logo logo
Prediction model study focusing on eHealth in the management of urinary incontinence: the Personalised Advantage Index as a decision-making aid. BMJ open OBJECTIVE:To develop a prediction model and illustrate the practical potential of personalisation of treatment decisions between app-based treatment and care as usual for urinary incontinence (UI). DESIGN:A prediction model study using data from a pragmatic, randomised controlled, non-inferiority trial. SETTING:Dutch primary care from 2015, with social media included from 2017. Enrolment ended on July 2018. PARTICIPANTS:Adult women were eligible if they had ≥2 episodes of UI per week, access to mobile apps and wanted treatment. Of the 350 screened women, 262 were eligible and randomised to app-based treatment or care as usual; 195 (74%) attended follow-up. PREDICTORS:Literature review and expert opinion identified 13 candidate predictors, categorised into two groups: Prognostic factors (independent of treatment type), such as UI severity, postmenopausal state, vaginal births, general physical health status, pelvic floor muscle function and body mass index; and modifiers (dependent on treatment type), such as age, UI type and duration, impact on quality of life, previous physical therapy, recruitment method and educational level. MAIN OUTCOME MEASURE:Primary outcome was symptom severity after a 4-month follow-up period, measured by the International Consultation on Incontinence Questionnaire the Urinary Incontinence Short Form. Prognostic factors and modifiers were combined into a final prediction model. For each participant, we then predicted treatment outcomes and calculated a Personalised Advantage Index (PAI). RESULTS:Baseline UI severity (prognostic) and age, educational level and impact on quality of life (modifiers) independently affected treatment effect of eHealth. The mean PAI was 0.99±0.79 points, being of clinical relevance in 21% of individuals. Applying the PAI also significantly improved treatment outcomes at the group level. CONCLUSIONS:Personalising treatment choice can support treatment decision making between eHealth and care as usual through the practical application of prediction modelling. Concerning eHealth for UI, this could facilitate the choice between app-based treatment and care as usual. TRIAL REGISTRATION NUMBER:NL4948t. 10.1136/bmjopen-2021-051827
A Personalized Management Approach of OHSS: Development of a Multiphase Prediction Model and Smartphone-Based App. Frontiers in endocrinology Objective:This study aimed to develop multiphase big-data-based prediction models of ovarian hyperstimulation syndrome (OHSS) and a smartphone app for risk calculation and patients' self-monitoring. Methods:Multiphase prediction models were developed from a retrospective cohort database of 21,566 women from January 2017 to December 2020 with controlled ovarian stimulation (COS). There were 17,445 women included in the final data analysis. Women were randomly assigned to either training cohort (n = 12,211) or validation cohort (n = 5,234). Their baseline clinical characteristics, COS-related characteristics, and embryo information were evaluated. The prediction models were divided into four phases: 1) prior to COS, 2) on the day of ovulation trigger, 3) after oocyte retrieval, and 4) prior to embryo transfer. The multiphase prediction models were built with stepwise regression and confirmed with LASSO regression. Internal validations were performed using the validation cohort and were assessed by discrimination and calibration, as well as clinical decision curves. A smartphone-based app "OHSS monitor" was constructed as part of the built-in app of the IVF-aid platform. The app had three modules, risk prediction module, symptom monitoring module, and treatment monitoring module. Results:The multiphase prediction models were developed with acceptable distinguishing ability to identify OHSS at-risk patients. The C-statistics of the first, second, third, and fourth phases in the training cohort were 0.628 (95% CI 0.598-0.658), 0.715 (95% CI 0.688-0.742), 0.792 (95% CI 0.770-0.815), and 0.814 (95% CI 0.793-0.834), respectively. The calibration plot showed the agreement of predictive and observed risks of OHSS, especially at the third- and fourth-phase prediction models in both training and validation cohorts. The net clinical benefits of the multiphase prediction models were also confirmed with a clinical decision curve. A smartphone-based app was constructed as a risk calculator based on the multiphase prediction models, and also as a self-monitoring tool for patients at risk. Conclusions:We have built multiphase prediction models based on big data and constructed a user-friendly smartphone-based app for the personalized management of women at risk of moderate/severe OHSS. The multiphase prediction models and user-friendly app can be readily used in clinical practice for clinical decision-support and self-management of patients. 10.3389/fendo.2022.911225
Development and validation of a machine-learning model for prediction of hypoxemia after extubation in intensive care units. Annals of translational medicine Background:Extubation is the process of removing tracheal tubes so that patients maintain oxygenation while they start to breathe spontaneously. However, hypoxemia after extubation is an important issue for critical care doctors and is associated with patients' oxygenation, circulation, recovery, and incidence of postoperative complications. Accuracy and specificity of most related conventional models remain unsatisfactory. We conducted a predictive analysis based on a supervised machine-learning algorithm for the precise prediction of hypoxemia after extubation in intensive care units (ICUs). Methods:Data were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database for patients over age 18 who underwent mechanical ventilation in the ICU. The primary outcome was hypoxemia after extubation, and it was defined as a partial pressure of oxygen <60 mmHg after extubation. Variables and individuals with missing values greater than 20% were excluded, and the remaining missing values were filled in using multiple imputation. The dataset was split into a training set (80%) and final test set (20%). All related clinical and laboratory variables were extracted, and logistics stepwise regression was performed to screen out the key features. Six different advanced machine-learning models, including logistics regression (LOG), random forest (RF), K-nearest neighbors (KNN), support-vector machine (SVM), eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM), were introduced for modelling. The best performance model in the first cross-validated dataset was further fine-tuned, and the final performance was assessed using the final test set. Results:A total of 14,777 patients were included in the study, and 1,864 of the patients' experienced hypoxemia after extubation. After training, the RF and LightGBM models were the strongest initial performers, and the area under the curve (AUC) using RF was 0.780 [95% confidence interval (CI), 0.755-0.805] and using LightGBM was 0.779 (95% CI, 0.752-0.806). The final AUC using RF was 0.792 (95% CI, 0.771-0.814) and using LightGBM was 0.792 (95% CI, 0.770-0.815). Conclusions:Our machine learning models have considerable potential for predicting hypoxemia after extubation, which help to reduce ICU morbidity and mortality. 10.21037/atm-22-2118
Derivation and validation of a machine learning risk score using biomarker and electronic patient data to predict progression of diabetic kidney disease. Diabetologia AIM:Predicting progression in diabetic kidney disease (DKD) is critical to improving outcomes. We sought to develop/validate a machine-learned, prognostic risk score (KidneyIntelX™) combining electronic health records (EHR) and biomarkers. METHODS:This is an observational cohort study of patients with prevalent DKD/banked plasma from two EHR-linked biobanks. A random forest model was trained, and performance (AUC, positive and negative predictive values [PPV/NPV], and net reclassification index [NRI]) was compared with that of a clinical model and Kidney Disease: Improving Global Outcomes (KDIGO) categories for predicting a composite outcome of eGFR decline of ≥5 ml/min per year, ≥40% sustained decline, or kidney failure within 5 years. RESULTS:In 1146 patients, the median age was 63 years, 51% were female, the baseline eGFR was 54 ml min [1.73 m], the urine albumin to creatinine ratio (uACR) was 6.9 mg/mmol, follow-up was 4.3 years and 21% had the composite endpoint. On cross-validation in derivation (n = 686), KidneyIntelX had an AUC of 0.77 (95% CI 0.74, 0.79). In validation (n = 460), the AUC was 0.77 (95% CI 0.76, 0.79). By comparison, the AUC for the clinical model was 0.62 (95% CI 0.61, 0.63) in derivation and 0.61 (95% CI 0.60, 0.63) in validation. Using derivation cut-offs, KidneyIntelX stratified 46%, 37% and 17% of the validation cohort into low-, intermediate- and high-risk groups for the composite kidney endpoint, respectively. The PPV for progressive decline in kidney function in the high-risk group was 61% for KidneyIntelX vs 40% for the highest risk strata by KDIGO categorisation (p < 0.001). Only 10% of those scored as low risk by KidneyIntelX experienced progression (i.e., NPV of 90%). The NRI for the high-risk group was 41% (p < 0.05). CONCLUSIONS:KidneyIntelX improved prediction of kidney outcomes over KDIGO and clinical models in individuals with early stages of DKD. 10.1007/s00125-021-05444-0
Development and validation of a machine-learning ALS survival model lacking vital capacity (VC-Free) for use in clinical trials during the COVID-19 pandemic. Amyotrophic lateral sclerosis & frontotemporal degeneration Vital capacity (VC) is routinely used for ALS clinical trial eligibility determinations, often to exclude patients unlikely to survive trial duration. However, spirometry has been limited by the COVID-19 pandemic. We developed a machine-learning survival model without the use of baseline VC and asked whether it could stratify clinical trial participants and a wider ALS clinic population. . A gradient boosting machine survival model lacking baseline VC (VC-Free) was trained using the PRO-ACT ALS database and compared to a multivariable model that included VC (VCI) and a univariable baseline %VC model (UNI). Discrimination, calibration-in-the-large and calibration slope were quantified. Models were validated using 10-fold internal cross validation, the VITALITY-ALS clinical trial placebo arm and data from the Emory University tertiary care clinic. Simulations were performed using each model to estimate survival of patients predicted to have a > 50% one year survival probability. . The VC-Free model suffered a minor performance decline compared to the VCI model yet retained strong discrimination for stratifying ALS patients. Both models outperformed the UNI model. The proportion of excluded vs. included patients who died through one year was on average 27% vs. 6% (VCI), 31% vs. 7% (VC-Free), and 13% vs. 10% (UNI). . The VC-Free model offers an alternative to the use of VC for eligibility determinations during the COVID-19 pandemic. The observation that the VC-Free model outperforms the use of VC in a broad ALS patient population suggests the use of prognostic strata in future, post-pandemic ALS clinical trial eligibility screening determinations. 10.1080/21678421.2021.1924207
Prognostic Assessment of COVID-19 in the Intensive Care Unit by Machine Learning Methods: Model Development and Validation. Pan Pan,Li Yichao,Xiao Yongjiu,Han Bingchao,Su Longxiang,Su Mingliang,Li Yansheng,Zhang Siqi,Jiang Dapeng,Chen Xia,Zhou Fuquan,Ma Ling,Bao Pengtao,Xie Lixin Journal of medical Internet research BACKGROUND:Patients with COVID-19 in the intensive care unit (ICU) have a high mortality rate, and methods to assess patients' prognosis early and administer precise treatment are of great significance. OBJECTIVE:The aim of this study was to use machine learning to construct a model for the analysis of risk factors and prediction of mortality among ICU patients with COVID-19. METHODS:In this study, 123 patients with COVID-19 in the ICU of Vulcan Hill Hospital were retrospectively selected from the database, and the data were randomly divided into a training data set (n=98) and test data set (n=25) with a 4:1 ratio. Significance tests, correlation analysis, and factor analysis were used to screen 100 potential risk factors individually. Conventional logistic regression methods and four machine learning algorithms were used to construct the risk prediction model for the prognosis of patients with COVID-19 in the ICU. The performance of these machine learning models was measured by the area under the receiver operating characteristic curve (AUC). Interpretation and evaluation of the risk prediction model were performed using calibration curves, SHapley Additive exPlanations (SHAP), Local Interpretable Model-Agnostic Explanations (LIME), etc, to ensure its stability and reliability. The outcome was based on the ICU deaths recorded from the database. RESULTS:Layer-by-layer screening of 100 potential risk factors finally revealed 8 important risk factors that were included in the risk prediction model: lymphocyte percentage, prothrombin time, lactate dehydrogenase, total bilirubin, eosinophil percentage, creatinine, neutrophil percentage, and albumin level. Finally, an eXtreme Gradient Boosting (XGBoost) model established with the 8 important risk factors showed the best recognition ability in the training set of 5-fold cross validation (AUC=0.86) and the verification queue (AUC=0.92). The calibration curve showed that the risk predicted by the model was in good agreement with the actual risk. In addition, using the SHAP and LIME algorithms, feature interpretation and sample prediction interpretation algorithms of the XGBoost black box model were implemented. Additionally, the model was translated into a web-based risk calculator that is freely available for public usage. CONCLUSIONS:The 8-factor XGBoost model predicts risk of death in ICU patients with COVID-19 well; it initially demonstrates stability and can be used effectively to predict COVID-19 prognosis in ICU patients. 10.2196/23128
Development and validation of three machine-learning models for predicting multiple organ failure in moderately severe and severe acute pancreatitis. Qiu Qiu,Nian Yong-Jian,Guo Yan,Tang Liang,Lu Nan,Wen Liang-Zhi,Wang Bin,Chen Dong-Feng,Liu Kai-Jun BMC gastroenterology BACKGROUND:Multiple organ failure (MOF) is a serious complication of moderately severe (MASP) and severe acute pancreatitis (SAP). This study aimed to develop and assess three machine-learning models to predict MOF. METHODS:Patients with MSAP and SAP who were admitted from July 2014 to June 2017 were included. Firstly, parameters with significant differences between patients with MOF and without MOF were screened out by univariate analysis. Then, support vector machine (SVM), logistic regression analysis (LRA) and artificial neural networks (ANN) models were constructed based on these factors, and five-fold cross-validation was used to train each model. RESULTS:A total of 263 patients were enrolled. Univariate analysis screened out sixteen parameters referring to blood volume, inflammatory, coagulation and renal function to construct machine-learning models. The predictive efficiency of the optimal combinations of features by SVM, LRA, and ANN was almost equal (AUC = 0.840, 0.832, and 0.834, respectively), as well as the Acute Physiology and Chronic Health Evaluation II score (AUC = 0.814, P > 0.05). The common important predictive factors were HCT, K-time, IL-6 and creatinine in three models. CONCLUSIONS:Three machine-learning models can be efficient prognostic tools for predicting MOF in MSAP and SAP. ANN is recommended, which only needs four common parameters. 10.1186/s12876-019-1016-y
Diabetes mellitus prediction and diagnosis from a data preprocessing and machine learning perspective. Computer methods and programs in biomedicine BACKGROUND AND OBJECTIVE:Diabetes mellitus is a metabolic disorder characterized by hyperglycemia, which results from the inadequacy of the body to secrete and respond to insulin. If not properly managed or diagnosed on time, diabetes can pose a risk to vital body organs such as the eyes, kidneys, nerves, heart, and blood vessels and so can be life-threatening. The many years of research in computational diagnosis of diabetes have pointed to machine learning to as a viable solution for the prediction of diabetes. However, the accuracy rate to date suggests that there is still much room for improvement. In this paper, we are proposing a machine learning framework for diabetes prediction and diagnosis using the PIMA Indian dataset and the laboratory of the Medical City Hospital (LMCH) diabetes dataset. We hypothesize that adopting feature selection and missing value imputation methods can scale up the performance of classification models in diabetes prediction and diagnosis. METHODS:In this paper, a robust framework for building a diabetes prediction model to aid in the clinical diagnosis of diabetes is proposed. The framework includes the adoption of Spearman correlation and polynomial regression for feature selection and missing value imputation, respectively, from a perspective that strengthens their performances. Further, different supervised machine learning models, the random forest (RF) model, support vector machine (SVM) model, and our designed twice-growth deep neural network (2GDNN) model are proposed for classification. The models are optimized by tuning the hyperparameters of the models using grid search and repeated stratified k-fold cross-validation and evaluated for their ability to scale to the prediction problem. RESULTS:Through experiments on the PIMA Indian and LMCH diabetes datasets, precision, sensitivity, F1-score, train-accuracy, and test-accuracy scores of 97.34%, 97.24%, 97.26%, 99.01%, 97.25 and 97.28%, 97.33%, 97.27%, 99.57%, 97.33, are achieved with the proposed 2GDNN model, respectively. CONCLUSION:The data preprocessing approaches and the classifiers with hyperparameter optimization proposed within the machine learning framework yield a robust machine learning model that outperforms state-of-the-art results in diabetes mellitus prediction and diagnosis. The source code for the models of the proposed machine learning framework has been made publicly available. 10.1016/j.cmpb.2022.106773
Development, Validation, and Evaluation of a Simple Machine Learning Model to Predict Cirrhosis Mortality. Kanwal Fasiha,Taylor Thomas J,Kramer Jennifer R,Cao Yumei,Smith Donna,Gifford Allen L,El-Serag Hashem B,Naik Aanand D,Asch Steven M JAMA network open Importance:Machine-learning algorithms offer better predictive accuracy than traditional prognostic models but are too complex and opaque for clinical use. Objective:To compare different machine learning methods in predicting overall mortality in cirrhosis and to use machine learning to select easily scored clinical variables for a novel cirrhosis prognostic model. Design, Setting, and Participants:This prognostic study used a retrospective cohort of adult patients with cirrhosis or its complications seen in 130 hospitals and affiliated ambulatory clinics in the integrated, national Veterans Affairs health care system from October 1, 2011, to September 30, 2015. Patients were followed up through December 31, 2018. Data were analyzed from October 1, 2017, to May 31, 2020. Exposures:Potential predictors included demographic characteristics; liver disease etiology, severity, and complications; use of health care resources; comorbid conditions; and comprehensive laboratory and medication data. Patients were randomly selected for model development (66.7%) and validation (33.3%). Three different statistical and machine learning methods were evaluated: gradient descent boosting, logistic regression with least absolute shrinkage and selection operator (LASSO) regularization, and logistic regression with LASSO constrained to select no more than 10 predictors (partial pathway model). Predictor inclusion and model performance were evaluated in a 5-fold cross-validation. Last, the predictors identified in the most parsimonious (the partial path) model were refit using maximum-likelihood estimation (Cirrhosis Mortality Model [CiMM]), and its predictive performance was compared with that of the widely used Model for End Stage Liver Disease with sodium (MELD-Na) score. Main Outcomes and Measures:All-cause mortality. Results:Of the 107 939 patients with cirrhosis (mean [SD] age, 62.7 [9.6] years; 96.6% male; 66.3% white, 18.4% African American), the annual mortality rate ranged from 8.8% to 15.3%. In total, 32.7% of patients died within 3 years, and 46.2% died within 5 years after the index date. Models predicting 1-year mortality had good discrimination for the gradient descent boosting (area under the receiver operating characteristics curve [AUC], 0.81; 95% CI, 0.80-0.82), logistic regression with LASSO regularization (AUC, 0.78; 95% CI, 0.77-0.79), and the partial path logistic model (AUC, 0.78; 95% CI, 0.76-0.78). All models showed good calibration. The final CiMM model with machine learning-derived clinical variables offered significantly better discrimination than the MELD-Na score, with AUCs of 0.78 (95% CI, 0.77-0.79) vs 0.67 (95% CI, 0.66-0.68) for 1-year mortality, respectively (DeLong z = 17.00; P < .001). Conclusions and Relevance:In this study, simple machine learning techniques performed as well as the more advanced ensemble gradient boosting. Using the clinical variables identified from simple machine learning in a cirrhosis mortality model produced a new score more transparent than machine learning and more predictive than the MELD-Na score. 10.1001/jamanetworkopen.2020.23780
