Risk factors and the utility of three different kinds of prediction models for postoperative fatigue after gastrointestinal tumor surgery.
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
BACKGROUND:Postoperative fatigue (POF) is a common complication after gastrointestinal tumor surgery, and it also brings negative effect on prognosis and life quality. However, there are no prediction models for POF, and studies of risk factors are not comprehensive. Therefore, the aim of this study is to investigate the risk factors and pick out the best prediction model for POF and to validate it. METHODS:A prospective study was conducted for patients undergoing elective gastrointestinal tumor surgery. Physiological, psychological, and socioeconomic factors were collected. Logistic regression, back-propagation artificial neural networks (BP-ANNs), and classification and regression tree (CART) were constructed and compared. RESULTS:A total of 598 patients consisting of 463 derivation sample and 135 validation sample were included. The incidence of POF in derivation sample, validation sample, and total were 58.3%, 57.0%, and 58.7%, respectively. Logistic regression results showed age, higher degree of education, lower personal monthly income, advanced cancer, hypoproteinemia, preoperative anxiety or depression, and limited social support were risk factors for POF. Receiver operating characteristic curve (ROC) and performance indices were used to test three models. BP-ANN was the best by the comparison of models, and its strong predictive performance was also validated. CONCLUSIONS:More attention should be paid on specific patients after gastrointestinal tumor surgery. BP-ANN is a powerful mathematical tool that could predict POF exactly. It may be used as a noninvasive screening tool to guide clinicians for early identification of high-risk patients and grading interventions.
Developing a visual model for predicting depression in patients with lung cancer.
Journal of clinical nursing
AIMS AND OBJECTIVES:To investigate and analyse the prevalence of depression among patients with lung cancer, identify risk factors of depression, and develop a visual, non-invasive, and straightforward clinical prediction model that can be used to predict the risk probability of depression in patients with lung cancer quantitatively. BACKGROUND:Depression is one of the common concomitant symptoms of patients with lung cancer, which can increase the risk of suicide. However, the current assessment tools cannot combine multiple risk factors to predict the risk probability of depression in patients. DESIGN:A cross-sectional study. METHODS:The clinical data from 297 patients with lung cancer in China were collected and analysed in this cross-sectional study. The clinical prediction model was constructed according to the results of the Chi-square test and the logistic regression analysis, evaluated by discrimination, calibration, and decision curve analysis, and visualised by a nomogram. This study was reported using the TRIPOD checklist. RESULTS:130 patients with lung cancer had depressive symptoms with a prevalence of 43.77%. A visual prediction model was constructed based on age, disease duration, exercise, stigma, and resilience. This model showed good discrimination at an AUC of 0.842. Calibration curve analysis indicated a good agreement between experimental and predicted values, and the decision curve analysis showed a high clinical utility. CONCLUSIONS:The visual prediction model developed in this study has excellent performance, which can accurately predict the occurrence of depression in patients with lung cancer at an early stage and assist the medical staff in taking targeted preventative measures. RELEVANCE TO CLINICAL PRACTICE:The visual, non-invasive, and simple nomogram can help clinical medical staff to calculate the risk probability of depression among patients with lung cancer, formulate personalised preventive care measures for high-risk groups as soon as possible, and improve the quality of life of patients.
Prediction of Persistent Pain Severity and Impact 12 Months After Breast Surgery Using Comprehensive Preoperative Assessment of Biopsychosocial Pain Modulators.
Schreiber Kristin L,Zinboonyahgoon Nantthansorn,Flowers K Mikayla,Hruschak Valerie,Fields Kara G,Patton Megan E,Schwartz Emily,Azizoddin Desiree,Soens Mieke,King Tari,Partridge Ann,Pusic Andrea,Golshan Mehra,Edwards Rob R
Annals of surgical oncology
BACKGROUND:Persistent post-mastectomy pain (PPMP) is a significant negative outcome occurring after breast surgery, and understanding which individual women are most at risk is essential to targeting of preventive efforts. The biopsychosocial model of pain suggests that factors from many domains may importantly modulate pain processing and predict the progression to pain persistence. METHODS:This prospective longitudinal observational cohort study used detailed and comprehensive psychosocial and psychophysical assessment to characterize individual pain-processing phenotypes in 259 women preoperatively. Pain severity and functional impact then were longitudinally assessed using both validated surgery-specific and general pain questionnaires to survey patients who underwent lumpectomy, mastectomy, or mastectomy with reconstruction in the first postsurgical year. An agnostic, multivariable modeling strategy identified consistent predictors of several pain outcomes at 12 months. RESULTS:The preoperative characteristics most consistently associated with PPMP outcomes were preexisting surgical area pain, less education, increased somatization, and baseline sleep disturbance, with axillary dissection emerging as the only consistent surgical variable to predict worse pain. Greater pain catastrophizing, negative affect, younger age, higher body mass index (BMI), and chemotherapy also were independently predictive of pain impact, but not severity. Sensory disturbance in the surgical area was predicted by a slightly different subset of factors, including higher preoperative temporal summation of pain. CONCLUSIONS:This comprehensive approach assessing consistent predictors of pain severity, functional impact, and sensory disturbance may inform personalized prevention of PPMP and also may allow stratification and enrichment in future preventive studies of women at higher risk of this outcome, including pharmacologic and behavioral interventions and regional anesthesia.
Predictive and Preventive Potential of Preoperative Gut Microbiota in Chronic Postoperative Pain in Breast Cancer Survivors.