A comparison of machine learning methods for predicting recurrence and death after curative-intent radiotherapy for non-small cell lung cancer: Development and validation of multivariable clinical prediction models. Hindocha Sumeet,Charlton Thomas G,Linton-Reid Kristofer,Hunter Benjamin,Chan Charleen,Ahmed Merina,Robinson Emily J,Orton Matthew,Ahmad Shahreen,McDonald Fiona,Locke Imogen,Power Danielle,Blackledge Matthew,Lee Richard W,Aboagye Eric O EBioMedicine BACKGROUND:Surveillance is universally recommended for non-small cell lung cancer (NSCLC) patients treated with curative-intent radiotherapy. High-quality evidence to inform optimal surveillance strategies is lacking. Machine learning demonstrates promise in accurate outcome prediction for a variety of health conditions. The purpose of this study was to utilise readily available patient, tumour, and treatment data to develop, validate and externally test machine learning models for predicting recurrence, recurrence-free survival (RFS) and overall survival (OS) at 2 years from treatment. METHODS:A retrospective, multicentre study of patients receiving curative-intent radiotherapy for NSCLC was undertaken. A total of 657 patients from 5 hospitals were eligible for inclusion. Data pre-processing derived 34 features for predictive modelling. Combinations of 8 feature reduction methods and 10 machine learning classification algorithms were compared, producing risk-stratification models for predicting recurrence, RFS and OS. Models were compared with 10-fold cross validation and an external test set and benchmarked against TNM-stage and performance status. Youden Index was derived from validation set ROC curves to distinguish high and low risk groups and Kaplan-Meier analyses performed. FINDINGS:Median follow-up time was 852 days. Parameters were well matched across training-validation and external test sets: Mean age was 73 and 71 respectively, and recurrence, RFS and OS rates at 2 years were 43% vs 34%, 54% vs 47% and 54% vs 47% respectively. The respective validation and test set AUCs were as follows: 1) RFS: 0·682 (0·575-0·788) and 0·681 (0·597-0·766), 2) Recurrence: 0·687 (0·582-0·793) and 0·722 (0·635-0·81), and 3) OS: 0·759 (0·663-0·855) and 0·717 (0·634-0·8). Our models were superior to TNM stage and performance status in predicting recurrence and OS. INTERPRETATION:This robust and ready to use machine learning method, validated and externally tested, sets the stage for future clinical trials entailing quantitative personalised risk-stratification and surveillance following curative-intent radiotherapy for NSCLC. FUNDING:A full list of funding bodies that contributed to this study can be found in the Acknowledgements section. 10.1016/j.ebiom.2022.103911
Development and External Validation of a Machine Learning Tool to Rule Out COVID-19 Among Adults in the Emergency Department Using Routine Blood Tests: A Large, Multicenter, Real-World Study. Plante Timothy B,Blau Aaron M,Berg Adrian N,Weinberg Aaron S,Jun Ik C,Tapson Victor F,Kanigan Tanya S,Adib Artur B Journal of medical Internet research BACKGROUND:Conventional diagnosis of COVID-19 with reverse transcription polymerase chain reaction (RT-PCR) testing (hereafter, PCR) is associated with prolonged time to diagnosis and significant costs to run the test. The SARS-CoV-2 virus might lead to characteristic patterns in the results of widely available, routine blood tests that could be identified with machine learning methodologies. Machine learning modalities integrating findings from these common laboratory test results might accelerate ruling out COVID-19 in emergency department patients. OBJECTIVE:We sought to develop (ie, train and internally validate with cross-validation techniques) and externally validate a machine learning model to rule out COVID 19 using only routine blood tests among adults in emergency departments. METHODS:Using clinical data from emergency departments (EDs) from 66 US hospitals before the pandemic (before the end of December 2019) or during the pandemic (March-July 2020), we included patients aged ≥20 years in the study time frame. We excluded those with missing laboratory results. Model training used 2183 PCR-confirmed cases from 43 hospitals during the pandemic; negative controls were 10,000 prepandemic patients from the same hospitals. External validation used 23 hospitals with 1020 PCR-confirmed cases and 171,734 prepandemic negative controls. The main outcome was COVID 19 status predicted using same-day routine laboratory results. Model performance was assessed with area under the receiver operating characteristic (AUROC) curve as well as sensitivity, specificity, and negative predictive value (NPV). RESULTS:Of 192,779 patients included in the training, external validation, and sensitivity data sets (median age decile 50 [IQR 30-60] years, 40.5% male [78,249/192,779]), AUROC for training and external validation was 0.91 (95% CI 0.90-0.92). Using a risk score cutoff of 1.0 (out of 100) in the external validation data set, the model achieved sensitivity of 95.9% and specificity of 41.7%; with a cutoff of 2.0, sensitivity was 92.6% and specificity was 59.9%. At the cutoff of 2.0, the NPVs at a prevalence of 1%, 10%, and 20% were 99.9%, 98.6%, and 97%, respectively. CONCLUSIONS:A machine learning model developed with multicenter clinical data integrating commonly collected ED laboratory data demonstrated high rule-out accuracy for COVID-19 status, and might inform selective use of PCR-based testing. 10.2196/24048
Machine-learning prediction of adolescent alcohol use: a cross-study, cross-cultural validation. Afzali Mohammad H,Sunderland Matthew,Stewart Sherry,Masse Benoit,Seguin Jean,Newton Nicola,Teesson Maree,Conrod Patricia Addiction (Abingdon, England) BACKGROUND AND AIMS:The experience of alcohol use among adolescents is complex, with international differences in age of purchase and individual differences in consumption and consequences. This latter underlines the importance of prediction modeling of adolescent alcohol use. The current study (a) compared the performance of seven machine-learning algorithms to predict different levels of alcohol use in mid-adolescence and (b) used a cross-cultural cross-study scheme in the training-validation-test process to display the predictive power of the best performing machine-learning algorithm. DESIGN:A comparison of seven machine-learning algorithms: logistic regression, support vector machines, random forest, neural network, lasso regression, ridge regression and elastic-net. SETTING:Canada and Australia. PARTICIPANTS:The Canadian sample is part of a 4-year follow-up (2012-16) of the Co-Venture cohort (n = 3826, baseline age 12.8 ± 0.4, 49.2% girls). The Australian sample is part of a 3-year follow-up (2012-15) of the Climate Schools and Preventure (CAP) cohort (n = 2190, baseline age 13.3 ± 0.3, 43.7% girls). MEASUREMENTS:The algorithms used several prediction indices, such as F prediction score, accuracy, precision, recall, negative predictive value and area under the curve (AUC). FINDINGS:Based on prediction indices, the elastic-net machine-learning algorithm showed the best predictive performance in both Canadian (AUC = 0.869 ± 0.066) and Australian (AUC = 0.855 ± 0.072) samples. Domain contribution analysis showed that the highest prediction accuracy indices yielded from models with only psychopathology (AUC = 0.816 ± 0.044/0.790 ± 0.071 in Canada/Australia) and only personality clusters (AUC = 0.776 ± 0.063/0.796 ± 0.066 in Canada/Australia). Similarly, regardless of the level of alcohol use, in both samples, externalizing psychopathologies, alcohol use at baseline and the sensation-seeking personality profile contributed to the prediction. CONCLUSIONS:Computerized screening software shows promise in predicting the risk of alcohol use among adolescents. 10.1111/add.14504
Development and Validation of an Explainable Machine Learning Model for Major Complications After Cytoreductive Surgery. JAMA network open Importance:Cytoreductive surgery (CRS) is one of the most complex operations in surgical oncology with significant morbidity, and improved risk prediction tools are critically needed. Machine learning models can potentially overcome the limitations of traditional multiple logistic regression (MLR) models and provide accurate risk estimates. Objective:To develop and validate an explainable machine learning model for predicting major postoperative complications in patients undergoing CRS. Design, Setting, and Participants:This prognostic study used patient data from tertiary care hospitals with expertise in CRS included in the US Hyperthermic Intraperitoneal Chemotherapy Collaborative Database between 1998 and 2018. Information from 147 variables was extracted to predict the risk of a major complication. An ensemble-based machine learning (gradient-boosting) model was optimized on 80% of the sample with subsequent validation on a 20% holdout data set. The machine learning model was compared with traditional MLR models. The artificial intelligence SHAP (Shapley additive explanations) method was used for interpretation of patient- and cohort-level risk estimates and interactions to define novel surgical risk phenotypes. Data were analyzed between November 2019 and August 2021. Exposures:Cytoreductive surgery. Main Outcomes and Measures:Area under the receiver operating characteristics (AUROC); area under the precision recall curve (AUPRC). Results:Data from a total 2372 patients were included in model development (mean age, 55 years [range, 11-95 years]; 1366 [57.6%] women). The optimized machine learning model achieved high discrimination (AUROC: mean cross-validation, 0.75 [range, 0.73-0.81]; test, 0.74) and precision (AUPRC: mean cross-validation, 0.50 [range, 0.46-0.58]; test, 0.42). Compared with the optimized machine learning model, the published MLR model performed worse (test AUROC and AUPRC: 0.54 and 0.18, respectively). Higher volume of estimated blood loss, having pelvic peritonectomy, and longer operative time were the top 3 contributors to the high likelihood of major complications. SHAP dependence plots demonstrated insightful nonlinear interactive associations between predictors and major complications. For instance, high estimated blood loss (ie, above 500 mL) was only detrimental when operative time exceeded 9 hours. Unsupervised clustering of patients based on similarity of sources of risk allowed identification of 6 distinct surgical risk phenotypes. Conclusions and Relevance:In this prognostic study using data from patients undergoing CRS, an optimized machine learning model demonstrated a superior ability to predict individual- and cohort-level risk of major complications vs traditional methods. Using the SHAP method, 6 distinct surgical phenotypes were identified based on sources of risk of major complications. 10.1001/jamanetworkopen.2022.12930
A Prediction Model for Cognitive Impairment Risk in Colorectal Cancer after Chemotherapy Treatment. Zhou Shu-Ping,Fei Su-Ding,Han Hui-Hui,Li Jing-Jing,Yang Shuang,Zhao Chun-Yang BioMed research international Background:A prediction model can be developed to predict the risk of cancer-related cognitive impairment in colorectal cancer patients after chemotherapy. Methods:A regression analysis was performed on 386 colorectal cancer patients who had undergone chemotherapy. Three prediction models (random forest, logistic regression, and support vector machine models) were constructed using collected clinical and pathological data of the patients. Calibration and ROC curves and -indexes were used to evaluate the selected models. A decision curve analysis (DCA) was used to determine the clinical utility of the line graph. Results:Three prediction models including a random forest, a logistic regression, and a support vector machine were constructed. The logistic regression model had the strongest predictive power with an area under the curve (AUC) of 0.799. Age, BMI, colostomy, complications, CRA, depression, diabetes, QLQ-C30 score, exercise, hypercholesterolemia, diet, marital status, education level, and pathological stage were included in the nomogram. The -index (0.826) and calibration curve showed that the nomogram had good predictive ability and the DCA curves indicated that the model had strong clinical utility. Conclusions:A prediction model with good predictive ability and practical clinical value can be developed for predicting the risk of cognitive impairment in colorectal cancer after chemotherapy. 10.1155/2021/6666453
A Risk Prediction Model Based on Machine Learning for Cognitive Impairment Among Chinese Community-Dwelling Elderly People With Normal Cognition: Development and Validation Study. Journal of medical Internet research BACKGROUND:Identifying cognitive impairment early enough could support timely intervention that may hinder or delay the trajectory of cognitive impairment, thus increasing the chances for successful cognitive aging. OBJECTIVE:We aimed to build a prediction model based on machine learning for cognitive impairment among Chinese community-dwelling elderly people with normal cognition. METHODS:A prospective cohort of 6718 older people from the Chinese Longitudinal Healthy Longevity Survey (CLHLS) register, followed between 2008 and 2011, was used to develop and validate the prediction model. Participants were included if they were aged 60 years or above, were community-dwelling elderly people, and had a cognitive Mini-Mental State Examination (MMSE) score ≥18. They were excluded if they were diagnosed with a severe disease (eg, cancer and dementia) or were living in institutions. Cognitive impairment was identified using the Chinese version of the MMSE. Several machine learning algorithms (random forest, XGBoost, naïve Bayes, and logistic regression) were used to assess the 3-year risk of developing cognitive impairment. Optimal cutoffs and adjusted parameters were explored in validation data, and the model was further evaluated in test data. A nomogram was established to vividly present the prediction model. RESULTS:The mean age of the participants was 80.4 years (SD 10.3 years), and 50.85% (3416/6718) were female. During a 3-year follow-up, 991 (14.8%) participants were identified with cognitive impairment. Among 45 features, the following four features were finally selected to develop the model: age, instrumental activities of daily living, marital status, and baseline cognitive function. The concordance index of the model constructed by logistic regression was 0.814 (95% CI 0.781-0.846). Older people with normal cognitive functioning having a nomogram score of less than 170 were considered to have a low 3-year risk of cognitive impairment, and those with a score of 170 or greater were considered to have a high 3-year risk of cognitive impairment. CONCLUSIONS:This simple and feasible cognitive impairment prediction model could identify community-dwelling elderly people at the greatest 3-year risk for cognitive impairment, which could help community nurses in the early identification of dementia. 10.2196/20298
Machine learning-based models for predicting clinical outcomes after surgery in unilateral primary aldosteronism. Scientific reports Unilateral subtype of primary aldosteronism (PA) is a common surgically curable form of endocrine hypertension. However, more than half of the patients with PA who undergo unilateral adrenalectomy suffer from persistent hypertension, which may discourage those with PA from undergoing adrenalectomy even when appropriate. The aim of this retrospective cross-sectional study was to develop machine learning-based models for predicting postoperative hypertensive remission using preoperative predictors that are readily available in routine clinical practice. A total of 107 patients with PA who achieved complete biochemical success after adrenalectomy were included and randomly assigned to the training and test datasets. Predictive models of complete clinical success were developed using supervised machine learning algorithms. Of 107 patients, 40 achieved complete clinical success after adrenalectomy in both datasets. Six clinical features associated with complete clinical success (duration of hypertension, defined daily dose (DDD) of antihypertensive medication, plasma aldosterone concentration (PAC), sex, body mass index (BMI), and age) were selected based on predictive performance in the machine learning-based model. The predictive accuracy and area under the curve (AUC) for the developed model in the test dataset were 77.3% and 0.884 (95% confidence interval: 0.737-1.000), respectively. In an independent external cohort, the performance of the predictive model was found to be comparable with an accuracy of 80.4% and AUC of 0.867 (95% confidence interval: 0.763-0.971). The duration of hypertension, DDD of antihypertensive medication, PAC, and BMI were non-linearly related to the prediction of complete clinical success. The developed predictive model may be useful in assessing the benefit of unilateral adrenalectomy and in selecting surgical treatment and antihypertensive medication for patients with PA in clinical practice. 10.1038/s41598-022-09706-8
Identification of exacerbation risk in patients with liver dysfunction using machine learning algorithms. Peng Junfeng,Zhou Mi,Chen Chuan,Xie Xiaohua,Luo Ching-Hsing PloS one The prediction of the liver failure (LF) and its proper diagnosis would lead to a reduction in the complications of the disease and prevents the progress of the disease. To improve the treatment of LF patients and reduce the cost of treatment, we build a machine learning model to forecast whether a patient would deteriorate after admission to the hospital. First, a total of 348 LF patients were included from May 2011 to March 2018 retrospectively in this study. Then, 15 key clinical indicators are selected as the input of the machine learning algorithm. Finally, machine learning and the Model for End-Stage Liver Disease (MELD) are used to forecast the LF deterioration. The area under the receiver operating characteristic (AUC) of MELD, GLMs, CART, SVM and NNET with 10 fold-cross validation was 0.670, 0.554, 0.794, 0.853 and 0.912 respectively. Additionally, the accuracy of MELD, GLMs, CART, SVM and NNET was 0.669, 0.456, 0.794, 0.853 and 0.912. The predictive performance of the developed machine model execept the GLMs exceeds the classic MELD model. The machine learning method could support the physicians to trigger the initiation of timely treatment for the LD patients. 10.1371/journal.pone.0239266
Cardiovascular risk prediction: from classical statistical methods to machine learning approaches. Sperti Michela,Malavolta Marta,Staunovo Polacco Federica,Dellavalle Annalisa,Ruggieri Rossella,Bergia Sara,Fazio Alice,Santoro Carmine,Deriu Marco A Minerva cardiology and angiology Nowadays, cardiovascular risk prediction scores are commonly used in primary prevention settings. Estimating the cardiovascular individual risk is of crucial importance for effective patient management and optimal therapy identification, with relevant consequences on secondary prevention settings. To reach this goal, a plethora of risk scores have been developed in the past, most of them assuming that each cardiovascular risk factor is linearly dependent on the outcome. However, the overall accuracy of these methods often remains insufficient to solve the problem at hand. In this scenario, machine learning techniques have repeatedly proved successful in improving cardiovascular risk predictions, being able to capture the non-linearity present in the data. In this concern, we present a detailed discussion concerning the application of classical versus machine learning-based cardiovascular risk scores in the clinical setting. This review aimed to give an overview of the current risk scores based on classical statistical approaches and machine learning techniques applied to predict the risk of several cardiovascular diseases, comparing them, discussing their similarities and differences, and highlighting their main drawbacks to aid the physician having a more critical understanding of these tools. 10.23736/S2724-5683.21.05868-3
Use of Machine Learning Approaches in Clinical Epidemiological Research of Diabetes. Basu Sanjay,Johnson Karl T,Berkowitz Seth A Current diabetes reports PURPOSE OF REVIEW:Machine learning approaches-which seek to predict outcomes or classify patient features by recognizing patterns in large datasets-are increasingly applied to clinical epidemiology research on diabetes. Given its novelty and emergence in fields outside of biomedical research, machine learning terminology, techniques, and research findings may be unfamiliar to diabetes researchers. Our aim was to present the use of machine learning approaches in an approachable way, drawing from clinical epidemiological research in diabetes published from 1 Jan 2017 to 1 June 2020. RECENT FINDINGS:Machine learning approaches using tree-based learners-which produce decision trees to help guide clinical interventions-frequently have higher sensitivity and specificity than traditional regression models for risk prediction. Machine learning approaches using neural networking and "deep learning" can be applied to medical image data, particularly for the identification and staging of diabetic retinopathy and skin ulcers. Among the machine learning approaches reviewed, researchers identified new strategies to develop standard datasets for rigorous comparisons across older and newer approaches, methods to illustrate how a machine learner was treating underlying data, and approaches to improve the transparency of the machine learning process. Machine learning approaches have the potential to improve risk stratification and outcome prediction for clinical epidemiology applications. Achieving this potential would be facilitated by use of universal open-source datasets for fair comparisons. More work remains in the application of strategies to communicate how the machine learners are generating their predictions. 10.1007/s11892-020-01353-5
Machine-Learning Provides Patient-Specific Prediction of Metastatic Risk Based on Innovative, Mechanobiology Assay. Rozen Rakefet,Weihs Daphne Annals of biomedical engineering Cancer mortality is mostly related to metastasis. Metastasis is currently prognosed via histopathology, disease-statistics, or genetics; those are potentially inaccurate, not rapidly available and require known markers. We had developed a rapid (~ 2 h) mechanobiology-based approach to provide early prognosis of the clinical likelihood for metastasis. Specifically, invasive cell-subsets seeded on impenetrable, physiological-stiffness polyacrylamide gels forcefully indent the gels, while non-invasive/benign cells do not. The number of indenting cells and their attained depths, the mechanical invasiveness, accurately define the metastatic risk of tumors and cell-lines. Utilizing our experimental database, we compare the capacity of several machine learning models to predict the metastatic risk. Models underwent supervised training on individual experiments using classification from literature and commercial-sources for established cell-lines and clinical histopathology reports for tumor samples. We evaluated 2-class models, separating invasive/non-invasive (e.g. benign) samples, and obtained sensitivity and specificity of 0.92 and 1, respectively; this surpasses other works. We also introduce a novel approach, using 5-class models (i.e. normal, benign, cancer-metastatic-non/low/high) that provided average sensitivity and specificity of 0.69 and 0.91. Combining our rapid, mechanical invasiveness assay with machine learning classification can provide accurate and early prognosis of metastatic risk, to support choice of treatments and disease management. 10.1007/s10439-020-02720-9
Machine learning methods are comparable to logistic regression techniques in predicting severe walking limitation following total knee arthroplasty. Pua Yong-Hao,Kang Hakmook,Thumboo Julian,Clark Ross Allan,Chew Eleanor Shu-Xian,Poon Cheryl Lian-Li,Chong Hwei-Chi,Yeo Seng-Jin Knee surgery, sports traumatology, arthroscopy : official journal of the ESSKA PURPOSE:Machine-learning methods are flexible prediction algorithms with potential advantages over conventional regression. This study aimed to use machine learning methods to predict post-total knee arthroplasty (TKA) walking limitation, and to compare their performance with that of logistic regression. METHODS:From the department's clinical registry, a cohort of 4026 patients who underwent elective, primary TKA between July 2013 and July 2017 was identified. Candidate predictors included demographics and preoperative clinical, psychosocial, and outcome measures. The primary outcome was severe walking limitation at 6 months post-TKA, defined as a maximum walk time ≤ 15 min. Eight common regression (logistic, penalized logistic, and ordinal logistic with natural splines) and ensemble machine learning (random forest, extreme gradient boosting, and SuperLearner) methods were implemented to predict the probability of severe walking limitation. Models were compared on discrimination and calibration metrics. RESULTS:At 6 months post-TKA, 13% of patients had severe walking limitation. Machine learning and logistic regression models performed moderately [mean area under the ROC curves (AUC) 0.73-0.75]. Overall, the ordinal logistic regression model performed best while the SuperLearner performed best among machine learning methods, with negligible differences between them (Brier score difference, < 0.001; 95% CI [- 0.0025, 0.002]). CONCLUSIONS:When predicting post-TKA physical function, several machine learning methods did not outperform logistic regression-in particular, ordinal logistic regression that does not assume linearity in its predictors. LEVEL OF EVIDENCE:Prognostic level II. 10.1007/s00167-019-05822-7
Machine learning in oral squamous cell carcinoma: Current status, clinical concerns and prospects for future-A systematic review. Alabi Rasheed Omobolaji,Youssef Omar,Pirinen Matti,Elmusrati Mohammed,Mäkitie Antti A,Leivo Ilmo,Almangush Alhadi Artificial intelligence in medicine BACKGROUND:Oral cancer can show heterogenous patterns of behavior. For proper and effective management of oral cancer, early diagnosis and accurate prediction of prognosis are important. To achieve this, artificial intelligence (AI) or its subfield, machine learning, has been touted for its potential to revolutionize cancer management through improved diagnostic precision and prediction of outcomes. Yet, to date, it has made only few contributions to actual medical practice or patient care. OBJECTIVES:This study provides a systematic review of diagnostic and prognostic application of machine learning in oral squamous cell carcinoma (OSCC) and also highlights some of the limitations and concerns of clinicians towards the implementation of machine learning-based models for daily clinical practice. DATA SOURCES:We searched OvidMedline, PubMed, Scopus, Web of Science, and Institute of Electrical and Electronics Engineers (IEEE) databases from inception until February 2020 for articles that used machine learning for diagnostic or prognostic purposes of OSCC. ELIGIBILITY CRITERIA:Only original studies that examined the application of machine learning models for prognostic and/or diagnostic purposes were considered. DATA EXTRACTION:Independent extraction of articles was done by two researchers (A.R. & O.Y) using predefine study selection criteria. We used the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) in the searching and screening processes. We also used Prediction model Risk of Bias Assessment Tool (PROBAST) for assessing the risk of bias (ROB) and quality of included studies. RESULTS:A total of 41 studies were published to have used machine learning to aid in the diagnosis/or prognosis of OSCC. The majority of these studies used the support vector machine (SVM) and artificial neural network (ANN) algorithms as machine learning techniques. Their specificity ranged from 0.57 to 1.00, sensitivity from 0.70 to 1.00, and accuracy from 63.4 % to 100.0 % in these studies. The main limitations and concerns can be grouped as either the challenges inherent to the science of machine learning or relating to the clinical implementations. CONCLUSION:Machine learning models have been reported to show promising performances for diagnostic and prognostic analyses in studies of oral cancer. These models should be developed to further enhance explainability, interpretability, and externally validated for generalizability in order to be safely integrated into daily clinical practices. Also, regulatory frameworks for the adoption of these models in clinical practices are necessary. 10.1016/j.artmed.2021.102060
Cytokine gene variants and socio-demographic characteristics as predictors of cervical cancer: A machine learning approach. Kaushik Manoj,Chandra Joshi Rakesh,Kushwah Atar Singh,Gupta Maneesh Kumar,Banerjee Monisha,Burget Radim,Dutta Malay Kishore Computers in biology and medicine Cervical cancer is still one of the most prevalent cancers in women and a significant cause of mortality. Cytokine gene variants and socio-demographic characteristics have been reported as biomarkers for determining the cervical cancer risk in the Indian population. This study was designed to apply a machine learning-based model using these risk factors for better prognosis and prediction of cervical cancer. This study includes the dataset of cytokine gene variants, clinical and socio-demographic characteristics of normal healthy control subjects, and cervical cancer cases. Different risk factors, including demographic details and cytokine gene variants, were analysed using different machine learning approaches. Various statistical parameters were used for evaluating the proposed method. After multi-step data processing and random splitting of the dataset, machine learning methods were applied and evaluated with 5-fold cross-validation and also tested on the unseen data records of a collected dataset for proper evaluation and analysis. The proposed approaches were verified after analysing various performance metrics. The logistic regression technique achieved the highest average accuracy of 82.25% and the highest average F1-score of 82.58% among all the methods. Ridge classifiers and the Gaussian Naïve Bayes classifier achieved the highest sensitivity-85%. The ridge classifier surpasses most of the machine learning classifiers with 84.78% accuracy and 97.83% sensitivity. The risk factors analysed in this study can be taken as biomarkers in developing a cervical cancer diagnosis system. The outcomes demonstrate that the machine learning assisted analysis of cytokine gene variants and socio-demographic characteristics can be utilised effectively for predicting the risk of developing cervical cancer. 10.1016/j.compbiomed.2021.104559
Investigating Risk Factors and Predicting Complications in Deep Brain Stimulation Surgery with Machine Learning Algorithms. Farrokhi Farrokh,Buchlak Quinlan D,Sikora Matt,Esmaili Nazanin,Marsans Maria,McLeod Pamela,Mark Jamie,Cox Emily,Bennett Christine,Carlson Jonathan World neurosurgery BACKGROUND:Deep brain stimulation (DBS) surgery is an option for patients experiencing medically resistant neurologic symptoms. DBS complications are rare; finding significant predictors requires a large number of surgeries. Machine learning algorithms may be used to effectively predict these outcomes. The aims of this study were to 1) investigate preoperative clinical risk factors and 2) build machine learning models to predict adverse outcomes. METHODS:This multicenter registry collected clinical and demographic characteristics of patients undergoing DBS surgery (n = 501) and tabulated occurrence of complications. Logistic regression was used to evaluate risk factors. Supervised learning algorithms were trained and validated on 70% and 30%, respectively, of both oversampled and original registry data. Performance was evaluated using area under the receiver operating characteristics curve (AUC), sensitivity, specificity, and accuracy. RESULTS:Logistic regression showed that the risk of complication was related to the operating institution in which the surgery was performed (odds ratio [OR] = 0.44, confidence interval [CI] = 0.25-0.78), body mass index (OR = 0.94, CI = 0.89-0.99), and diabetes (OR = 2.33, CI = 1.18-4.60). Patients with diabetes were almost 3× more likely to return to the operating room (OR = 2.78, CI = 1.31-5.88). Patients with a history of smoking were 4× more likely to experience postoperative infection (OR = 4.20, CI = 1.21-14.61). Supervised learning algorithms demonstrated high discrimination performance when predicting any complication (AUC = 0.86), a complication within 12 months (AUC = 0.91), return to the operating room (AUC = 0.88), and infection (AUC = 0.97). Age, body mass index, procedure side, gender, and a diagnosis of Parkinson disease were influential features. CONCLUSIONS:Multiple significant complication risk factors were identified, and supervised learning algorithms effectively predicted adverse outcomes in DBS surgery. 10.1016/j.wneu.2019.10.063
Machine Learning Models in Type 2 Diabetes Risk Prediction: Results from a Cross-sectional Retrospective Study in Chinese Adults. Xiong Xiao-Lu,Zhang Rong-Xin,Bi Yan,Zhou Wei-Hong,Yu Yun,Zhu Da-Long Current medical science Type 2 diabetes mellitus (T2DM) has become a prevalent health problem in China, especially in urban areas. Early prevention strategies are needed to reduce the associated mortality and morbidity. We applied the combination of rules and different machine learning techniques to assess the risk of development of T2DM in an urban Chinese adult population. A retrospective analysis was performed on 8000 people with non-diabetes and 3845 people with T2DM in Nanjing. Multilayer Perceptron (MLP), AdaBoost (AD), Trees Random Forest (TRF), Support Vector Machine (SVM), and Gradient Tree Boosting (GTB) machine learning techniques with 10 cross validation methods were used with the proposed model for the prediction of the risk of development of T2DM. The performance of these models was evaluated with accuracy, precision, sensitivity, specificity, and area under receiver operating characteristic (ROC) curve (AUC). After comparison, the prediction accuracy of the different five machine models was 0.87, 0.86, 0.86, 0.86 and 0.86 respectively. The combination model using the same voting weight of each component was built on T2DM, which was performed better than individual models. The findings indicate that, combining machine learning models could provide an accurate assessment model for T2DM risk prediction. 10.1007/s11596-019-2077-4
Predicting the Risk of Hypertension Based on Several Easy-to-Collect Risk Factors: A Machine Learning Method. Zhao Huanhuan,Zhang Xiaoyu,Xu Yang,Gao Lisheng,Ma Zuchang,Sun Yining,Wang Weimin Frontiers in public health Hypertension is a widespread chronic disease. Risk prediction of hypertension is an intervention that contributes to the early prevention and management of hypertension. The implementation of such intervention requires an effective and easy-to-implement hypertension risk prediction model. This study evaluated and compared the performance of four machine learning algorithms on predicting the risk of hypertension based on easy-to-collect risk factors. A dataset of 29,700 samples collected through a physical examination was used for model training and testing. Firstly, we identified easy-to-collect risk factors of hypertension, through univariate logistic regression analysis. Then, based on the selected features, 10-fold cross-validation was utilized to optimize four models, random forest (RF), CatBoost, MLP neural network and logistic regression (LR), to find the best hyper-parameters on the training set. Finally, the performance of models was evaluated by AUC, accuracy, sensitivity and specificity on the test set. The experimental results showed that the RF model outperformed the other three models, and achieved an AUC of 0.92, an accuracy of 0.82, a sensitivity of 0.83 and a specificity of 0.81. In addition, Body Mass Index (BMI), age, family history and waist circumference (WC) are the four primary risk factors of hypertension. These findings reveal that it is feasible to use machine learning algorithms, especially RF, to predict hypertension risk without clinical or genetic data. The technique can provide a non-invasive and economical way for the prevention and management of hypertension in a large population. 10.3389/fpubh.2021.619429
Machine learning for characterizing risk of type 2 diabetes mellitus in a rural Chinese population: the Henan Rural Cohort Study. Scientific reports With the development of data mining, machine learning offers opportunities to improve discrimination by analyzing complex interactions among massive variables. To test the ability of machine learning algorithms for predicting risk of type 2 diabetes mellitus (T2DM) in a rural Chinese population, we focus on a total of 36,652 eligible participants from the Henan Rural Cohort Study. Risk assessment models for T2DM were developed using six machine learning algorithms, including logistic regression (LR), classification and regression tree (CART), artificial neural networks (ANN), support vector machine (SVM), random forest (RF) and gradient boosting machine (GBM). The model performance was measured in an area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, negative predictive value and area under precision recall curve. The importance of variables was identified based on each classifier and the shapley additive explanations approach. Using all available variables, all models for predicting risk of T2DM demonstrated strong predictive performance, with AUCs ranging between 0.811 and 0.872 using laboratory data and from 0.767 to 0.817 without laboratory data. Among them, the GBM model performed best (AUC: 0.872 with laboratory data and 0.817 without laboratory data). Performance of models plateaued when introduced 30 variables to each model except CART model. Among the top-10 variables across all methods were sweet flavor, urine glucose, age, heart rate, creatinine, waist circumference, uric acid, pulse pressure, insulin, and hypertension. New important risk factors (urinary indicators, sweet flavor) were not found in previous risk prediction methods, but determined by machine learning in our study. Through the results, machine learning methods showed competence in predicting risk of T2DM, leading to greater insights on disease risk factors with no priori assumption of causality. 10.1038/s41598-020-61123-x