Anesthesia and analgesia
BACKGROUND:Evidence suggests a potential relationship between gut microbiota and chronic postoperative pain (CPP). This study aimed to explore the predictive and preventive potential of preoperative gut microbiota in CPP in breast cancer survivors. METHODS:In the clinical experiments, we designed a nested case-control study to compared preoperative gut microbiota of breast cancer survivors with and without CPP using 16s rRNA sequencing. The primary outcome was clinically meaningful pain in or around the operative area 3 months after surgery. Logistic prediction models based on previously identified risk factors for CPP in breast cancer survivors were tested with and without differential bacteria to evaluate the model's potential for improvement with the addition of gut microbiota information. In the animal experiments, preoperative fecal microbiota was transplanted from patients with and without CPP to mice, and a spared nerve injury (SNI) model was used to mimic neuropathic pain in CPP. Mechanical hyperalgesia and the expression of markers of spinal microglia and peroxisome proliferator-activated receptor-γ (PPAR-γ) were assessed. RESULTS:Sixty-six CPP patients and 66 matched controls were analyzed. Preoperative gut microbiota composition was significantly different in the 2 groups at phylus, family, and genera levels. The discrimination of the clinical prediction model (determined by area under the receiver operating characteristic curve) improved by 0.039 and 0.099 after the involvement of differential gut microbiota at the family and genus levels, respectively. After fecal microbiota transplantation (FMT), "CPP microbiota" recipient mice exhibited significantly increased mechanical hyperalgesia and decreased expression of Ppar-γ and arginase-1 (Arg-1) in the spinal cord. CONCLUSIONS:Preoperative gut microbiota has the potential to predict and prevent the development of CPP and plays a causal role in its development via the PPAR-γ-microglia pathway in the spinal cord. Thus, it could be targeted to develop a prevention strategy for CPP in breast cancer survivors.
Cost-Sensitive KNN Algorithm for Cancer Prediction Based on Entropy Analysis.
Song Chaohong,Li Xinran
Entropy (Basel, Switzerland)
Early diagnosis of cancer is beneficial in the formulation of the best treatment plan; it can improve the survival rate and the quality of patient life. However, imaging detection and needle biopsy usually used not only find it difficult to effectively diagnose tumors at early stage, but also do great harm to the human body. Since the changes in a patient's health status will cause changes in blood protein indexes, if cancer can be diagnosed by the changes in blood indexes in the early stage of cancer, it can not only conveniently track and detect the treatment process of cancer, but can also reduce the pain of patients and reduce the costs. In this paper, 39 serum protein markers were taken as research objects. The difference of the entropies of serum protein marker sequences in different types of patients was analyzed, and based on this, a cost-sensitive analysis model was established for the purpose of improving the accuracy of cancer recognition. The results showed that there were significant differences in entropy of different cancer patients, and the complexity of serum protein markers in normal people was higher than that in cancer patients. Although the dataset was rather imbalanced, containing 897 instances, including 799 normal instances, 44 liver cancer instances, and 54 ovarian cancer instances, the accuracy of our model still reached 95.21%. Other evaluation indicators were also stable and satisfactory; precision, recall, F1 and AUC reach 0.807, 0.833, 0.819 and 0.92, respectively. This study has certain theoretical and practical significance for cancer prediction and clinical application and can also provide a research basis for the intelligent medical treatment.
Development and internal validation of prediction models for colorectal cancer survivors to estimate the 1-year risk of low health-related quality of life in multiple domains.
Révész Dóra,van Kuijk Sander M J,Mols Floortje,van Duijnhoven Fränzel J B,Winkels Renate M,Hoofs Huub,Kant I Jmert,Smits Luc J,Breukink Stéphanie O,van de Poll-Franse Lonneke V,Kampman Ellen,Beijer Sandra,Weijenberg Matty P,Bours Martijn J L
BMC medical informatics and decision making
BACKGROUND:Many colorectal cancer (CRC) survivors experience persisting health problems post-treatment that compromise their health-related quality of life (HRQoL). Prediction models are useful tools for identifying survivors at risk of low HRQoL in the future and for taking preventive action. Therefore, we developed prediction models for CRC survivors to estimate the 1-year risk of low HRQoL in multiple domains. METHODS:In 1458 CRC survivors, seven HRQoL domains (EORTC QLQ-C30: global QoL; cognitive, emotional, physical, role, social functioning; fatigue) were measured prospectively at study baseline and 1 year later. For each HRQoL domain, scores at 1-year follow-up were dichotomized into low versus normal/high. Separate multivariable logistic prediction models including biopsychosocial predictors measured at baseline were developed for the seven HRQoL domains, and internally validated using bootstrapping. RESULTS:Average time since diagnosis was 5 years at study baseline. Prediction models included both non-modifiable predictors (age, sex, socio-economic status, time since diagnosis, tumor stage, chemotherapy, radiotherapy, stoma, micturition, chemotherapy-related, stoma-related and gastrointestinal complaints, comorbidities, social inhibition/negative affectivity, and working status) and modifiable predictors (body mass index, physical activity, smoking, meat consumption, anxiety/depression, pain, and baseline fatigue and HRQoL scores). Internally validated models showed good calibration and discrimination (AUCs: 0.83-0.93). CONCLUSIONS:The prediction models performed well for estimating 1-year risk of low HRQoL in seven domains. External validation is needed before models can be applied in practice.
Identifying predictive factors for neuropathic pain after breast cancer surgery using machine learning.
Juwara Lamin,Arora Navpreet,Gornitsky Mervyn,Saha-Chaudhuri Paramita,Velly Ana M
International journal of medical informatics
INTRODUCTION:Neuropathic pain (NP) remains a major debilitating condition affecting more than 26% of breast cancer survivors worldwide. NP is diagnosed using a validated 10-items Douleur Neuropathique - 4 screening questionnaire which is administered 3 months after surgery and requires patient-doctor interaction. To develop an effective prognosis model admissible soon after surgery, without the need for patient-doctor interaction, we sought to  identify specific pain characteristics that can help determine which patients may be susceptible to NP after BC surgery, and 2) assess the utility of machine learning models developed in objective  as a knowledge discovery tool for downstream analysis. METHODS:The dataset is from a prospective cohort study of female patients scheduled to undergo breast cancer surgery for the first time at the Jewish General Hospital, Montreal, Canada between November 2014 and March 2019. NP was assessed at 3 months after surgery using Douleur Neuropathique - 4 interview scores (in short, DN4-interview; range: 0-7). For the primary analysis, we constructed six ML algorithms (least square, ridge, elastic net, random forest, gradient boosting, and neural net) to identify the most relevant predictors for DN4-interview score; and compared model performance based on root mean square error (RMSE). For the secondary analysis, we built a logistic classification model for neuropathic pain (DN4-interview score ≥ 3 versus DN4-interview score < 3) using the relevant-consensus-predictors from the primary analysis. RESULTS:Anxiety, type of surgery, preoperative baseline pain and acute pain on movement were identified as the most relevant predictors for DN4 - interview score. The least square regression model (RMSE = 1.43) is comparable in performance with random forest (RMSE = 1.39) and neural network model (RMSE = 1.50). The Gradient boosting model (RMSE = 1.16) outperformed the models compared including the penalized regression models (ridge regressions, RMSE = 1.28; and elastic net, RMSE = 1.31). In the secondary analysis, the preferred logistic regression classier for NP had an area under the curve (AUC) of 0.68 (95% CI = 0.57 to 0.79). Anxiety was significantly associated with the likelihood of NP (odds ratio = 2.18; 95% CI = 1.05-4.49). In comparison to their counterparts, the odds of NP were higher in participants with acute pain on movement or with present preoperative baseline pain or participants who performed total mastectomy surgery, but the differences were not statistically significant. CONCLUSIONS:Modern machine learning models show improvements over traditional least square regression in predicting of DN4-interview score. Penalized regression methods and the Gradient boosting model out-perform other models. As a predictor discovery tool, machine learning algorithms identify relevant predictors for DN4-interview score that remain statistically significant indicators of neuropathic pain in the classification model. Anxiety, type of surgery and acute pain on movement remain the most useful predictors for neuropathic pain.
Statistical validation of the relationships of cancer pain relief with various factors using ordered logistic regression analysis.
Kanbayashi Yuko,Okamoto Kousuke,Ogaru Takanori,Hosokawa Toyoshi,Takagi Tatsuya
The Clinical journal of pain
OBJECTIVE:To clarify the relationships of cancer pain with various factors that prevent pain control statistically. METHODS:The participants were 71 terminal cancer patients admitted to the Department of Hematology/Oncology or Department of Gastroenterology/Hepatology, University Hospital, Kyoto Prefectural University of Medicine in whose pain control a pharmacist was involved as part of her clinical duties from January 2004 to November 2006. The effectiveness of pain control was evaluated using a 5-point verbal rating scale (0=excellent, 1=good, 2=moderate, 3=poor, and 4=very poor) by interviewing the patients. As pain was rated using a graded scale and as many factors were involved in pain, analysis was performed using ordered logistic regression analysis. Moreover, prediction of an optimal model was performed by leave-one-out cross-validation to eliminate unnecessary variables. A program to perform leave-one-out cross-validation by ordered logistic regression analysis was prepared, independent variables used in the model were increased one by one, and calculation was performed in all combinations. Then, the optimal model was predicted by calculating the percent accuracy of predictions and Spearman rank correlation coefficient. RESULTS:Nausea [odds ratio (OR)=1.948, P=0.0232], sex (OR=2.322, P=0.0030), and bone metastasis (OR=2.367, P=0.0017) remained as variables significantly correlated with pain when the number of independent variables was 5, and sex (OR=2.167, P=0.006) and bone metastasis (OR=2.093, P=0.005) remained when the number of variables was 6. DISCUSSION:The statistical identification of factors preventing pain control is considered to contribute to the establishment of an evidence-based approach to cancer pain relief.
A decision tree prediction model for a short-term outcome of delirium in patients with advanced cancer receiving pharmacological interventions: A secondary analysis of a multicenter and prospective observational study (Phase-R).
Palliative & supportive care
OBJECTIVE:There is no widely used prognostic model for delirium in patients with advanced cancer. The present study aimed to develop a decision tree prediction model for a short-term outcome. METHOD:This is a secondary analysis of a multicenter and prospective observational study conducted at 9 psycho-oncology consultation services and 14 inpatient palliative care units in Japan. We used records of patients with advanced cancer receiving pharmacological interventions with a baseline Delirium Rating Scale Revised-98 (DRS-R98) severity score of ≥10. A DRS-R98 severity score of <10 on day 3 was defined as the study outcome. The dataset was randomly split into the training and test dataset. A decision tree model was developed using the training dataset and potential predictors. The area under the curve (AUC) of the receiver operating characteristic curve was measured both in 5-fold cross-validation and in the independent test dataset. Finally, the model was visualized using the whole dataset. RESULTS:Altogether, 668 records were included, of which 141 had a DRS-R98 severity score of <10 on day 3. The model achieved an average AUC of 0.698 in 5-fold cross-validation and 0.718 (95% confidence interval, 0.627-0.810) in the test dataset. The baseline DRS-R98 severity score (cutoff of 15), hypoxia, and dehydration were the important predictors, in this order. SIGNIFICANCE OF RESULTS:We developed an easy-to-use prediction model for the short-term outcome of delirium in patients with advanced cancer receiving pharmacological interventions. The baseline severity of delirium and precipitating factors of delirium were important for prediction.
Predictors of Pain Palliation After Radiation Therapy for Painful Tumors: A Prospective Observational Study.