Machine-Learning-Derived Model for the Stratification of Cardiovascular risk in Patients with Ischemic Stroke. Ntaios George,Sagris Dimitrios,Kallipolitis Athanasios,Karagkiozi Efstathia,Korompoki Eleni,Manios Efstathios,Plagianakos Vasileios,Vemmos Konstantinos,Maglogiannis Ilias Journal of stroke and cerebrovascular diseases : the official journal of National Stroke Association Background Stratification of cardiovascular risk in patients with ischemic stroke is important as it may inform management strategies. We aimed to develop a machine-learning-derived prognostic model for the prediction of cardiovascular risk in ischemic stroke patients. MATERIALS AND METHODS:Two prospective stroke registries with consecutive acute ischemic stroke patients were used as training/validation and test datasets. The outcome assessed was major adverse cardiovascular event, defined as non-fatal stroke, non-fatal myocardial infarction, and cardiovascular death during 2-year follow-up. The variables selection was performed with the LASSO technique. The algorithms XGBoost (Extreme Gradient Boosting), Random Forest and Support Vector Machines were selected according to their performance. The evaluation of the classifier was performed by bootstrapping the dataset 1000 times and performing cross-validation by splitting in 60% for the training samples and 40% for the validation samples. RESULTS:The model included age, gender, atrial fibrillation, heart failure, peripheral artery disease, arterial hypertension, statin treatment before stroke onset, prior anticoagulant treatment (in case of atrial fibrillation), creatinine, cervical artery stenosis, anticoagulant treatment at discharge (in case of atrial fibrillation), and statin treatment at discharge. The best accuracy was measured by the XGBoost classifier. In the validation dataset, the area under the curve was 0.648 (95%CI:0.619-0.675) and the balanced accuracy was 0.58 ± 0.14. In the test dataset, the corresponding values were 0.59 and 0.576. CONCLUSIONS:We propose an externally validated machine-learning-derived model which includes readily available parameters and can be used for the estimation of cardiovascular risk in ischemic stroke patients. 10.1016/j.jstrokecerebrovasdis.2021.106018
Identifying the predictive effectiveness of a genetic risk score for incident hypertension using machine learning methods among populations in rural China. Hypertension research : official journal of the Japanese Society of Hypertension Current studies have shown the controversial effect of genetic risk scores (GRSs) in hypertension prediction. Machine learning methods are used extensively in the medical field but rarely in the mining of genetic information. This study aims to determine whether genetic information can improve the prediction of incident hypertension using machine learning approaches in a prospective study. The study recruited 4592 subjects without hypertension at baseline from a cohort study conducted in rural China. A polygenic risk score (PGGRS) was calculated using 13 SNPs. According to a ratio of 7:3, subjects were randomly allocated to the train and test datasets. Models with and without the PGGRS were established using the train dataset with Cox regression, artificial neural network (ANN), random forest (RF), and gradient boosting machine (GBM) methods. The discrimination and reclassification of models were estimated using the test dataset. The PGGRS showed a significant association with the risk of incident hypertension (HR (95% CI), 1.046 (1.004, 1.090), P = 0.031) irrespective of baseline blood pressure. Models that did not include the PGGRS achieved AUCs (95% CI) of 0.785 (0.763, 0.807), 0.790 (0.768, 0.811), 0.838 (0.817, 0.857), and 0.854 (0.835, 0.873) for the Cox, ANN, RF, and GBM methods, respectively. The addition of the PGGRS led to the improvement of the AUC by 0.001, 0.008, 0.023, and 0.017; IDI by 1.39%, 2.86%, 4.73%, and 4.68%; and NRI by 25.05%, 13.01%, 44.87%, and 22.94%, respectively. Incident hypertension risk was better predicted by the traditional+PGGRS model, especially when machine learning approaches were used, suggesting that genetic information may have the potential to identify new hypertension cases using machine learning methods in resource-limited areas. CLINICAL TRIAL REGISTRATION: The Henan Rural Cohort Study has been registered at the Chinese Clinical Trial Register (Registration number: ChiCTR-OOC-15006699). http://www.chictr.org.cn/showproj.aspx?proj=11375 . 10.1038/s41440-021-00738-7
Using Machine Learning Approaches to Predict Short-Term Risk of Cardiotoxicity Among Patients with Colorectal Cancer After Starting Fluoropyrimidine-Based Chemotherapy. Li Chao,Chen Li,Chou Chiahung,Ngorsuraches Surachat,Qian Jingjing Cardiovascular toxicology Cardiotoxicity is a severe side effect for colorectal cancer (CRC) patients undergoing fluoropyrimidine-based chemotherapy. To develop and compare machine learning algorithms to predict cardiotoxicity risk among nationally representative CRC patients receiving fluoropyrimidine, CRC Patients with at least one claim of fluoropyrimidine after their cancer diagnosis were included. The outcome was the 30-day cardiotoxicity from the first day of starting fluoropyrimidine. The machine learning models including extreme gradient boosting (XGBoost), random forest (RF), and logistic regression (LR) were developed using 2006-2011 SEER-Medicare data, and model performances were evaluated using 2012-2014 data. Precision, F1 score, and area under the receiver operating characteristics curve (AUC) were measured to evaluate model performances. Feature importance plots were obtained to quantify the predictor importance. Among 36,030 CRC patients, 18.74% of them developed cardiotoxicity within 30 days since the first fluoropyrimidine. The XGBoost approach had better prediction performance with higher precision (0.619) and F1 score (0.406) in predicting the 30-day cardiotoxicity, compared to the RF (precision, 0.607 and F1 score, 0.395) and LR (precision, 0.610 and F1 score, 0.398). According to the DeLong's test for AUC difference, the XGBoost significantly outperformed the RF and LR (XGBoost, 0.816 vs. RF, 0.804, P < 0.001; XGBoost vs. LR, 0.812, P = 0.003, respectively). Feature importance plots identified pre-existing cardiac conditions, surgery, older age as top significant risk factors for cardiotoxicity events among CRC patients after receiving fluoropyrimidine. In summary, the developed machine learning models can accurately predict the occurrence of 30-day cardiotoxicity among CRC patients receiving fluoropyrimidine-based chemotherapy. 10.1007/s12012-021-09708-4
Predicting women with depressive symptoms postpartum with machine learning methods. Andersson Sam,Bathula Deepti R,Iliadis Stavros I,Walter Martin,Skalkidou Alkistis Scientific reports Postpartum depression (PPD) is a detrimental health condition that affects 12% of new mothers. Despite negative effects on mothers' and children's health, many women do not receive adequate care. Preventive interventions are cost-efficient among high-risk women, but our ability to identify these is poor. We leveraged the power of clinical, demographic, and psychometric data to assess if machine learning methods can make accurate predictions of postpartum depression. Data were obtained from a population-based prospective cohort study in Uppsala, Sweden, collected between 2009 and 2018 (BASIC study, n = 4313). Sub-analyses among women without previous depression were performed. The extremely randomized trees method provided robust performance with highest accuracy and well-balanced sensitivity and specificity (accuracy 73%, sensitivity 72%, specificity 75%, positive predictive value 33%, negative predictive value 94%, area under the curve 81%). Among women without earlier mental health issues, the accuracy was 64%. The variables setting women at most risk for PPD were depression and anxiety during pregnancy, as well as variables related to resilience and personality. Future clinical models that could be implemented directly after delivery might consider including these variables in order to identify women at high risk for postpartum depression to facilitate individualized follow-up and cost-effectiveness. 10.1038/s41598-021-86368-y
Machine learning for identification of frailty in Canadian primary care practices. International journal of population data science INTRODUCTION:Frailty is a medical syndrome, commonly affecting people aged 65 years and over and is characterized by a greater risk of adverse outcomes following illness or injury. Electronic medical records contain a large amount of longitudinal data that can be used for primary care research. Machine learning can fully utilize this wide breadth of data for the detection of diseases and syndromes. The creation of a frailty case definition using machine learning may facilitate early intervention, inform advanced screening tests, and allow for surveillance. OBJECTIVES:The objective of this study was to develop a validated case definition of frailty for the primary care context, using machine learning. METHODS:Physicians participating in the Canadian Primary Care Sentinel Surveillance Network across Canada were asked to retrospectively identify the level of frailty present in a sample of their own patients (total n 5,466), collected from 2015-2019. Frailty levels were dichotomized using a cut-off of 5. Extracted features included previously prescribed medications, billing codes, and other routinely collected primary care data. We used eight supervised machine learning algorithms, with performance assessed using a hold-out test set. A balanced training dataset was also created by oversampling. Sensitivity analyses considered two alternative dichotomization cut-offs. Model performance was evaluated using area under the receiver-operating characteristic curve, F1, accuracy, sensitivity, specificity, negative predictive value and positive predictive value. RESULTS:The prevalence of frailty within our sample was 18.4%. Of the eight models developed to identify frail patients, an XGBoost model achieved the highest sensitivity (78.14%) and specificity (74.41%). The balanced training dataset did not improve classification performance. Sensitivity analyses did not show improved performance for cut-offs other than 5. CONCLUSION:Supervised machine learning was able to create well performing classification models for frailty. Future research is needed to assess frailty inter-rater reliability, and link multiple data sources for frailty identification. 10.23889/ijpds.v6i1.1650
Machine Learning for Early Lung Cancer Identification Using Routine Clinical and Laboratory Data. Gould Michael K,Huang Brian Z,Tammemagi Martin C,Kinar Yaron,Shiff Ron American journal of respiratory and critical care medicine Most lung cancers are diagnosed at an advanced stage. Presymptomatic identification of high-risk individuals can prompt earlier intervention and improve long-term outcomes. To develop a model to predict a future diagnosis of lung cancer on the basis of routine clinical and laboratory data by using machine learning. We assembled data from 6,505 case patients with non-small cell lung cancer (NSCLC) and 189,597 contemporaneous control subjects and compared the accuracy of a novel machine learning model with a modified version of the well-validated 2012 Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial risk model (mPLCOm2012), by using the area under the receiver operating characteristic curve (AUC), sensitivity, and diagnostic odds ratio (OR) as measures of model performance. Among ever-smokers in the test set, a machine learning model was more accurate than the mPLCOm2012 for identifying NSCLC 9-12 months before clinical diagnosis ( < 0.00001) and demonstrated an AUC of 0.86, a diagnostic OR of 12.3, and a sensitivity of 40.1% at a predefined specificity of 95%. In comparison, the mPLCOm2012 demonstrated an AUC of 0.79, an OR of 7.4, and a sensitivity of 27.9% at the same specificity. The machine learning model was more accurate than standard eligibility criteria for lung cancer screening and more accurate than the mPLCOm2012 when applied to a screening-eligible population. Influential model variables included known risk factors and novel predictors such as white blood cell and platelet counts. A machine learning model was more accurate for early diagnosis of NSCLC than either standard eligibility criteria for screening or the mPLCOm2012, demonstrating the potential to help prevent lung cancer deaths through early detection. 10.1164/rccm.202007-2791OC
Prediction models applying machine learning to oral cavity cancer outcomes: A systematic review. Adeoye John,Tan Jia Yan,Choi Siu-Wai,Thomson Peter International journal of medical informatics OBJECTIVES:Machine learning platforms are now being introduced into modern oncological practice for classification and prediction of patient outcomes. To determine the current status of the application of these learning models as adjunctive decision-making tools in oral cavity cancer management, this systematic review aims to summarize the accuracy of machine-learning based models for disease outcomes. METHODS:Electronic databases including PubMed, Scopus, EMBASE, Cochrane Library, LILACS, SciELO, PsychINFO, and Web of Science were searched up until December 21, 2020. Pertinent articles detailing the development and accuracy of machine learning prediction models for oral cavity cancer outcomes were selected in a two-stage process. Quality assessment was conducted using the Quality in Prognosis Studies (QUIPS) tool and results of base studies were qualitatively synthesized by all authors. Outcomes of interest were malignant transformation of precancer lesions, cervical lymph node metastasis, as well as treatment response, and prognosis of oral cavity cancer. RESULTS:Twenty-seven articles out of 950 citations identified from electronic and manual searching were included in this study. Five studies had low bias concerns on the QUIPS tool. Prediction of malignant transformation, cervical lymph node metastasis, treatment response, and prognosis were reported in three, six, eight, and eleven articles respectively. Accuracy of these learning models on the internal or external validation sets ranged from 0.85 to 0.97 for malignant transformation prediction, 0.78-0.91 for cervical lymph node metastasis prediction, 0.64-1.00 for treatment response prediction, and 0.71-0.99 for prognosis prediction. In general, most trained algorithms predicting these outcomes performed better than alternate methods of prediction. We also found that models including molecular markers in training data had better accuracy estimates for malignant transformation, treatment response, and prognosis prediction. CONCLUSION:Machine learning algorithms have a satisfactory to excellent accuracy for predicting three of four oral cavity cancer outcomes i.e., malignant transformation, nodal metastasis, and prognosis. However, considering the training approach of many available classifiers, these models may not be streamlined enough for clinical application currently. 10.1016/j.ijmedinf.2021.104557
Depression Prediction by Using Ecological Momentary Assessment, Actiwatch Data, and Machine Learning: Observational Study on Older Adults Living Alone. Kim Heejung,Lee SungHee,Lee SangEun,Hong Soyun,Kang HeeJae,Kim Namhee JMIR mHealth and uHealth BACKGROUND:Although geriatric depression is prevalent, diagnosis using self-reporting instruments has limitations when measuring the depressed mood of older adults in a community setting. Ecological momentary assessment (EMA) by using wearable devices could be used to collect data to classify older adults into depression groups. OBJECTIVE:The objective of this study was to develop a machine learning algorithm to predict the classification of depression groups among older adults living alone. We focused on utilizing diverse data collected through a survey, an Actiwatch, and an EMA report related to depression. METHODS:The prediction model using machine learning was developed in 4 steps: (1) data collection, (2) data processing and representation, (3) data modeling (feature engineering and selection), and (4) training and validation to test the prediction model. Older adults (N=47), living alone in community settings, completed an EMA to report depressed moods 4 times a day for 2 weeks between May 2017 and January 2018. Participants wore an Actiwatch that measured their activity and ambient light exposure every 30 seconds for 2 weeks. At baseline and the end of the 2-week observation, depressive symptoms were assessed using the Korean versions of the Short Geriatric Depression Scale (SGDS-K) and the Hamilton Depression Rating Scale (K-HDRS). Conventional classification based on binary logistic regression was built and compared with 4 machine learning models (the logit, decision tree, boosted trees, and random forest models). RESULTS:On the basis of the SGDS-K and K-HDRS, 38% (18/47) of the participants were classified into the probable depression group. They reported significantly lower scores of normal mood and physical activity and higher levels of white and red, green, and blue (RGB) light exposures at different degrees of various 4-hour time frames (all P<.05). Sleep efficiency was chosen for modeling through feature selection. Comparing diverse combinations of the selected variables, daily mean EMA score, daily mean activity level, white and RGB light at 4:00 pm to 8:00 pm exposure, and daily sleep efficiency were selected for modeling. Conventional classification based on binary logistic regression had a good model fit (accuracy: 0.705; precision: 0.770; specificity: 0.859; and area under receiver operating characteristic curve or AUC: 0.754). Among the 4 machine learning models, the logit model had the best fit compared with the others (accuracy: 0.910; precision: 0.929; specificity: 0.940; and AUC: 0.960). CONCLUSIONS:This study provides preliminary evidence for developing a machine learning program to predict the classification of depression groups in older adults living alone. Clinicians should consider using this method to identify underdiagnosed subgroups and monitor daily progression regarding treatment or therapeutic intervention in the community setting. Furthermore, more efforts are needed for researchers and clinicians to diversify data collection methods by using a survey, EMA, and a sensor. 10.2196/14149
A Machine Learning Model to Successfully Predict Future Diagnosis of Chronic Myelogenous Leukemia With Retrospective Electronic Health Records Data. Hauser Ronald G,Esserman Denise,Beste Lauren A,Ong Shawn Y,Colomb Denis G,Bhargava Ankur,Wadia Roxanne,Rose Michal G American journal of clinical pathology BACKGROUND:Chronic myelogenous leukemia (CML) is a clonal stem cell disorder accounting for 15% of adult leukemias. We aimed to determine if machine learning models could predict CML using blood cell counts prior to diagnosis. METHODS:We identified patients with a diagnostic test for CML (BCR-ABL1) and at least 6 consecutive prior years of differential blood cell counts between 1999 and 2020 in the largest integrated health care system in the United States. Blood cell counts from different time periods prior to CML diagnostic testing were used to train, validate, and test machine learning models. RESULTS:The sample included 1,623 patients with BCR-ABL1 positivity rate 6.2%. The predictive ability of machine learning models improved when trained with blood cell counts closer to time of diagnosis: 2 to 5 years area under the curve (AUC), 0.59 to 0.67, 0.5 to 1 years AUC, 0.75 to 0.80, at diagnosis AUC, 0.87 to 0.92. CONCLUSIONS:Blood cell counts collected up to 5 years prior to diagnostic workup of CML successfully predicted the BCR-ABL1 test result. These findings suggest a machine learning model trained with blood cell counts could lead to diagnosis of CML earlier in the disease course compared to usual medical care. 10.1093/ajcp/aqab086
Machine learning with D-dimer in the risk stratification for pulmonary embolism: a derivation and internal validation study. European heart journal. Acute cardiovascular care AIM:To develop a machine learning model to predict the diagnosis of pulmonary embolism (PE). METHODS AND RESULTS:We undertook a derivation and internal validation study to develop a risk prediction model for use in patients being investigated for possible PE. The machine learning technique, generalized logistic regression using elastic net, was chosen following an assessment of seven machine learning techniques and on the basis that it optimized the area under the receiver operator characteristic curve (AUC) and Brier score. Models were developed both with and without the addition of D-dimer. A total of 3347 patients were included in the study of whom, 219 (6.5%) had PE. Four clinical variables (O2 saturation, previous deep venous thrombosis or PE, immobilization or surgery, and alternative diagnosis equal or more likely than PE) plus D-dimer contributed to the machine learning models. The addition of D-dimer improved the AUC by 0.16 (95% confidence interval 0.13-0.19), from 0.73 to 0.89 (0.87-0.91) and decreased the Brier score by 14% (10-18%). More could be ruled out with a higher positive likelihood ratio than by the Wells score combined with D-dimer, revised Geneva score combined with D-dimer, or the Pulmonary Embolism Rule-out Criteria score. Machine learning with D-dimer maintained a low-false-negative rate at a true-negative rate of nearly 53%, which was better performance than any of the other alternatives. CONCLUSION:A machine learning model outperformed traditional risk scores for the risk stratification of PE in the emergency department. However, external validation is needed. 10.1093/ehjacc/zuab089