Saito Tetsuo,Toya Ryo,Tomitaka Etsushi,Matsuyama Tomohiko,Ninomura Satoshi,Oya Natsuo
International journal of radiation oncology, biology, physics
PURPOSE:Although radiation therapy (RT) is an important part of treatment for cancer pain, prediction of the patient's pain response remains difficult. We evaluated the characteristics of patients, their tumors, and their pain to identify the predictors of pain palliation after RT for painful tumors. METHODS:Our 3-center prospective observational study included patients scheduled for palliative or curative RT for painful tumors. Brief Pain Inventory data were collected at the start of RT and 1, 2, and 3 months thereafter. The pain response was assessed using the International Consensus Endpoint. The Mann-Whitney U-test was used to compare responders and nonresponders based on changes in the BPI scores. Predictors of the pain response were evaluated using the Fine-Gray model, in which death without a pain response was recorded as a competing risk. The independent variables were 11 a priori selected potential predictors with clinical relevance. RESULTS:Of 302 analyzable patients, 262 (87%) had solid and 40 (13%) had hematologic tumors. The median total radiation dose was 30 Gy (range, 6-70.4 Gy). The pain response rate was 52% for 264 (87%) evaluable patients at 1-, 57% for 228 (75%) such patients at 2-, and 58% for 182 (60%) evaluable patients at 3-month follow-up. At 2-month follow-up, responders experienced a greater decrease in all 7 pain interference subscales of the Brief Pain Inventory compared to nonresponders. Multivariable analysis demonstrated that hematologic tumors (hazard ratio [HR], 1.85; 95% confidence interval [CI], 1.15-2.98), a neuropathic component of the index pain (HR, 1.50; 95% CI, 1.05-2.14), and opioid analgesic use before RT (HR, 0.65; 95% CI, 0.47-0.91) were independent significant predictors of pain response. CONCLUSIONS:Patients with hematologic tumors, a neuropathic component of the index pain, and no treatment with opioid analgesics before RT were more likely to experience pain palliation after RT.
Predicting anxiety in cancer survivors presenting to primary care - A machine learning approach accounting for physical comorbidity.
Haun Markus W,Simon Laura,Sklenarova Halina,Zimmermann-Schlegel Verena,Friederich Hans-Christoph,Hartmann Mechthild
BACKGROUND:The purpose of this study was to explore predictors for anxiety as the most common form of psychological distress in cancer survivors while accounting for physical comorbidity. METHODS:We conducted a secondary data analysis of a large study within the German National Cancer Plan which enrolled primary care cancer survivors diagnosed with colon, prostatic, or breast cancer. We selected candidate predictors based on a systematic MEDLINE search. Using supervised machine learning, we developed a prediction model for anxiety by splitting the data into a 70% training set and a 30% test set and further split the training set into 10-folds for cross-validating the hyperparameter tuning step during model selection. We fit six different regression models, selected the model that maximized the root mean square error (RMSE) and fit the selected model to the entire training set. Finally, we evaluated the model performance on the holdout test set. RESULTS:In total, data from 496 cancer survivors were analyzed. The LASSO model (α = 1.0) with weakly penalized model complexity (λ = 0.015) slightly outperformed all other models (RMSE = 0.370). Physical symptoms, namely, fatigue/weakness (β = 0.18), insomnia (β = 0.12), and pain (β = 0.04), were the most important predictors, while the degree of physical comorbidity was negligible. CONCLUSIONS:Prediction of clinically significant anxiety in cancer survivors using readily available predictors is feasible. The findings highlight the need for considering cancer survivors' physical functioning regardless of the degree of comorbidity when assessing their psychological well-being. The generalizability of the model to other populations should be investigated in future external validations.
Contributions of risk and protective factors to prediction of psychological symptoms after traumatic experiences.
Carlson Eve B,Palmieri Patrick A,Field Nigel P,Dalenberg Constance J,Macia Kathryn S,Spain David A
OBJECTIVE:Traumatic experiences cause considerable suffering and place a burden on society due to lost productivity, increases in suicidality, violence, criminal behavior, and psychological disorder. The impact of traumatic experiences is complicated because many factors affect individuals' responses. By employing several methodological improvements, we sought to identify risk factors that would account for a greater proportion of variance in later disorder than prior studies. METHOD:In a sample of 129 traumatically injured hospital patients and family members of injured patients, we studied pre-trauma, time of trauma, and post-trauma psychosocial risk and protective factors hypothesized to influence responses to traumatic experiences and posttraumatic (PT) symptoms (including symptoms of PTSD, depression, negative thinking, and dissociation) two months after trauma. RESULTS:The risk factors were all significantly correlated with later PT symptoms, with post-trauma life stress, post-trauma social support, and acute stress symptoms showing the strongest relationships. A hierarchical regression, in which the risk factors were entered in 6 steps based on their occurrence in time, showed the risks accounted for 72% of the variance in later symptoms. Most of the variance in PT symptoms was shared among many risk factors, and pre-trauma and post-trauma risk factors accounted for the most variance. CONCLUSIONS:Collectively, the risk factors accounted for more variance in later PT symptoms than in previous studies. These risk factors may identify individuals at risk for PT psychological disorders and targets for treatment.
A Machine-Learning-Algorithm-Based Prediction Model for Psychotic Symptoms in Patients with Depressive Disorder.
Journal of personalized medicine
Psychotic symptoms are rarely concurrent with the clinical manifestations of depression. Additionally, whether psychotic major depression is a subtype of major depression or a clinical syndrome distinct from non-psychotic major depression remains controversial. Using data from the Research on Asian Psychotropic Prescription Patterns for Antidepressants, we developed a machine-learning-algorithm-based prediction model for concurrent psychotic symptoms in patients with depressive disorders. The advantages of machine learning algorithms include the easy identification of trends and patterns, handling of multi-dimensional and multi-faceted data, and wide application. Among 1171 patients with depressive disorders, those with psychotic symptoms were characterized by significantly higher rates of depressed mood, loss of interest and enjoyment, reduced energy and diminished activity, reduced self-esteem and self-confidence, ideas of guilt and unworthiness, psychomotor agitation or retardation, disturbed sleep, diminished appetite, and greater proportions of moderate and severe degrees of depression compared to patients without psychotic symptoms. The area under the curve was 0.823. The overall accuracy was 0.931 (95% confidence interval: 0.897-0.956). Severe depression (degree of depression) was the most important variable in the prediction model, followed by diminished appetite, subthreshold (degree of depression), ideas or acts of self-harm or suicide, outpatient status, age, psychomotor retardation or agitation, and others. In conclusion, the machine-learning-based model predicted concurrent psychotic symptoms in patients with major depression in connection with the "severity psychosis" hypothesis.