Prediction of depressive symptoms onset and long-term trajectories in home-based older adults using machine learning techniques. Aging & mental health OBJECTIVES:Our aim was to explore the possibility of using machine learning (ML) in predicting the onset and trajectories of depressive symptom in home-based older adults over a 7-year period. METHODS:Depressive symptom data (collected in the year 2011, 2013, 2015 and 2018) of home-based older Chinese ( = 2650) recruited in the China Health and Retirement Longitudinal Study (CHARLS) were included in the current analysis. The latent class growth modeling (LCGM) and growth mixture modeling (GMM) were used to classify different trajectory classes. Based on the identified trajectory patterns, three ML classification algorithms (i.e. gradient boosting decision tree, support vector machine and random forest) were evaluated with a 10-fold cross-validation procedure and a metric of the area under the receiver operating characteristic curve (AUC). RESULTS:Four trajectories were identified for the depressive symptoms: no symptoms (63.9%), depressive symptoms onset {incident increasing symptoms [new-onset increasing (16.8%)], chronic symptoms [slowly decreasing (12.5%), persistent high (6.8%)]}. Among the analyzed baseline variables, the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) score, cognition, sleep time, self-reported memory were the top five important predictors across all trajectories. The mean AUCs of the three predictive models had a range from 0.661 to 0.892. CONCLUSIONS:ML techniques can be robust in predicting depressive symptom onset and trajectories over a 7-year period with easily accessible sociodemographic and health information. Supplemental data for this article is available online at http://dx.doi.org/10.1080/13607863.2022.2031868. 10.1080/13607863.2022.2031868
Neuropsychiatric symptoms as predictors of conversion from MCI to dementia: a machine learning approach. Mallo Sabela C,Valladares-Rodriguez Sonia,Facal David,Lojo-Seoane Cristina,Fernández-Iglesias Manuel J,Pereiro Arturo X International psychogeriatrics OBJECTIVES:To use a Machine Learning (ML) approach to compare Neuropsychiatric Symptoms (NPS) in participants of a longitudinal study who developed dementia and those who did not. DESIGN:Mann-Whitney U and ML analysis. Nine ML algorithms were evaluated using a 10-fold stratified validation procedure. Performance metrics (accuracy, recall, F-1 score, and Cohen's kappa) were computed for each algorithm, and graphic metrics (ROC and precision-recall curves) and features analysis were computed for the best-performing algorithm. SETTING:Primary care health centers. PARTICIPANTS:128 participants: 78 cognitively unimpaired and 50 with MCI. MEASUREMENTS:Diagnosis at baseline, months from the baseline assessment until the 3rd follow-up or development of dementia, gender, age, Charlson Comorbidity Index, Neuropsychiatric Inventory-Questionnaire (NPI-Q) individual items, NPI-Q total severity, and total stress score and Geriatric Depression Scale-15 items (GDS-15) total score. RESULTS:30 participants developed dementia, while 98 did not. Most of the participants who developed dementia were diagnosed at baseline with amnestic multidomain MCI. The Random Forest Plot model provided the metrics that best predicted conversion to dementia (e.g. accuracy=.88, F1=.67, and Cohen's kappa=.63). The algorithm indicated the importance of the metrics, in the following (decreasing) order: months from first assessment, age, the diagnostic group at baseline, total NPI-Q severity score, total NPI-Q stress score, and GDS-15 total score. CONCLUSIONS:ML is a valuable technique for detecting the risk of conversion to dementia in MCI patients. Some NPS proxies, including NPI-Q total severity score, NPI-Q total stress score, and GDS-15 total score, were deemed as the most important variables for predicting conversion, adding further support to the hypothesis that some NPS are associated with a higher risk of dementia in MCI. 10.1017/S1041610219001030
Using machine-learning to predict sudden gains in treatment for major depressive disorder. Aderka Idan M,Kauffmann Amitay,Shalom Jonathan G,Beard Courtney,Björgvinsson Thröstur Behaviour research and therapy OBJECTIVE:Sudden gains during psychotherapy have been found to consistently predict treatment outcome but evidence on predictors of sudden gains has been equivocal. To address this gap, the present study utilized three machine learning algorithms to predict sudden gains during treatment for major depressive disorder. METHOD:We examined predictors of sudden gains in two large samples of individuals receiving treatment in a partial hospital setting (n = 726 and n = 788; total N = 1514). Predictors included age, gender, marital status, education, employment status, previous hospitalization, comorbid diagnoses, and pretreatment measures of depressive and generalized anxiety symptoms. We used three machine learning models: a Random Forest model, a Random Forest model with an adaptive boosting meta-algorithm, and a Support Vector Machine model. RESULTS:In both samples, sudden gains were identified and found to significantly predict outcome. However, none of the machine learning algorithms was able to identify robust predictors of sudden gains. Thus, even though some models achieved fair prediction of sudden gains in the training subset, prediction in the test subset was poor. CONCLUSIONS:Despite the use of a large sample and three machine-learning models, we were unable to identify robust demographic and pretreatment clinical predictors of sudden gains. Implications for clinical decision making and future studies are discussed. 10.1016/j.brat.2021.103929
Prediction of diagnosis and prognosis of COVID-19 disease by blood gas parameters using decision trees machine learning model: a retrospective observational study. Huyut Mehmet Tahir,Üstündağ Hilal Medical gas research The coronavirus disease 2019 (COVID-19) epidemic went down in history as a pandemic caused by corona-viruses that emerged in 2019 and spread rapidly around the world. The different symptoms of COVID-19 made it difficult to understand which variables were more influential on the diagnosis, course and mortality of the disease. Machine learning models can accurately assess hidden patterns among risk factors by analyzing large-datasets to quickly predict diagnosis, prognosis and mortality of diseases. Because of this advantage, the use of machine learning models as decision support systems in health services is increasing. The aim of this study is to determine the diagnosis and prognosis of COVID-19 disease with blood-gas data using the Chi-squared Automatic Interaction Detector (CHAID) decision-tree-model, one of the machine learning methods, which is a subfield of artificial intelligence. This study was carried out on a total of 686 patients with COVID-19 (n = 343) and non-COVID-19 (n = 343) treated at Erzincan-Mengücek-Gazi-Training and Research-Hospital between April 1, 2020 and March 1, 2021. Arterial blood gas values of all patients were obtained from the hospital registry system. While the total-accuracyratio of the decision-tree-model was 65.0% in predicting the prognosis of the disease, it was 68.2% in the diagnosis of the disease. According to the results obtained, the low ionized-calcium value (< 1.10 mM) significantly predicted the need for intensive care of COVID-19 patients. At admission, low-carboxyhemoglobin (< 1.00%), high-pH (> 7.43), low-sodium (< 135.0 mM), hematocrit (< 40.0%), and methemoglobin (< 1.30%) values are important biomarkers in the diagnosis of COVID-19 and the results were promising. The findings in the study may aid in the early-diagnosis of the disease and the intensive-care treatment of patients who are severe. The study was approved by the Ministry of Health and Erzincan University Faculty of Medicine Clinical Research Ethics Committee. 10.4103/2045-9912.326002
Machine Learning Methods to Extract Documentation of Breast Cancer Symptoms From Electronic Health Records. Forsyth Alexander W,Barzilay Regina,Hughes Kevin S,Lui Dickson,Lorenz Karl A,Enzinger Andrea,Tulsky James A,Lindvall Charlotta Journal of pain and symptom management CONTEXT:Clinicians document cancer patients' symptoms in free-text format within electronic health record visit notes. Although symptoms are critically important to quality of life and often herald clinical status changes, computational methods to assess the trajectory of symptoms over time are woefully underdeveloped. OBJECTIVES:To create machine learning algorithms capable of extracting patient-reported symptoms from free-text electronic health record notes. METHODS:The data set included 103,564 sentences obtained from the electronic clinical notes of 2695 breast cancer patients receiving paclitaxel-containing chemotherapy at two academic cancer centers between May 1996 and May 2015. We manually annotated 10,000 sentences and trained a conditional random field model to predict words indicating an active symptom (positive label), absence of a symptom (negative label), or no symptom at all (neutral label). Sentences labeled by human coder were divided into training, validation, and test data sets. Final model performance was determined on 20% test data unused in model development or tuning. RESULTS:The final model achieved precision of 0.82, 0.86, and 0.99 and recall of 0.56, 0.69, and 1.00 for positive, negative, and neutral symptom labels, respectively. The most common positive symptoms were pain, fatigue, and nausea. Machine-based labeling of 103,564 sentences took two minutes. CONCLUSION:We demonstrate the potential of machine learning to gather, track, and analyze symptoms experienced by cancer patients during chemotherapy. Although our initial model requires further optimization to improve the performance, further model building may yield machine learning methods suitable to be deployed in routine clinical care, quality improvement, and research applications. 10.1016/j.jpainsymman.2018.02.016
Predicting persistent depressive symptoms in older adults: A machine learning approach to personalised mental healthcare. Hatton Christopher M,Paton Lewis W,McMillan Dean,Cussens James,Gilbody Simon,Tiffin Paul A Journal of affective disorders BACKGROUND:Depression causes significant physical and psychosocial morbidity. Predicting persistence of depressive symptoms could permit targeted prevention, and lessen the burden of depression. Machine learning is a rapidly expanding field, and such approaches offer powerful predictive abilities. We investigated the utility of a machine learning approach to predict the persistence of depressive symptoms in older adults. METHOD:Baseline demographic and psychometric data from 284 patients were used to predict the likelihood of older adults having persistent depressive symptoms after 12 months, using a machine learning approach ('extreme gradient boosting'). Predictive performance was compared to a conventional statistical approach (logistic regression). Data were drawn from the 'treatment-as-usual' arm of the CASPER (CollAborative care and active surveillance for Screen-Positive EldeRs with subthreshold depression) trial. RESULTS:Predictive performance was superior using machine learning compared to logistic regression (mean AUC 0.72 vs. 0.67, p < 0.0001). Using machine learning, an average of 89% of those predicted to have PHQ-9 scores above threshold at 12 months actually did, compared to 78% using logistic regression. However, mean negative predictive values were somewhat lower for the machine learning approach (45% vs. 35%). LIMITATIONS:A relatively small sample size potentially limited the predictive power of the algorithm. In addition, PHQ-9 scores were used as an indicator of persistent depressive symptoms, and whilst well validated, a clinical interview would have been preferable. CONCLUSIONS:Overall, our findings support the potential application of machine learning in personalised mental healthcare. 10.1016/j.jad.2018.12.095
Relative importance of symptoms, cognition, and other multilevel variables for psychiatric disease classifications by machine learning. Walsh-Messinger Julie,Jiang Haoran,Lee Hyejoo,Rothman Karen,Ahn Hongshik,Malaspina Dolores Psychiatry research This study used machine-learning algorithms to make unbiased estimates of the relative importance of various multilevel data for classifying cases with schizophrenia (n = 60), schizoaffective disorder (n = 19), bipolar disorder (n = 20), unipolar depression (n = 14), and healthy controls (n = 51) into psychiatric diagnostic categories. The Random Forest machine learning algorithm, which showed best efficacy (92.9% SD: 0.06), was used to generate variable importance ranking of positive, negative, and general psychopathology symptoms, cognitive indexes, global assessment of function (GAF), and parental ages at birth for sorting participants into diagnostic categories. Symptoms were ranked most influential for separating cases from healthy controls, followed by cognition and maternal age. To separate schizophrenia/schizoaffective disorder from bipolar/unipolar depression, GAF was most influential, followed by cognition and paternal age. For classifying schizophrenia from all other psychiatric disorders, low GAF and paternal age were similarly important, followed by cognition, psychopathology and maternal age. Controls misclassified as schizophrenia cases showed lower nonverbal abilities, mild negative and general psychopathology symptoms, and younger maternal or older paternal age. The importance of symptoms for classification of cases and lower GAF for diagnosing schizophrenia, notably more important and distinct from cognition and symptoms, concurs with current practices. The high importance of parental ages is noteworthy and merits further study. 10.1016/j.psychres.2019.03.048
Validation of Random Forest Machine Learning Models to Predict Dementia-Related Neuropsychiatric Symptoms in Real-World Data. Journal of Alzheimer's disease : JAD BACKGROUND:Neuropsychiatric symptoms (NPS) are the leading cause of the social burden of dementia but their role is underestimated. OBJECTIVE:The objective of the study was to validate predictive models to separately identify psychotic and depressive symptoms in patients diagnosed with dementia using clinical databases representing the whole population to inform decision-makers. METHODS:First, we searched the electronic health records of 4,003 patients with dementia to identify NPS. Second, machine learning (random forest) algorithms were applied to build separate predictive models for psychotic and depressive symptom clusters in the training set (N = 3,003). Third, calibration and discrimination were assessed in the test set (N = 1,000) to assess the performance of the models. RESULTS:Neuropsychiatric symptoms were noted in the electronic health record of 58% of patients. The area under the receiver operating curve reached 0.80 for the psychotic cluster model and 0.74 for the depressive cluster model. The Kappa index and accuracy also showed better discrimination in the psychotic model. Calibration plots indicated that both types of model had less predictive accuracy when the probability of neuropsychiatric symptoms was <25%. The most important variables in the psychotic cluster model were use of risperidone, level of sedation, use of quetiapine and haloperidol and the number of antipsychotics prescribed. In the depressive cluster model, the most important variables were number of antidepressants prescribed, escitalopram use, level of sedation, and age. CONCLUSION:Given their relatively good performance, the predictive models can be used to estimate prevalence of NPS in population databases. 10.3233/JAD-200345
Automatic Grading of Stroke Symptoms for Rapid Assessment Using Optimized Machine Learning and 4-Limb Kinematics: Clinical Validation Study. Park Eunjeong,Lee Kijeong,Han Taehwa,Nam Hyo Suk Journal of medical Internet research BACKGROUND:Subtle abnormal motor signs are indications of serious neurological diseases. Although neurological deficits require fast initiation of treatment in a restricted time, it is difficult for nonspecialists to detect and objectively assess the symptoms. In the clinical environment, diagnoses and decisions are based on clinical grading methods, including the National Institutes of Health Stroke Scale (NIHSS) score or the Medical Research Council (MRC) score, which have been used to measure motor weakness. Objective grading in various environments is necessitated for consistent agreement among patients, caregivers, paramedics, and medical staff to facilitate rapid diagnoses and dispatches to appropriate medical centers. OBJECTIVE:In this study, we aimed to develop an autonomous grading system for stroke patients. We investigated the feasibility of our new system to assess motor weakness and grade NIHSS and MRC scores of 4 limbs, similar to the clinical examinations performed by medical staff. METHODS:We implemented an automatic grading system composed of a measuring unit with wearable sensors and a grading unit with optimized machine learning. Inertial sensors were attached to measure subtle weaknesses caused by paralysis of upper and lower limbs. We collected 60 instances of data with kinematic features of motor disorders from neurological examination and demographic information of stroke patients with NIHSS 0 or 1 and MRC 7, 8, or 9 grades in a stroke unit. Training data with 240 instances were generated using a synthetic minority oversampling technique to complement the imbalanced number of data between classes and low number of training data. We trained 2 representative machine learning algorithms, an ensemble and a support vector machine (SVM), to implement auto-NIHSS and auto-MRC grading. The optimized algorithms performed a 5-fold cross-validation and were searched by Bayes optimization in 30 trials. The trained model was tested with the 60 original hold-out instances for performance evaluation in accuracy, sensitivity, specificity, and area under the receiver operating characteristics curve (AUC). RESULTS:The proposed system can grade NIHSS scores with an accuracy of 83.3% and an AUC of 0.912 using an optimized ensemble algorithm, and it can grade with an accuracy of 80.0% and an AUC of 0.860 using an optimized SVM algorithm. The auto-MRC grading achieved an accuracy of 76.7% and a mean AUC of 0.870 in SVM classification and an accuracy of 78.3% and a mean AUC of 0.877 in ensemble classification. CONCLUSIONS:The automatic grading system quantifies proximal weakness in real time and assesses symptoms through automatic grading. The pilot outcomes demonstrated the feasibility of remote monitoring of motor weakness caused by stroke. The system can facilitate consistent grading with instant assessment and expedite dispatches to appropriate hospitals and treatment initiation by sharing auto-MRC and auto-NIHSS scores between prehospital and hospital responses as an objective observation. 10.2196/20641