Assessment of Changes in Symptoms Is Feasible and Prognostic in the Last Weeks of Life: An International Multicenter Cohort Study.
Suh Sang-Yeon,Won Seon-Hye,Hiratsuka Yusuke,Choi Sung-Eun,Cheng Shao-Yi,Mori Masanori,Chen Ping-Jen,Yamaguchi Takashi,Morita Tatsuya,Tsuneto Satoru,LeBlanc Thomas W,Kim Sun-Hyun,Yoon Seok-Joon,Lee Eon Sook,Hwang Sun Wook
Journal of palliative medicine
Symptoms are not typically part of established various prognostic factors and scoring systems but are among the most frequently assessed issues in patient care. To evaluate that, changes in symptoms can provide additional useful prognostic information. A secondary analysis of an international cohort study in Japan, Korea, and Taiwan. Subjects were adult patients with advanced cancer ( = 2074) who were admitted to 37 palliative care units (PCUs) in 3 countries from January 2017 to September 2018. Symptoms (dyspnea, fatigue, dry mouth, and drowsiness) were assessed at admission and one-week later. Dyspnea was assessed by the presence of resting and exertional dyspnea, whereas other symptoms were assessed using the Integrated Palliative care Outcome Scales (IPOS) (range 0-4). For analysis, we grouped patients by symptom change, as either Improved, Stable, or Worsened (by having at least a one increment decrease, no change, or at least a one increment increase, respectively). Worsened groups had the shortest survival (median survival 15-21 days) compared with those with Improved (median survival 23-31 days) and Stable symptoms (median survival 27-29 days) across all four symptoms (dyspnea, fatigue, dry mouth, and drowsiness). Survival differences were statistically significantly different across all three groups for all symptoms (all < 0.001). Interestingly, Improved symptoms were associated with similar survival compared with Stable groups, with no statistical differences. Worsened symptoms at one week after admission were useful predictors of survival for patients with advanced cancer in PCUs during the final weeks of life. Longitudinal assessments are needed to reflect passage of time as well as impact of treatments.
Development and preliminary validation of a risk prediction model for chemotherapy-related nausea and vomiting.
Molassiotis A,Stamataki Z,Kontopantelis E
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
BACKGROUND:A number of risk factors have been implicated in the development of chemotherapy-induced nausea/vomiting (CINV). Our aim was to develop a risk prediction model and identify patients at high risk for developing CINV before their chemotherapy treatment. PATIENTS AND METHODS:A multisite, observational, prospective longitudinal design was used. Participants were 336 chemotherapy-naïve cancer patients providing 791 assessments. They completed measures to assess potential risk factors for CINV, including socio-demographic and clinical/treatment-related characteristics, symptom distress, expectations for CINV and state-trait anxiety. CINV was measured with the MASCC Antiemesis Tool. Participants were divided randomly to a training set (=286) and a test set (=50). Random-effects models were run to ascertain the contribution of risk factors in the development of CINV using the training sample. Specificity and sensitivity of the model were assessed in both sets of samples. RESULTS:Younger age, history of nausea/vomiting, trait anxiety and fatigue were linked with higher levels of CINV, and use of moderately and low emetogenic chemotherapy were linked with lower CINV. The model's specificity were 55.4 and 50.0 % and sensitivity were 80.3 and 79.0 % in the training and test sample, respectively. A dynamic web-based tool is freely available for use by clinicians. CONCLUSION:This model of risk prediction for CINV can be an aid to clinical decision-making and assist clinicians to rationalise antiemetic use with their patients.
Approach to evaluation of fever in ambulatory cancer patients receiving chemotherapy: A systematic review.
Krzyzanowska M K,Walker-Dilks C,Morris A M,Gupta R,Halligan R,Kouroukis C T,McCann K,Atzema C L
Cancer treatment reviews
PURPOSE:To define the optimal model of care for patients receiving outpatient chemotherapy who experience a fever. Fever is a common symptom in patients receiving chemotherapy, but the approach to evaluation of fever is not standardized. METHODS:We conducted a search for existing guidelines and a systematic review of the primary literature from database inception to November 2015. Full-text reports and conference abstracts were considered for inclusion. The search focused on the following topics: the relationship between temperature and poor outcome; predictors for the development of febrile neutropenia (FN); the timing, location, and personnel involved in fever assessment; and the provision of information to patients receiving chemotherapy. RESULTS:Eight guidelines and 38 studies were included. None of the guidelines were directly relevant to the target population because they dealt primarily with the management of FN after diagnosis. The primary studies tended to include fever as one of many symptoms assessed in the setting of chemotherapy. Temperature level was a weak predictor of poor outcomes. We did not find validated prediction models for identifying patients at risk of FN among patients receiving chemotherapy. Several studies presented approaches to symptom management that included fever among the symptoms, but results were not mature enough to merit widespread adoption. CONCLUSION:Despite the frequency and risks of fever in the setting of chemotherapy, there is limited evidence to define who needs urgent assessment, where the assessment should be performed, and how quickly. Future research in this area is greatly needed to inform new models of care.
Early symptoms and sensations as predictors of lung cancer: a machine learning multivariate model.