Use of Patient-Reported Symptoms from an Online Symptom Tracking Tool for Dementia Severity Staging: Development and Validation of a Machine Learning Approach. Shehzad Aaqib,Rockwood Kenneth,Stanley Justin,Dunn Taylor,Howlett Susan E Journal of medical Internet research BACKGROUND:SymptomGuide Dementia (DGI Clinical Inc) is a publicly available online symptom tracking tool to support caregivers of persons living with dementia. The value of such data are enhanced when the specific dementia stage is identified. OBJECTIVE:We aimed to develop a supervised machine learning algorithm to classify dementia stages based on tracked symptoms. METHODS:We employed clinical data from 717 people from 3 sources: (1) a memory clinic; (2) long-term care; and (3) an open-label trial of donepezil in vascular and mixed dementia (VASPECT). Symptoms were captured with SymptomGuide Dementia. A clinician classified participants into 4 groups using either the Functional Assessment Staging Test or the Global Deterioration Scale as mild cognitive impairment, mild dementia, moderate dementia, or severe dementia. Individualized symptom profiles from the pooled data were used to train machine learning models to predict dementia severity. Models trained with 6 different machine learning algorithms were compared using nested cross-validation to identify the best performing model. Model performance was assessed using measures of balanced accuracy, precision, recall, Cohen κ, area under the receiver operating characteristic curve (AUROC), and area under the precision-recall curve (AUPRC). The best performing algorithm was used to train a model optimized for balanced accuracy. RESULTS:The study population was mostly female (424/717, 59.1%), older adults (mean 77.3 years, SD 10.6, range 40-100) with mild to moderate dementia (332/717, 46.3%). Age, duration of symptoms, 37 unique dementia symptoms, and 10 symptom-derived variables were used to distinguish dementia stages. A model trained with a support vector machine learning algorithm using a one-versus-rest approach showed the best performance. The correct dementia stage was identified with 83% balanced accuracy (Cohen κ=0.81, AUPRC 0.91, AUROC 0.96). The best performance was seen when classifying severe dementia (AUROC 0.99). CONCLUSIONS:A supervised machine learning algorithm exhibited excellent performance in identifying dementia stages based on dementia symptoms reported in an online environment. This novel dementia staging algorithm can be used to describe dementia stage based on user-reported symptoms. This type of symptom recording offers real-world data that reflect important symptoms in people with dementia. 10.2196/20840
Preictal state detection using prodromal symptoms: A machine learning approach. Cousyn Louis,Navarro Vincent,Chavez Mario Epilepsia A reliable identification of a high-risk state for upcoming seizures may allow for preemptive treatment and improve the quality of patients' lives. We evaluated the ability of prodromal symptoms to predict preictal states using a machine learning (ML) approach. Twenty-four patients with drug-resistant epilepsy were admitted for continuous video-electroencephalographic monitoring and filled out a daily four-point questionnaire on prodromal symptoms. Data were then classified into (1) a preictal group for questionnaires completed in a 24-h period prior to at least one seizure (n  = 58) and (2) an interictal group for questionnaires completed in a 24-h period without seizures (n  = 190). Our prediction model was based on a support vector machine classifier and compared to a Fisher's linear classifier. The combination of all the prodromal symptoms yielded a good prediction performance (area under the curve [AUC] = .72, 95% confidence interval [CI] = .61-.81). This performance was significantly enhanced by selecting a subset of the most relevant symptoms (AUC = .80, 95% CI = .69-.88). In comparison, the linear classifier systematically failed (AUCs < .6). Our findings indicate that the ML analysis of prodromal symptoms is a promising approach to identifying preictal states prior to seizures. This could pave the way for development of clinical strategies in seizure prevention and even a noninvasive alarm system. 10.1111/epi.16804
Machine Learning-Based Automatic Rating for Cardinal Symptoms of Parkinson Disease. Park Kye Won,Lee Eun-Jae,Lee Jun Seong,Jeong Jinhoon,Choi Nari,Jo Sungyang,Jung Mina,Do Ja Yeon,Kang Dong-Wha,Lee June-Goo,Chung Sun Ju Neurology OBJECTIVE:We developed and investigated the feasibility of a machine learning-based automated rating for the 2 cardinal symptoms of Parkinson disease (PD): resting tremor and bradykinesia. METHODS:Using OpenPose, a deep learning-based human pose estimation program, we analyzed video clips for resting tremor and finger tapping of the bilateral upper limbs of 55 patients with PD (110 arms). Key motion parameters, including resting tremor amplitude and finger tapping speed, amplitude, and fatigue, were extracted to develop a machine learning-based automatic Unified Parkinson's Disease Rating Scale (UPDRS) rating using support vector machine (SVM) method. To evaluate the performance of this model, we calculated weighted κ and intraclass correlation coefficients (ICCs) between the model and the gold standard rating by a movement disorder specialist who is trained and certified by the Movement Disorder Society for UPDRS rating. These values were compared to weighted κ and ICC between a nontrained human rater and the gold standard rating. RESULTS:For resting tremors, the SVM model showed a very good to excellent reliability range with the gold standard rating (κ 0.791; ICC 0.927), with both values higher than that of nontrained human rater (κ 0.662; ICC 0.861). For finger tapping, the SVM model showed a very good reliability range with the gold standard rating (κ 0.700 and ICC 0.793), which was comparable to that for nontrained human raters (κ 0.627; ICC 0.797). CONCLUSION:Machine learning-based algorithms that automatically rate PD cardinal symptoms are feasible, with more accurate results than nontrained human ratings. CLASSIFICATION OF EVIDENCE:This study provides Class II evidence that machine learning-based automated rating of resting tremor and bradykinesia in people with PD has very good reliability compared to a rating by a movement disorder specialist. 10.1212/WNL.0000000000011654
Machine Learning Approaches to Identify Factors Associated with Women's Vasomotor Symptoms Using General Hospital Data. Ryu Ki Jin,Yi Kyong Wook,Kim Yong Jin,Shin Jung Ho,Hur Jun Young,Kim Tak,Seo Jong Bae,Lee Kwang Sig,Park Hyuntae Journal of Korean medical science BACKGROUND:To analyze the factors associated with women's vasomotor symptoms (VMS) using machine learning. METHODS:Data on 3,298 women, aged 40-80 years, who attended their general health check-up from January 2010 to December 2012 were obtained from Korea University Anam Hospital in Seoul, Korea. Five machine learning methods were applied and compared for the prediction of VMS, measured by the Menopause Rating Scale. Variable importance, the effect of a variable on model performance, was used for identifying the major factors associated with VMS. RESULTS:In terms of the mean squared error, the random forest (0.9326) was much better than linear regression (12.4856) and artificial neural networks with one, two, and three hidden layers (1.5576, 1.5184, and 1.5833, respectively). Based on the variable importance from the random forest, the most important factors associated with VMS were age, menopause age, thyroid-stimulating hormone, and monocyte, triglyceride, gamma glutamyl transferase, blood urea nitrogen, cancer antigen 19-9, C-reactive protein, and low-density lipoprotein cholesterol levels. Indeed, the following variables were ranked within the top 20 in terms of variable importance: cancer antigen 125, total cholesterol, insulin, free thyroxine, forced vital capacity, alanine aminotransferase, forced expired volume in 1 second, height, homeostatic model assessment for insulin resistance, and carcinoembryonic antigen. CONCLUSION:Machine learning provides an invaluable decision support system for the prediction of VMS. For managing VMS, comprehensive consideration is needed regarding thyroid function, lipid profile, liver function, inflammation markers, insulin resistance, monocyte count, cancer antigens, and lung function. 10.3346/jkms.2021.36.e122
Predicting the development of adverse cardiac events in patients with hypertrophic cardiomyopathy using machine learning. Kochav Stephanie M,Raita Yoshihiko,Fifer Michael A,Takayama Hiroo,Ginns Jonathan,Maurer Mathew S,Reilly Muredach P,Hasegawa Kohei,Shimada Yuichi J International journal of cardiology BACKGROUND:Only a subset of patients with hypertrophic cardiomyopathy (HCM) develop adverse cardiac events - e.g., end-stage heart failure, cardiovascular death. Current risk stratification methods are imperfect, limiting identification of high-risk patients with HCM. Our aim was to improve the prediction of adverse cardiac events in patients with HCM using machine learning methods. METHODS:We applied modern machine learning methods to a prospective cohort of adults with HCM. The outcome was a composite of death due to heart failure, heart transplant, and sudden death. As the reference model, we constructed logistic regression model using known predictors. We determined 20 predictive characteristics based on random forest classification and a priori knowledge, and developed 4 machine learning models. Results Of 183 patients in the cohort, the mean age was 53 (SD = 17) years and 45% were female. During the median follow-up of 2.2 years (interquartile range, 0.6-3.8), 33 subjects (18%) developed an outcome event, the majority of which (85%) was heart transplant. The predictive accuracy of the reference model was 73% (sensitivity 76%, specificity 72%) while that of the machine learning model was 85% (e.g., sensitivity 88%, specificity 84% with elastic net regression). All 4 machine learning models significantly outperformed the reference model - e.g., area under the receiver-operating-characteristic curve 0.79 with the reference model vs. 0.93 with elastic net regression (p < 0.001). CONCLUSIONS:Compared with conventional risk stratification, the machine learning models demonstrated a superior ability to predict adverse cardiac events. These modern machine learning methods may enhance identification of high-risk HCM subpopulations. 10.1016/j.ijcard.2020.11.003
Performance of Machine Learning Algorithms for Predicting Progression to Dementia in Memory Clinic Patients. James Charlotte,Ranson Janice M,Everson Richard,Llewellyn David J JAMA network open Importance:Machine learning algorithms could be used as the basis for clinical decision-making aids to enhance clinical practice. Objective:To assess the ability of machine learning algorithms to predict dementia incidence within 2 years compared with existing models and determine the optimal analytic approach and number of variables required. Design, Setting, and Participants:This prognostic study used data from a prospective cohort of 15 307 participants without dementia at baseline to perform a secondary analysis of factors that could be used to predict dementia incidence. Participants attended National Alzheimer Coordinating Center memory clinics across the United States between 2005 and 2015. Analyses were conducted from March to May 2021. Exposures:258 variables spanning domains of dementia-related clinical measures and risk factors. Main Outcomes and Measures:The main outcome was incident all-cause dementia diagnosed within 2 years of baseline assessment. Results:In a sample of 15 307 participants (mean [SD] age, 72.3 [9.8] years; 9129 [60%] women and 6178 [40%] men) without dementia at baseline, 1568 (10%) received a diagnosis of dementia within 2 years of their initial assessment. Compared with 2 existing models for dementia risk prediction (ie, Cardiovascular Risk Factors, Aging, and Incidence of Dementia Risk Score, and the Brief Dementia Screening Indicator), machine learning algorithms were superior in predicting incident all-cause dementia within 2 years. The gradient-boosted trees algorithm had a mean (SD) overall accuracy of 92% (1%), sensitivity of 0.45 (0.05), specificity of 0.97 (0.01), and area under the curve of 0.92 (0.01) using all 258 variables. Analysis of variable importance showed that only 6 variables were required for machine learning algorithms to achieve an accuracy of 91% and area under the curve of at least 0.89. Machine learning algorithms also identified up to 84% of participants who received an initial dementia diagnosis that was subsequently reversed to mild cognitive impairment or cognitively unimpaired, suggesting possible misdiagnosis. Conclusions and Relevance:These findings suggest that machine learning algorithms could accurately predict incident dementia within 2 years in patients receiving care at memory clinics using only 6 variables. These findings could be used to inform the development and validation of decision-making aids in memory clinics. 10.1001/jamanetworkopen.2021.36553
Using machine learning to identify diabetes patients with canagliflozin prescriptions at high-risk of lower extremity amputation using real-world data. Yang Lanting,Gabriel Nico,Hernandez Inmaculada,Winterstein Almut G,Guo Jingchuan Pharmacoepidemiology and drug safety AIMS:Canagliflozin, a sodium-glucose cotransporter 2 inhibitor indicated for lowering glucose, has been increasingly used in diabetes patients because of its beneficial effects on cardiovascular and renal outcomes. However, clinical trials have documented an increased risk of lower extremity amputations (LEA) associated with canagliflozin. We applied machine learning methods to predict LEA among diabetes patients treated with canagliflozin. METHODS:Using claims data from a 5% random sample of Medicare beneficiaries, we identified 13 904 diabetes individuals initiating canagliflozin between April 2013 and December 2016. The samples were randomly and equally split into training and testing sets. We identified 41 predictor candidates using information from the year prior to canagliflozin initiation, and applied four machine learning approaches (elastic net, least absolute shrinkage and selection operator [LASSO], gradient boosting machine and random forests) to predict LEA risk after canagliflozin initiation. RESULTS:The incidence rate of LEA was 0.57% over a median 1.5 years follow-up. LASSO produced the best prediction, yielding a C-statistic of 0.81 (95% CI: 0.76, 0.86). Among individuals categorized in the top 5% of the risk score, the actual incidence rate of LEA was 3.74%. Among the 16 factors selected by LASSO, history of LEA [adjusted odds ratio (aOR): 33.6 (13.8, 81.9)] and loop diuretic use [aOR: 3.6 (1.8,7.3)] had the strongest associations with LEA incidence. CONCLUSIONS:Our machine learning model efficiently predicted the risk of LEA among diabetes patients undergoing canagliflozin treatment. The risk score may support optimized treatment decisions and thus improve health outcomes of diabetes patients. 10.1002/pds.5206
Machine learning versus traditional risk stratification methods in acute coronary syndrome: a pooled randomized clinical trial analysis. Gibson William J,Nafee Tarek,Travis Ryan,Yee Megan,Kerneis Mathieu,Ohman Magnus,Gibson C Michael Journal of thrombosis and thrombolysis Traditional statistical models allow population based inferences and comparisons. Machine learning (ML) explores datasets to develop algorithms that do not assume linear relationships between variables and outcomes and that may account for higher order interactions to make individualized outcome predictions. To evaluate the performance of machine learning models compared to traditional risk stratification methods for the prediction of major adverse cardiovascular events (MACE) and bleeding in patients with acute coronary syndrome (ACS) that are treated with antithrombotic therapy. Data on 24,178 ACS patients were pooled from four randomized controlled trials. The super learner ensemble algorithm selected weights for 23 machine learning models and was compared to traditional models. The efficacy endpoint was a composite of cardiovascular death, myocardial infarction, or stroke. The safety endpoint was a composite of TIMI major and minor bleeding or bleeding requiring medical attention. For the MACE outcome, the super learner model produced a higher c-statistic (0.734) than logistic regression (0.714), the TIMI risk score (0.489), and a new cardiovascular risk score developed in the dataset (0.644). For the bleeding outcome, the super learner demonstrated a similar c-statistic as the logistic regression model (0.670 vs. 0.671). The machine learning risk estimates were highly calibrated with observed efficacy and bleeding outcomes (Hosmer-Lemeshow p value = 0.692 and 0.970, respectively). The super learner algorithm was highly calibrated on both efficacy and safety outcomes and produced the highest c-statistic for prediction of MACE compared to traditional risk stratification methods. This analysis demonstrates a contemporary application of machine learning to guide patient-level antithrombotic therapy treatment decisions.Clinical Trial Registration ATLAS ACS-2 TIMI 46: https://clinicaltrials.gov/ct2/show/NCT00402597. Unique Identifier: NCT00402597. ATLAS ACS-2 TIMI 51: https://clinicaltrials.gov/ct2/show/NCT00809965. Unique Identifier: NCT00809965. GEMINI ACS-1: https://clinicaltrials.gov/ct2/show/NCT02293395. Unique Identifier: NCT02293395. PIONEER-AF PCI: https://clinicaltrials.gov/ct2/show/NCT01830543. Unique Identifier: NCT01830543. 10.1007/s11239-019-01940-8
Predicting Australian Adults at High Risk of Cardiovascular Disease Mortality Using Standard Risk Factors and Machine Learning. Sajeev Shelda,Champion Stephanie,Beleigoli Alline,Chew Derek,Reed Richard L,Magliano Dianna J,Shaw Jonathan E,Milne Roger L,Appleton Sarah,Gill Tiffany K,Maeder Anthony International journal of environmental research and public health Effective cardiovascular disease (CVD) prevention relies on timely identification and intervention for individuals at risk. Conventional formula-based techniques have been demonstrated to over- or under-predict the risk of CVD in the Australian population. This study assessed the ability of machine learning models to predict CVD mortality risk in the Australian population and compare performance with the well-established Framingham model. Data is drawn from three Australian cohort studies: the North West Adelaide Health Study (NWAHS), the Australian Diabetes, Obesity, and Lifestyle study, and the Melbourne Collaborative Cohort Study (MCCS). Four machine learning models for predicting 15-year CVD mortality risk were developed and compared to the 2008 Framingham model. Machine learning models performed significantly better compared to the Framingham model when applied to the three Australian cohorts. Machine learning based models improved prediction by 2.7% to 5.2% across three Australian cohorts. In an aggregated cohort, machine learning models improved prediction by up to 5.1% (area-under-curve (AUC) 0.852, 95% CI 0.837-0.867). Net reclassification improvement (NRI) was up to 26% with machine learning models. Machine learning based models also showed improved performance when stratified by sex and diabetes status. Results suggest a potential for improving CVD risk prediction in the Australian population using machine learning models. 10.3390/ijerph18063187
Using machine-learning algorithms to identify patients at high risk of upper gastrointestinal lesions for endoscopy. Liu Yongjia,Lin Da,Li Lan,Chen Yu,Wen Jiayao,Lin Yiguang,He Xingxiang Journal of gastroenterology and hepatology BACKGROUND AND AIM:Endoscopic screening for early detection of upper gastrointestinal (UGI) lesions is important. However, population-based endoscopic screening is difficult to implement in populous countries. By identifying high-risk individuals from the general population, the screening targets can be narrowed to individuals who are in most need of an endoscopy. This study was designed to develop an artificial intelligence (AI)-based model to predict patient risk of UGI lesions to identify high-risk individuals for endoscopy. METHODS:A total of 620 patients (from 5300 participants) were equally allocated into 10 parts for 10-fold cross validation experiments. The machine-learning predictive models for UGI lesion risk were constructed using random forest, logistic regression, decision tree, and support vector machine (SVM) algorithms. A total of 48 variables covering lifestyles, social-economic status, clinical symptoms, serological results, and pathological data were used in the model construction. RESULTS:The accuracies of the four models were between 79.3% and 93.4% in the training set and between 77.2% and 91.2% in the testing dataset (logistics regression: 77.2%; decision tree: 87.3%; random forest: 88.2%; SVM: 91.2%;). The AUCs of four models showed impressive predictive ability. Comparing the four models with the different algorithms, the SVM model featured the best sensitivity and specificity in all datasets tested. CONCLUSIONS:Machine-learning algorithms can accurately and reliably predict the risk of UGI lesions based on readily available parameters. The predictive models have the potential to be used clinically for identifying patients with high risk of UGI lesions and stratifying patients for necessary endoscopic screening. 10.1111/jgh.15530
Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection. Journal of translational medicine BACKGROUND:Surgical resection is the only potentially curative treatment for pancreatic ductal adenocarcinoma (PDAC) and the survival of patients after radical resection is closely related to relapse. We aimed to develop models to predict the risk of relapse using machine learning methods based on multiple clinical parameters. METHODS:Data were collected and analysed of 262 PDAC patients who underwent radical resection at 3 institutions between 2013 and 2017, with 183 from one institution as a training set, 79 from the other 2 institution as a validation set. We developed and compared several predictive models to predict 1- and 2-year relapse risk using machine learning approaches. RESULTS:Machine learning techniques were superior to conventional regression-based analyses in predicting risk of relapse of PDAC after radical resection. Among them, the random forest (RF) outperformed other methods in the training set. The highest accuracy and area under the receiver operating characteristic curve (AUROC) for predicting 1-year relapse risk with RF were 78.4% and 0.834, respectively, and for 2-year relapse risk were 95.1% and 0.998. However, the support vector machine (SVM) model showed better performance than the others for predicting 1-year relapse risk in the validation set. And the k neighbor algorithm (KNN) model achieved the highest accuracy and AUROC for predicting 2-year relapse risk. CONCLUSIONS:By machine learning, this study has developed and validated comprehensive models integrating clinicopathological characteristics to predict the relapse risk of PDAC after radical resection which will guide the development of personalized surveillance programs after surgery. 10.1186/s12967-021-02955-7
Using machine learning-based analytics of daily activities to identify modifiable risk factors for falling in Parkinson's disease. Panyakaew Pattamon,Pornputtapong Natapol,Bhidayasiri Roongroj Parkinsonism & related disorders BACKGROUND:Although risk factors that lead to falling in Parkinson's disease (PD) have been previously studied, the established predictors are mostly non-modifiable. A novel method for fall risk assessment may provide more insight into preventable high-risk activities to reduce future falls. OBJECTIVES:To explore the prediction of falling in PD patients using a machine learning-based approach. METHOD:305 PD patients, with or without a history of falls within the past month, were recruited. Data including clinical demographics, medications, and balance confidence, scaled by the 16-item Activities-Specific Balance Confidence Scale (ABC-16), were entered into the supervised machine learning models using XGBoost to explore the prediction of fallers/recurrent fallers in two separate models. RESULTS:99 (32%) patients were fallers and 58 (19%) were recurrent fallers. The accuracy of the model to predict falls was 72% (p = 0.001). The most important factors were item 7 (sweeping the floor), item 5 (reaching on tiptoes), and item 12 (walking in a crowded mall) in the ABC-16 scale, followed by disease stage and duration. When recurrent falls were analysed, the models had higher accuracy (81%, p = 0.02). The strongest predictors of recurrent falls were item 12, 5, and 10 (walking across parking lot), followed by disease stage and current age. CONCLUSION:Our machine learning-based study demonstrated that predictors of falling combined demographics of PD with environmental factors, including high-risk activities that require cognitive attention and changes in vertical and lateral orientations. This enables physicians to focus on modifiable factors and appropriately implement fall prevention strategies for individual patients. 10.1016/j.parkreldis.2020.11.014
A Postural Assessment Utilizing Machine Learning Prospectively Identifies Older Adults at a High Risk of Falling. Forth Katharine E,Wirfel Kelly L,Adams Sasha D,Rianon Nahid J,Lieberman Aiden Erez,Madansingh Stefan I Frontiers in medicine Falls are the leading cause of accidental death in older adults. Each year, 28.7% of US adults over 65 years experience a fall resulting in over 300,000 hip fractures and $50 billion in medical costs. Annual fall risk assessments have become part of the standard care plan for older adults. However, the effectiveness of these assessments in identifying at-risk individuals remains limited. This study characterizes the performance of a commercially available, automated method, for assessing fall risk using machine learning. Participants ( = 209) were recruited from eight senior living facilities and from adults living in the community (five local community centers in Houston, TX) to participate in a 12-month retrospective and a 12-month prospective cohort study. Upon enrollment, each participant stood for 60 s, with eyes open, on a commercial balance measurement platform which uses force-plate technology to capture center-of-pressure (60 Hz frequency). Linear and non-linear components of the center-of-pressure were analyzed using a machine-learning algorithm resulting in a postural stability (PS) score (range 1-10). A higher PS score indicated greater stability. Participants were contacted monthly for a year to track fall events and determine fall circumstances. Reliability among repeated trials, past and future fall prediction, as well as survival analyses, were assessed. Measurement reliability was found to be high (ICC(2,1) [95% CI]=0.78 [0.76-0.81]). Individuals in the high-risk range (1-3) were three times more likely to fall within a year than those in low-risk (7-10). They were also an order of magnitude more likely (12/104 vs. 1/105) to suffer a spontaneous fall i.e., a fall where no cause was self-reported. Survival analyses suggests a fall event within 9 months (median) for high risk individuals. We demonstrate that an easy-to-use, automated method for assessing fall risk can reliably predict falls a year in advance. Objective identification of at-risk patients will aid clinicians in providing individualized fall prevention care. 10.3389/fmed.2020.591517
A Machine Learning-Based Fall Risk Assessment Model for Inpatients. Computers, informatics, nursing : CIN Falls are one of the most common accidents among inpatients and may result in extended hospitalization and increased medical costs. Constructing a highly accurate fall prediction model could effectively reduce the rate of patient falls, further reducing unnecessary medical costs and patient injury. This study applied data mining techniques on a hospital's electronic medical records database comprising a nursing information system to construct inpatient-fall-prediction models for use during various stages of inpatient care. The inpatient data were collected from 15 inpatient wards. To develop timely and effective fall prediction models for inpatients, we retrieved the data of multiple-time assessment variables at four points during hospitalization. This study used various supervised machine learning algorithms to build classification models. Four supervised learning and two classifier ensemble techniques were selected for model development. The results indicated that Bagging+RF classifiers yielded optimal prediction performance at all four points during hospitalization. This study suggests that nursing personnel should be aware of patients' risk factors based on comprehensive fall risk assessment and provide patients with individualized fall prevention interventions to reduce inpatient fall rates. 10.1097/CIN.0000000000000727
XGBoost based machine learning approach to predict the risk of fall in older adults using gait outcomes. Noh Byungjoo,Youm Changhong,Goh Eunkyoung,Lee Myeounggon,Park Hwayoung,Jeon Hyojeong,Kim Oh Yoen Scientific reports This study aimed to identify the optimal features of gait parameters to predict the fall risk level in older adults. The study included 746 older adults (age: 63-89 years). Gait tests (20 m walkway) included speed modification (slower, preferred, and faster-walking) while wearing the inertial measurement unit sensors embedded in the shoe-type data loggers on both outsoles. A metric was defined to classify the fall risks, determined based on a set of questions determining the history of falls and fear of falls. The extreme gradient boosting (XGBoost) model was built from gait features to predict the factor affecting the risk of falls. Moreover, the definition of the fall levels was classified into high- and low-risk groups. At all speeds, three gait features were identified with the XGBoost (stride length, walking speed, and stance phase) that accurately classified the fall risk levels. The model accuracy in classifying fall risk levels ranged between 67-70% with 43-53% sensitivity and 77-84% specificity. Thus, we identified the optimal gait features for accurate fall risk level classification in older adults. The XGBoost model could inspire future works on fall prevention and the fall-risk assessment potential through the gait analysis of older adults. 10.1038/s41598-021-91797-w
Latest Research Trends in Fall Detection and Prevention Using Machine Learning: A Systematic Review. Usmani Sara,Saboor Abdul,Haris Muhammad,Khan Muneeb A,Park Heemin Sensors (Basel, Switzerland) Falls are unusual actions that cause a significant health risk among older people. The growing percentage of people of old age requires urgent development of fall detection and prevention systems. The emerging technology focuses on developing such systems to improve quality of life, especially for the elderly. A fall prevention system tries to predict and reduce the risk of falls. In contrast, a fall detection system observes the fall and generates a help notification to minimize the consequences of falls. A plethora of technical and review papers exist in the literature with a primary focus on fall detection. Similarly, several studies are relatively old, with a focus on wearables only, and use statistical and threshold-based approaches with a high false alarm rate. Therefore, this paper presents the latest research trends in fall detection and prevention systems using Machine Learning (ML) algorithms. It uses recent studies and analyzes datasets, age groups, ML algorithms, sensors, and location. Additionally, it provides a detailed discussion of the current trends of fall detection and prevention systems with possible future directions. This overview can help researchers understand the current systems and propose new methodologies by improving the highlighted issues. 10.3390/s21155134
An Interpretable Machine Learning Approach to Predict Fall Risk Among Community-Dwelling Older Adults: a Three-Year Longitudinal Study. Journal of general internal medicine BACKGROUND:Adverse health effects resulting from falls are a major public health concern. Although studies have identified risk factors for falls, none have examined long-term prediction of fall risk. Furthermore, recent evidence suggests that there are additional risk factors, such as psychosocial factors. OBJECTIVE:In this 3-year longitudinal study, we evaluated a predictive model for risk of fall among community-dwelling older adults using machine learning methods. DESIGN:A 3-year follow-up prospective longitudinal study (from 2010 to 2013). SETTING:Twenty-four municipalities in nine of the 47 prefectures (provinces) of Japan. PARTICIPANTS:Community-dwelling individuals aged ≥65 years who were functionally independent at baseline (n = 61,883). METHODS:The baseline survey was conducted from August 2010 to January 2012, and the follow-up survey was conducted from October to December 2013. Both surveys were conducted involving self-reported questionnaires. The measured outcome at the follow-up survey was self-reported multiple falls during the previous year. The 142 variables included in the baseline survey were regarded as candidate predictors. The random-forest-based Boruta algorithm was used to select predictors, and the eXtreme Gradient Boosting algorithm with 10 repetitions of nested k-fold cross-validation was used for modeling and model evaluation. Furthermore, we used shapley additive explanations to gain insight into the behavior of the prediction model. KEY RESULTS:Fourteen out of 142 candidate features were selected as predictors. Among these predictors, experience of falling as of the baseline survey was the most important feature, followed by self-rated health and age. Moreover, sense of coherence was newly identified as a risk factor for falls. CONCLUSIONS:This study suggests that machine learning tools can be adapted to explore new associative factors, make accurate predictions, and provide actionable insights for fall prevention strategies. 10.1007/s11606-022-07394-8
Symptom Prediction and Mortality Risk Calculation for COVID-19 Using Machine Learning. Jamshidi Elham,Asgary Amirhossein,Tavakoli Nader,Zali Alireza,Dastan Farzaneh,Daaee Amir,Badakhshan Mohammadtaghi,Esmaily Hadi,Jamaldini Seyed Hamid,Safari Saeid,Bastanhagh Ehsan,Maher Ali,Babajani Amirhesam,Mehrazi Maryam,Sendani Kashi Mohammad Ali,Jamshidi Masoud,Sendani Mohammad Hassan,Rahi Sahand Jamal,Mansouri Nahal Frontiers in artificial intelligence Early prediction of symptoms and mortality risks for COVID-19 patients would improve healthcare outcomes, allow for the appropriate distribution of healthcare resources, reduce healthcare costs, aid in vaccine prioritization and self-isolation strategies, and thus reduce the prevalence of the disease. Such publicly accessible prediction models are lacking, however. Based on a comprehensive evaluation of existing machine learning (ML) methods, we created two models based solely on the age, gender, and medical histories of 23,749 hospital-confirmed COVID-19 patients from February to September 2020: a symptom prediction model (SPM) and a mortality prediction model (MPM). The SPM predicts 12 symptom groups for each patient: respiratory distress, consciousness disorders, chest pain, paresis or paralysis, cough, fever or chill, gastrointestinal symptoms, sore throat, headache, vertigo, loss of smell or taste, and muscular pain or fatigue. The MPM predicts the death of COVID-19-positive individuals. The SPM yielded ROC-AUCs of 0.53-0.78 for symptoms. The most accurate prediction was for consciousness disorders at a sensitivity of 74% and a specificity of 70%. 2,440 deaths were observed in the study population. MPM had a ROC-AUC of 0.79 and could predict mortality with a sensitivity of 75% and a specificity of 70%. About 90% of deaths occurred in the top 21 percentile of risk groups. To allow patients and clinicians to use these models easily, we created a freely accessible online interface at www.aicovid.net. The ML models predict COVID-19-related symptoms and mortality using information that is readily available to patients as well as clinicians. Thus, both can rapidly estimate the severity of the disease, allowing shared and better healthcare decisions with regard to hospitalization, self-isolation strategy, and COVID-19 vaccine prioritization in the coming months. 10.3389/frai.2021.673527
Prediction of Chronic Obstructive Pulmonary Disease Exacerbation Events by Using Patient Self-reported Data in a Digital Health App: Statistical Evaluation and Machine Learning Approach. Chmiel Francis P,Burns Dan K,Pickering John Brian,Blythin Alison,Wilkinson Thomas Ma,Boniface Michael J JMIR medical informatics BACKGROUND:Self-reporting digital apps provide a way of remotely monitoring and managing patients with chronic conditions in the community. Leveraging the data collected by these apps in prognostic models could provide increased personalization of care and reduce the burden of care for people who live with chronic conditions. This study evaluated the predictive ability of prognostic models for the prediction of acute exacerbation events in people with chronic obstructive pulmonary disease by using data self-reported to a digital health app. OBJECTIVE:The aim of this study was to evaluate if data self-reported to a digital health app can be used to predict acute exacerbation events in the near future. METHODS:This is a retrospective study evaluating the use of symptom and chronic obstructive pulmonary disease assessment test data self-reported to a digital health app (myCOPD) in predicting acute exacerbation events. We include data from 2374 patients who made 68,139 self-reports. We evaluated the degree to which the different variables self-reported to the app are predictive of exacerbation events and developed both heuristic and machine learning models to predict whether the patient will report an exacerbation event within 3 days of self-reporting to the app. The model's predictive ability was evaluated based on self-reports from an independent set of patients. RESULTS:Users self-reported symptoms, and standard chronic obstructive pulmonary disease assessment tests displayed correlation with future exacerbation events. Both a baseline model (area under the receiver operating characteristic curve [AUROC] 0.655, 95% CI 0.689-0.676) and a machine learning model (AUROC 0.727, 95% CI 0.720-0.735) showed moderate ability in predicting exacerbation events, occurring within 3 days of a given self-report. Although the baseline model obtained a fixed sensitivity and specificity of 0.551 (95% CI 0.508-0.596) and 0.759 (95% CI 0.752-0.767) respectively, the sensitivity and specificity of the machine learning model can be tuned by dichotomizing the continuous predictions it provides with different thresholds. CONCLUSIONS:Data self-reported to health care apps designed to remotely monitor patients with chronic obstructive pulmonary disease can be used to predict acute exacerbation events with moderate performance. This could increase personalization of care by allowing preemptive action to be taken to mitigate the risk of future exacerbation events. 10.2196/26499
Prediction of post-stroke urinary tract infection risk in immobile patients using machine learning: an observational cohort study. Zhu C,Xu Z,Gu Y,Zheng S,Sun X,Cao J,Song B,Jin J,Liu Y,Wen X,Cheng S,Li J,Wu X The Journal of hospital infection BACKGROUND:Urinary tract infection (UTI) is one of major nosocomial infections significantly affecting the outcomes of immobile stroke patients. Previous studies have identified several risk factors, but it is still challenging to accurately estimate personal UTI risk. AIM:To develop predictive models for UTI risk identification for immobile stroke patients. METHODS:Research data were collected from our previous multicentre study. Derivation cohort included 3982 immobile stroke patients collected from November 1, 2015 to June 30, 2016; external validation cohort included 3837 patients collected from November 1, 2016 to July 30, 2017. Six machine learning models and an ensemble learning model were derived, based on 80% of derivation cohort, and effectiveness was evaluated with the remaining 20%. Shapley additive explanation values were used to determine feature importance and examine the clinical significance of prediction models. FINDINGS:In all, 2.59% (103/3982) patients were diagnosed with UTI in derivation cohort, 1.38% (53/3837) in external cohort. The ensemble learning model performed the best in area under the receiver operating characteristic (ROC) curve in internal validation (82.2%); second best in external validation (80.8%). In addition, the ensemble learning model performed the best sensitivity in both internal and external validation sets (80.9% and 81.1%, respectively). Seven UTI risk factors (pneumonia, glucocorticoid use, female sex, mixed cerebrovascular disease, increased age, prolonged length of stay, and duration of catheterization) were also identified. CONCLUSION:This ensemble learning model demonstrated promising performance. Future work should continue to develop a more concise scoring tool based on machine learning models and prospectively examining the model in practical use, thus improving clinical outcomes. 10.1016/j.jhin.2022.01.002