Levitsky Adrian,Pernemalm Maria,Bernhardson Britt-Marie,Forshed Jenny,Kölbeck Karl,Olin Maria,Henriksson Roger,Lehtiö Janne,Tishelman Carol,Eriksson Lars E
The aim of this study was to identify a combination of early predictive symptoms/sensations attributable to primary lung cancer (LC). An interactive e-questionnaire comprised of pre-diagnostic descriptors of first symptoms/sensations was administered to patients referred for suspected LC. Respondents were included in the present analysis only if they later received a primary LC diagnosis or had no cancer; and inclusion of each descriptor required ≥4 observations. Fully-completed data from 506/670 individuals later diagnosed with primary LC (n = 311) or no cancer (n = 195) were modelled with orthogonal projections to latent structures (OPLS). After analysing 145/285 descriptors, meeting inclusion criteria, through randomised seven-fold cross-validation (six-fold training set: n = 433; test set: n = 73), 63 provided best LC prediction. The most-significant LC-positive descriptors included a cough that varied over the day, back pain/aches/discomfort, early satiety, appetite loss, and having less strength. Upon combining the descriptors with the background variables current smoking, a cold/flu or pneumonia within the past two years, female sex, older age, a history of COPD (positive LC-association); antibiotics within the past two years, and a history of pneumonia (negative LC-association); the resulting 70-variable model had accurate cross-validated test set performance: area under the ROC curve = 0.767 (descriptors only: 0.736/background predictors only: 0.652), sensitivity = 84.8% (73.9/76.1%, respectively), specificity = 55.6% (66.7/51.9%, respectively). In conclusion, accurate prediction of LC was found through 63 early symptoms/sensations and seven background factors. Further research and precision in this model may lead to a tool for referral and LC diagnostic decision-making.
Prospective validation of a prediction tool for identifying patients at high risk for chemotherapy-induced nausea and vomiting.
Dranitsaris George,Bouganim Nathaniel,Milano Carolyn,Vandermeer Lisa,Dent Susan,Wheatley-Price Paul,Laporte Jenny,Oxborough Karen-Ann,Clemons Mark
The journal of supportive oncology
BACKGROUND:Even with modern antiemetic regimens, up to 20% of cancer patients suffer from moderate to severe chemotherapy-induced nausea and vomiting (CINV) (> or = grade 2). We previously developed chemotherapy cycle-based risk predictive models for > or = grade 2 acute and delayed CINV. In this study, the prospective validation of the prediction models and associated scoring systems is described. OBJECTIVE:Our objective was to prospectively validate prediction models designed to identify patients at high risk for moderate to severe CINV. METHODS:Patients receiving chemotherapy were provided with CINV symptom diaries. Prior to each cycle of chemotherapy, the acute and delayed CINV scoring systems were used to stratify patients into low- and high-risk groups. Logistic regression was used to compare the occurrence of > or = grade 2 CINV between patients considered by the model to be at high vs low risk. The external validity of each system was assessed via an area under the receiver operating characteristic (AUROC) curve analysis. RESULTS:Outcome data were collected from 97 patients following 401 cycles of chemotherapy. The incidence of > or =grade 2 acute and delayed CINV was 13.5% and 21.4%, respectively. There was a significant correlation between the risk score and the probability of developing acute and delayed CINV following chemotherapy. Both the acute and delayed scoring systems had good predictive accuracy when applied to the validation sample (acute, AUROC = 0.70, 95% CI, 0.62-0.77; delayed, AUROC = 0.75, 95% CI, 0.69-0.80). Patients who were identified as high risk were 3.1 (P = .006) and 4.2 (P< .001) times more likely to develop - grade 2 acute and delayed CINV than were those identified as low risk. CONCLUSION:This study demonstrates that the scoring systems are able to accurately identify patients at high risk for acute and delayed CINV.
Cancer-related coping processes as predictors of depressive symptoms, trajectories, and episodes.
Stanton Annette L,Wiley Joshua F,Krull Jennifer L,Crespi Catherine M,Weihs Karen L
Journal of consulting and clinical psychology
OBJECTIVE:Although numerous studies address the relationships of depression with coping processes directed toward approaching or avoiding stressful experiences, the large majority are cross-sectional in design, assess coping processes at only one timepoint, or solely include prediction of the linear slope of depressive symptoms. In this research, coping processes were investigated as predictors of depressive symptoms, symptom trajectory classes (consistently high, recovery, consistently low), and major depressive episodes (MDEs) over 12 months in the cancer context. METHOD:Women (N = 460) within 4 months of breast cancer diagnosis completed assessments of cancer-related coping processes, depressive symptoms, and MDEs at 7 points across 1 year. RESULTS:Beyond sociodemographic and medical variables, coping through cancer-related avoidance an average of 2 months after diagnosis was associated with likelihood of being in the high depressive symptom trajectory class and occurrence of a MDE during the year. Less decline in avoidant coping over time also predicted poor outcomes. In contrast, high initial engagement in approach-oriented coping, as well as increases in coping through emotional expression and acceptance, were associated with lower depressive symptoms across assessments and higher likelihood of being in the recovery or low trajectory class. CONCLUSIONS:Greater engagement in cancer-related avoidant coping was associated with all three indicators of depression, and greater approach-oriented coping was related to more favorable outcomes (except MDE). Sustained or increasing coping through emotional expression or acceptance predicted recovery from initially high depressive symptoms. Approach- and avoidance-oriented coping processes constitute malleable targets for preventive and ameliorative approaches. (PsycINFO Database Record (c) 2018 APA, all rights reserved).
Follow-Up of Cancer Patients Receiving Anti-PD-(L)1 Therapy Using an Electronic Patient-Reported Outcomes Tool (KISS): Prospective Feasibility Cohort Study.
Iivanainen Sanna,Alanko Tuomo,Vihinen Pia,Konkola Teemu,Ekstrom Jussi,Virtanen Henri,Koivunen Jussi
JMIR formative research
BACKGROUND:Immune checkpoint inhibitors (ICIs) have become a standard of care for various tumor types. Their unique spectrum of side effects demands continuous and long-lasting assessment of symptoms. Electronic patient-reported outcome (ePRO) follow-up has been shown to improve survival and quality of life of cancer patients treated with chemotherapy. OBJECTIVE:This study aimed to investigate whether ePRO follow-up of cancer patients treated with ICIs is feasible. The study analyzed (1) the variety of patient reported symptoms, (2) etiology of alerts, (3) symptom correlations, and (4) patient compliance. METHODS:In this prospective, one-arm, multi-institutional study, we recruited adult cancer patients whose advanced cancer was treated with anti-programmed cell death protein 1 (PD)- ligand (L)1 agents in outpatient settings. The ePRO tool consisted of a weekly questionnaire evaluating the presence of typical side effects, with an algorithm assessing the severity of the symptom according to National Cancer Institute Common Terminology Criteria for Adverse Events and an urgency algorithm sending alerts to the care team. A patient experience survey was conducted monthly. The patients were followed up to 6 months or until disease progression. RESULTS:A total of 889 symptom questionnaires was completed by 37 patients (lung cancer, n=15; melanoma, n=9; genitourinary cancer, n=9; head and neck cancer, n=4). Patients showed good adherence to ePRO follow-up. The most common grade 1 symptoms were fatigue (28%) and itching (13%), grade 2 symptoms were loss of appetite (12%) and nausea (12%), and grade 3-4 symptoms were cough (6%) and loss of appetite (4%). The most common reasons for alerts were loss of appetite and shortness of breath. In the treatment benefit analysis, positive correlations were seen between clinical benefit and itching as well as progressive disease and chest pain. CONCLUSIONS:According to the results, ePRO follow-up of cancer patients receiving ICIs is feasible. ePROs capture a wide range of symptoms. Some symptoms correlate to treatment benefit, suggesting that individual prediction models could be generated. TRIAL REGISTRATION:Clinical Trials Register, NCT3928938; https://clinicaltrials.gov/ct2/show/NCT03928938.
Clinical Relevance and Prognostic Value of Inflammatory Biomarkers: A prospective Study in Terminal Cancer Patients Receiving Palliative Care.
Cunha Gabriella da Costa,Rosa Karla Santos da Costa,Wiegert Emanuelly Varea Maria,de Oliveira Livia Costa
Journal of pain and symptom management
CONTEXT:Inflammatory biomarkers have prognostic value in cancer patients, but the feasibility of their use with terminal cancer patients and the related cutoff points are poorly explored. OBJECTIVES:To describe the percentiles values of inflammatory biomarkers; to identify their cutoff points in relation to death; and to determine the prognostic value of C-reactive protein (CRP), leukocytes, neutrophils, neutrophil/lymphocyte ratio (NLR), CRP/albumin ratio (CAR), and modified Glasgow Prognostic Score for death within 90 days, in terminal cancer patients receiving palliative care. METHODS:Prospective cohort study that included patients who received palliative care at the Palliative Care Unit of the National Cancer Institute (Brazil) between October 2019 and March 2020. Receiver operating characteristic curves were used to identify the optimal cutoff points of the inflammatory biomarkers for the prediction of death in 90 days. Kaplan-Meier curves and Cox regression were used to verify the prognostic value of these cutoff points and concordance statistic (C-statistic) was used to test their predictive accuracy. RESULTS:A total 205 patients (mean age: 62.5 years; female: 59%) were included in the study. The optimal cutoff points were CRP ≥6.7mg/L, CAR ≥2.0, leukocytes ≥9300/μL, neutrophils ≥7426/μL and NLR ≥6.0. All biomarkers showed prognostic value and good predictive accuracy when their cutoff points were used, especially CAR, which presented excellent discrimination power (C-statistic: 0.80). CONCLUSION:The inflammatory biomarkers analyzed are independent predictive factors for death within 90 days in terminal cancer patients. CAR appears to be the most useful parameter for predicting survival in these patients.
Computational prediction of state anxiety in Asian patients with cancer susceptible to chemotherapy-induced nausea and vomiting.
Yap Kevin Yi-Lwern,Low Xiu Hui,Chui Wai Keung,Chan Alexandre,
Journal of clinical psychopharmacology
State anxiety, a risk factor for chemotherapy-induced nausea and vomiting (CINV), is a subjective symptom and difficult to quantify. Clinicians need appropriate anxiety measures to assess patients' risks of CINV. This study aimed to determine the anxiety characteristics that can predict CINV based on computational analysis of an objective assessment tool. A single-center, prospective, observational study was carried out between January 2007 and July 2010. Patients with breast, head and neck, and gastrointestinal cancers were recruited and treated with a variety of chemotherapy protocols and appropriate antiemetics. Chemotherapy-induced nausea and vomiting characteristics and antiemetic use were recorded using a standardized diary, whereas patients' anxiety characteristics were evaluated using the Beck Anxiety Inventory. Principal component (PC) analysis was performed to analyze the anxiety characteristics. A subset known as principal variables, which had the highest PC weightings, was identified for patients with and without complete response, complete protection, and complete control. Chemotherapy-induced nausea and vomiting events and anxiety characteristics of 710 patients were collated; 51%, 30%, and 20% were on anthracycline-, oxaliplatin-, and cisplatin-based therapies, respectively. Most patients suffered from delayed CINV, with decreasing proportions achieving complete response (58%), complete protection (42%), and complete control (27%). Seven symptoms (fear of dying, fear of the worst, unable to relax, hot/cold sweats, nervousness, faintness, numbness) were identified as potential CINV predictors. This study demonstrates the usefulness of PC analysis, an unsupervised machine learning technique, to identify 7 anxiety characteristics that are useful as clinical CINV predictors. Clinicians should be aware of these characteristics when assessing CINV in patients on emetogenic chemotherapies.
Prediction of evening fatigue severity in outpatients receiving chemotherapy: less may be more.