Machine learning-based risk prediction of intrahospital clinical outcomes in patients undergoing TAVI. Gomes Bruna,Pilz Maximilian,Reich Christoph,Leuschner Florian,Konstandin Mathias,Katus Hugo A,Meder Benjamin Clinical research in cardiology : official journal of the German Cardiac Society BACKGROUND:Currently, patient selection in TAVI is based upon a multidisciplinary heart team assessment of patient comorbidities and surgical risk stratification. In an era of increasing need for precision medicine and quickly expanding TAVI indications, machine learning has shown promise in making accurate predictions of clinical outcomes. This study aims to predict different intrahospital clinical outcomes in patients undergoing TAVI using a machine learning-based approach. The main clinical outcomes include all-cause mortality, stroke, major vascular complications, paravalvular leakage, and new pacemaker implantations. METHODS AND RESULTS:The dataset consists of 451 consecutive patients undergoing elective TAVI between February 2014 and June 2016. The applied machine learning methods were neural networks, support vector machines, and random forests. Their performance was evaluated using five-fold nested cross-validation. Considering all 83 features, the performance of all machine learning models in predicting all-cause intrahospital mortality (AUC 0.94-0.97) was significantly higher than both the STS risk score (AUC 0.64), the STS/ACC TAVR score (AUC 0.65), and all machine learning models using baseline characteristics only (AUC 0.72-0.82). Using an extreme boosting gradient, baseline troponin T was found to be the most important feature among all input variables. Overall, after feature selection, there was a slightly inferior performance. Stroke, major vascular complications, paravalvular leakage, and new pacemaker implantations could not be accurately predicted. CONCLUSIONS:Machine learning has the potential to improve patient selection and risk management of interventional cardiovascular procedures, as it is capable of making superior predictions compared to current logistic risk scores. 10.1007/s00392-020-01691-0
A machine learning approach for mortality prediction only using non-invasive parameters. Zhang Guang,Xu JiaMeng,Yu Ming,Yuan Jing,Chen Feng Medical & biological engineering & computing At present, the traditional scoring methods generally utilize laboratory measurements to predict mortality. It results in difficulties of early mortality prediction in the rural areas lack of professional laboratorians and medical laboratory equipment. To improve the efficiency, accuracy, and applicability of mortality prediction in the remote areas, a novel mortality prediction method based on machine learning algorithms is proposed, which only uses non-invasive parameters readily available from ordinary monitors and manual measurement. A new feature selection method based on the Bayes error rate is developed to select valuable features. Based on non-invasive parameters, four machine learning models were trained for early mortality prediction. The subjects contained in this study suffered from general critical diseases including but not limited to cancer, bone fracture, and diarrhea. Comparison tests among five traditional scoring methods and these four machine learning models with and without laboratory measurement variables are performed. Only using the non-invasive parameters, the LightGBM algorithms have an excellent performance with the largest accuracy of 0.797 and AUC of 0.879. There is no apparent difference between the mortality prediction performance with and without laboratory measurement variables for the four machine learning methods. After reducing the number of feature variables to no more than 50, the machine learning models still outperform the traditional scoring systems, with AUC higher than 0.83. The machine learning approaches only using non-invasive parameters achieved an excellent mortality prediction performance and can equal those using extra laboratory measurements, which can be applied in rural areas and remote battlefield for mortality risk evaluation. Graphical abstract. 10.1007/s11517-020-02174-0
Machine learning risk score for prediction of gestational diabetes in early pregnancy in Tianjin, China. Liu Hongwei,Li Jing,Leng Junhong,Wang Hui,Liu Jinnan,Li Weiqin,Liu Hongyan,Wang Shuo,Ma Jun,Chan Juliana Cn,Yu Zhijie,Hu Gang,Li Changping,Yang Xilin Diabetes/metabolism research and reviews AIMS:This study aimed to develop a machine learning-based prediction model for gestational diabetes mellitus (GDM) in early pregnancy in Chinese women. MATERIALS AND METHODS:We used an established population-based prospective cohort of 19,331 pregnant women registered as pregnant before the 15th gestational week in Tianjin, China, from October 2010 to August 2012. The dataset was randomly divided into a training set (70%) and a test set (30%). Risk factors collected at registration were examined and used to construct the prediction model in the training dataset. Machine learning, that is, the extreme gradient boosting (XGBoost) method, was employed to develop the model, while a traditional logistic model was also developed for comparison purposes. In the test dataset, the performance of the developed prediction model was assessed by calibration plots for calibration and area under the receiver operating characteristic curve (AUR) for discrimination. RESULTS:In total, 1484 (7.6%) women developed GDM. Pre-pregnancy body mass index, maternal age, fasting plasma glucose at registration, and alanine aminotransferase were selected as risk factors. The machine learning XGBoost model-predicted probability of GDM was similar to the observed probability in the test data set, while the logistic model tended to overestimate the risk at the highest risk level (Hosmer-Lemeshow test p value: 0.243 vs. 0.099). The XGBoost model achieved a higher AUR than the logistic model (0.742 vs. 0.663, p < 0.001). This XGBoost model was deployed through a free, publicly available software interface (https://liuhongwei.shinyapps.io/gdm_risk_calculator/). CONCLUSION:The XGBoost model achieved better performance than the logistic model. 10.1002/dmrr.3397
Application of machine learning approaches for osteoporosis risk prediction in postmenopausal women. Shim Jae-Geum,Kim Dong Woo,Ryu Kyoung-Ho,Cho Eun-Ah,Ahn Jin-Hee,Kim Jeong-In,Lee Sung Hyun Archives of osteoporosis Many predictive tools have been reported for assessing osteoporosis risk. The development and validation of osteoporosis risk prediction models were supported by machine learning. INTRODUCTION:Osteoporosis is a silent disease until it results in fragility fractures. However, early diagnosis of osteoporosis provides an opportunity to detect and prevent fractures. We aimed to develop machine learning approaches to achieve high predictive ability for osteoporosis risk that could help primary care providers identify which women are at increased risk of osteoporosis and should therefore undergo further testing with bone densitometry. METHODS:We included all postmenopausal Korean women from the Korea National Health and Nutrition Examination Surveys (KNHANES V-1, V-2) conducted in 2010 and 2011. Machine learning models using methods such as the k-nearest neighbors (KNN), decision tree (DT), random forest (RF), gradient boosting machine (GBM), support vector machine (SVM), artificial neural networks (ANN), and logistic regression (LR) were developed to predict osteoporosis risk. We analyzed the effect of applying the machine learning algorithms to the raw data and featuring the selected data only where the statistically significant variables were included as model inputs. The accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were used to evaluate performance among the seven models. RESULTS:A total of 1792 patients were included in this study, of which 613 had osteoporosis. The raw data consisted of 19 variables and achieved performances (in terms of AUROCs) of 0.712, 0.684, 0.727, 0.652, 0.724, 0.741, and 0.726 for KNN, DT, RF, GBM, SVM, ANN, and LR with fivefold cross-validation, respectively. The feature selected data consisted of nine variables and achieved performances (in terms of AUROCs) of 0.713, 0.685, 0.734, 0.728, 0.728, 0.743, and 0.727 for KNN, DT, RF, GBM, SVM, ANN, and LR with fivefold cross-validation, respectively. CONCLUSION:In this study, we developed and compared seven machine learning models to accurately predict osteoporosis risk. The ANN model performed best when compared to the other models, having the highest AUROC value. Applying the ANN model in the clinical environment could help primary care providers stratify osteoporosis patients and improve the prevention, detection, and early treatment of osteoporosis. 10.1007/s11657-020-00802-8
Machine learning for early prediction of in-hospital cardiac arrest in patients with acute coronary syndromes. Clinical cardiology BACKGROUND:Previous studies have used machine leaning to predict clinical deterioration to improve outcome prediction. However, no study has used machine learning to predict cardiac arrest in patients with acute coronary syndrome (ACS). Algorithms are required to generate high-performance models for predicting cardiac arrest in ACS patients with multivariate features. HYPOTHESIS:Machine learning algorithms will significantly improve outcome prediction of cardiac arrest in ACS patients. METHODS:This retrospective cohort study reviewed 166 ACS patients who had in-hospital cardiac arrest. Eight machine learning algorithms were trained using multivariate clinical features obtained 24 h prior to the onset of cardiac arrest. All machine learning models were compared to each other and to existing risk prediction scores (Global Registry of Acute Coronary Events, National Early Warning Score, and Modified Early Warning Score) using the area under the receiver operating characteristic curve (AUROC). RESULTS:The XGBoost model provided the best performance with regard to AUC (0.958 [95%CI: 0.938-0.978]), accuracy (88.9%), sensitivity (73%), negative predictive value (89%), and F1 score (80%) compared with other machine learning models. The K-nearest neighbor model generated the best specificity (99.3%) and positive predictive value (93.8%) metrics, but had low and unacceptable values for sensitivity and AUC. Most, but not all, machine learning models outperformed the existing risk prediction scores. CONCLUSIONS:The XGBoost model, which was generated based on a machine learning algorithm, has high potential to be used to predict cardiac arrest in ACS patients. This proposed model significantly improves outcome prediction compared to existing risk prediction scores. 10.1002/clc.23541
Role of artificial intelligence in cardiovascular risk prediction and outcomes: comparison of machine-learning and conventional statistical approaches for the analysis of carotid ultrasound features and intra-plaque neovascularization. Johri Amer M,Mantella Laura E,Jamthikar Ankush D,Saba Luca,Laird John R,Suri Jasjit S The international journal of cardiovascular imaging The aim of this study was to compare machine learning (ML) methods with conventional statistical methods to investigate the predictive ability of carotid plaque characteristics for assessing the risk of coronary artery disease (CAD) and cardiovascular (CV) events. Focused carotid B-mode ultrasound, contrast-enhanced ultrasound, and coronary angiography were performed on 459 participants. These participants were followed for 30 days. Plaque characteristics such as carotid intima-media thickness (cIMT), maximum plaque height (MPH), total plaque area (TPA), and intraplaque neovascularization (IPN) were measured at baseline. Two ML-based algorithms-random forest (RF) and random survival forest (RSF) were used for CAD and CV event prediction. The performance of these algorithms was compared against (i) univariate and multivariate analysis for CAD prediction using the area-under-the-curve (AUC) and (ii) Cox proportional hazard model for CV event prediction using the concordance index (c-index). There was a significant association between CAD and carotid plaque characteristics [cIMT (odds ratio (OR) = 1.49, p = 0.03), MPH (OR = 2.44, p < 0.0001), TPA (OR = 1.61, p < 0.0001), and IPN (OR = 2.78, p < 0.0001)]. IPN alone reported significant CV event prediction (hazard ratio = 1.24, p < 0.0001). CAD prediction using the RF algorithm reported an improvement in AUC by ~ 3% over the univariate analysis with IPN alone (0.97 vs. 0.94, p < 0.0001). Cardiovascular event prediction using RSF demonstrated an improvement in the c-index by ~ 17.8% over the Cox-based model (0.86 vs. 0.73). Carotid imaging phenotypes and IPN were associated with CAD and CV events. The ML-based system is superior to the conventional statistically-derived approaches for CAD prediction and survival analysis. 10.1007/s10554-021-02294-0
Risk prediction for delayed clearance of high-dose methotrexate in pediatric hematological malignancies by machine learning. Zhan Min,Chen Zebin,Ding Changcai,Qu Qiang,Wang Guoqiang,Liu Sixi,Wen Feiqiu International journal of hematology This study aimed to establish a predictive model to identify children with hematologic malignancy at high risk for delayed clearance of high-dose methotrexate (HD-MTX) based on machine learning. A total of 205 patients were recruited. Five variables (hematocrit, risk classification, dose, SLC19A1 rs2838958, sex) and three variables (SLC19A1 rs2838958, sex, dose) were statistically significant in univariable analysis and, separately, multivariate logistic regression. The data was randomly split into a "training cohort" and a "validation cohort". A nomogram for prediction of delayed HD-MTX clearance was constructed using the three variables in the training dataset and validated in the validation dataset. Five machine learning algorithms (cart classification and regression trees, naïve Bayes, support vector machine, random forest, C5.0 decision tree) combined with different resampling methods were used for model building with five or three variables. When developed machine learning models were evaluated in the validation dataset, the C5.0 decision tree combined with the synthetic minority oversampling technique (SMOTE) using five variables had the highest area under the receiver operating characteristic curve (AUC 0.807 [95% CI 0.724-0.889]), a better performance than the nomogram (AUC 0.69 [95% CI 0.594-0.787]). The results support potential clinical application of machine learning for patient risk classification. 10.1007/s12185-021-03184-w
Machine learning approaches for the prediction of postoperative complication risk in liver resection patients. Zeng Siyu,Li Lele,Hu Yanjie,Luo Li,Fang Yuanchen BMC medical informatics and decision making BACKGROUND:For liver cancer patients, the occurrence of postoperative complications increases the difficulty of perioperative nursing, prolongs the hospitalization time of patients, and leads to large increases in hospitalization costs. The ability to identify influencing factors and to predict the risk of complications in patients with liver cancer after surgery could assist doctors to make better clinical decisions. OBJECTIVE:The aim of the study was to develop a postoperative complication risk prediction model based on machine learning algorithms, which utilizes variables obtained before or during the liver cancer surgery, to predict when complications present with clinical symptoms and the ways of reducing the risk of complications. METHODS:The study subjects were liver cancer patients who had undergone liver resection. There were 175 individuals, and 13 variables were recorded. 70% of the data were used for the training set, and 30% for the test set. The performance of five machine learning models, logistic regression, decision trees-C5.0, decision trees-CART, support vector machines, and random forests, for predicting postoperative complication risk in liver resection patients were compared. The significant influencing factors were selected by combining results of multiple methods, based on which the prediction model of postoperative complications risk was created. The results were analyzed to give suggestions of how to reduce the risk of complications. RESULTS:Random Forest gave the best performance from the decision curves analysis. The decision tree-C5.0 algorithm had the best performance of the five machine learning algorithms if ACC and AUC were used as evaluation indicators, producing an area under the receiver operating characteristic curve value of 0.91 (95% CI 0.77-1), with an accuracy of 92.45% (95% CI 85-100%), the sensitivity of 87.5%, and specificity of 94.59%. The duration of operation, patient's BMI, and length of incision were significant influencing factors of postoperative complication risk in liver resection patients. CONCLUSIONS:To reduce the risk of complications, it appears to be important that the patient's BMI should be above 22.96 before the operation, and the duration of the operation should be minimized. 10.1186/s12911-021-01731-3
Machine learning for the prediction of acute kidney injury in critical care patients with acute cerebrovascular disease. Renal failure PURPOSE:Acute kidney injury (AKI) is a common complication and associated with a poor clinical outcome. In this study, we developed and validated a model for predicting the risk of AKI through machine learning methods in critical care patients with acute cerebrovascular disease. METHODS:This study was a retrospective study based on two different cohorts. Five machine learning methods were used to develop AKI risk prediction models. We used six popular metrics (AUROC, F2-Score, accuracy, sensitivity, specificity and precision) to evaluate the performance of these models. RESULTS:We identified 2935 patients in the MIMIC-III database and 499 patients in our local database to develop and validate the AKI risk prediction model. The incidence of AKI in these two different cohorts was 18.3% and 61.7%, respectively. Analysis showed that several laboratory parameters (serum creatinine, hemoglobin, white blood cell count, bicarbonate, blood urea nitrogen, sodium, albumin, and platelet count), age, and length of hospital stay, were the top ten important factors associated with AKI. The analysis demonstrated that the XGBoost had higher AUROC (0.880, 95%CI: 0.831-0.929), indicating that the XGBoost model was better at predicting AKI risk in patients with acute cerebrovascular disease than other models. CONCLUSIONS:This study developed machine learning methods to identify critically ill patients with acute cerebrovascular disease who are at a high risk of developing AKI. This result suggested that machine learning techniques had the potential to improve the prediction of AKI risk models in critical care. 10.1080/0886022X.2022.2036619
Machine Learning-based Prediction Models for Diagnosis and Prognosis in Inflammatory Bowel Diseases: A Systematic Review. Journal of Crohn's & colitis BACKGROUND AND AIMS:There is increasing interest in machine learning-based prediction models in inflammatory bowel diseases [IBD]. We synthesised and critically appraised studies comparing machine learning vs traditional statistical models, using routinely available clinical data for risk prediction in IBD. METHODS:Through a systematic review till January 1, 2021, we identified cohort studies that derived and/or validated machine learning models, based on routinely collected clinical data in patients with IBD, to predict the risk of harbouring or developing adverse clinical outcomes, and reported its predictive performance against a traditional statistical model for the same outcome. We appraised the risk of bias in these studies using the Prediction model Risk of Bias ASsessment [PROBAST] tool. RESULTS:We included 13 studies on machine learning-based prediction models in IBD, encompassing themes of predicting treatment response to biologics and thiopurines and predicting longitudinal disease activity and complications and outcomes in patients with acute severe ulcerative colitis. The most common machine learning models used were tree-based algorithms, which are classification approaches achieved through supervised learning. Machine learning models outperformed traditional statistical models in risk prediction. However, most models were at high risk of bias, and only one was externally validated. CONCLUSIONS:Machine learning-based prediction models based on routinely collected data generally perform better than traditional statistical models in risk prediction in IBD, though frequently have high risk of bias. Future studies examining these approaches are warranted, with special focus on external validation and clinical applicability. 10.1093/ecco-jcc/jjab155