Fatigue : biomedicine, health & behavior
BACKGROUND:Fatigue is the most common and debilitating symptom experienced by oncology patients undergoing chemotherapy. Little is known about patient characteristics that predict changes in fatigue severity over time. PURPOSE:To predict the severity of evening fatigue in the week following the administration of chemotherapy using machine learning approaches. METHODS:Outpatients with breast, gastrointestinal, gynecological, or lung cancer (=1217) completed questionnaires one week prior to and one week following administration of chemotherapy. Evening fatigue was measured with the Lee Fatigue Scale (LFS). Separate prediction models for evening fatigue severity were created using clinical, symptom, and psychosocial adjustment characteristics and either evening fatigue scores or individual fatigue item scores. Prediction models were created using two regression and three machine learning approaches. RESULTS:Random forest (RF) models provided the best fit across all models. For the RF model using individual LFS item scores, two of the 13 individual LFS items (i.e., "worn out", "exhausted") were the strongest predictors. CONCLUSION:This study is the first to use machine learning techniques to predict evening fatigue severity in the week following chemotherapy from fatigue scores obtained in the week prior to chemotherapy. Our findings suggest that the language used to assess clinical fatigue in oncology patients is important and that two simple questions may be used to predict evening fatigue severity.
Predicting late symptoms of head and neck cancer treatment using LSTM and patient reported outcomes.
Proceedings. International Database Engineering and Applications Symposium
Patient-Reported Outcome (PRO) surveys are used to monitor patients' symptoms during and after cancer treatment. Acute symptoms refer to those experienced during treatment and late symptoms refer to those experienced after treatment. While most patients experience severe symptoms during treatment, these usually subside in the late stage. However, for some patients, late toxicities persist negatively affecting the patient's quality of life (QoL). In the case of head and neck cancer patients, PRO surveys are recorded every week during the patient's visit to the clinic and at different follow-up times after the treatment has concluded. In this paper, we model the PRO data as a time-series and apply Long-Short Term Memory (LSTM) neural networks for predicting symptom severity in the late stage. The PRO data used in this project corresponds to MD Anderson Symptom Inventory (MDASI) questionnaires collected from head and neck cancer patients treated at the MD Anderson Cancer Center. We show that the LSTM model is effective in predicting symptom ratings under the RMSE and NRMSE metrics. Our experiments show that the LSTM model also outperforms other machine learning models and time-series prediction models for these data.
Symptom presentations and other characteristics of colorectal cancer patients and the diagnostic performance of the Auckland Regional Grading Criteria for Suspected Colorectal Cancer in the South Auckland population.
Hsiang John C,Bai Wayne,Lal Dinesh
The New Zealand medical journal
AIM:This study reviews the presenting symptoms of colorectal cancer in the ethnically diverse Middlemore Hospital referral population of South Auckland, New Zealand. The performance of the newly introduced Auckland Regional Grading Criteria as prediction tool for selecting colorectal cancer cases referred from primary care was evaluated in this group. METHOD:Retrospective review of all colorectal cancer (CRC) cases diagnosed between January 2006 and January 2011. Information extracted from case note review was used to grade patients using the Auckland Regional Grading Criteria. RESULTS:A total of 799 patients were included. The commonest symptoms were: rectal bleeding (25.5-42.3%) and change in bowel habit (20.6-26.8%). Low-risk symptoms including abdominal pain (16.3-46.8%) and weight loss (18.4-26.1%) were not uncommon. 64.4% of Maori and 64.9% of Pacific patients had stage III or IV cancers. Pacific patients had more stage IV disease, 37.7% (p<0.001) and were less likely to undergo tumour resection, 26.0% (p<0.001). The Auckland Regional Grading Criteria would miss 24.7% of the patients with CRC in the referral population. CONCLUSION:While rectal bleeding and change in bowel habit are frequent presenting symptoms, low-risk atypical symptoms including constipation, weight loss and abdominal pain were not uncommon. Significant proportion of Pacific patients present with late-stage disease. The current Auckland Regional grading criteria would miss significant proportion of our study population with colorectal cancer.
Risk factors of falls among inpatients with cancer.
Jun M D,Lee K M,Park S A
International nursing review
AIM:To investigate the risk factors and predictors of falls according to the general characteristics, conscious state, physical condition and treatment of hospitalized patients with cancer. BACKGROUND:Inpatients with cancer experience falls more frequently than those without cancer, and the degree of injuries is more severe among inpatients with cancer. A specific fall prevention strategy is needed for each patient. Prevention of falls in patients with cancer is very important for improving the quality of nursing care. METHODS:This retrospective study included matched case-control patients. We evaluated patients between January 1, 2013, and December 31, 2014. A total of 356 patients (fall group, 178; non-fall group, 178) were included. For fall prediction, logistic regression was performed on the variables that were statistically significant in the univariate analysis. RESULTS:The variables that were significant predictors of falls were the use of an assistive device, history of falls and fatigue. DISCUSSION:The predictors of falls in patients with cancer include physical conditions and general characteristics. Fall prevention strategies in patients with cancer should be planned individually with multifaceted aspects, including physical symptom management. LIMITATIONS:The study was conducted at a single cancer center in Korea; thus, our results cannot be generalized. Additionally, in Korea, it is common to have family members or private caregivers for patient care, and this might have influenced the results. CONCLUSION AND IMPLICATIONS FOR NURSING AND HEALTH POLICY:The predictive factors for falls reflect the nature of the patient's environment, culture and disease. Falls have a negative effect on patient safety and can significantly influence quality of life. Policies for patient safety need more specialized and customized approaches.
PROgnostic Model for Advanced Cancer (PRO-MAC).
Hum Allyn,Wong Yoko Kin Yoke,Yee Choon Meng,Lee Chung Seng,Wu Huei Yaw,Koh Mervyn Yong Hwang
BMJ supportive & palliative care
OBJECTIVE:To develop and validate a simple prognostic tool for early prediction of survival of patients with advanced cancer in a tertiary care setting. DESIGN:Prospective cohort study with 2 years' follow-up. SETTING:Single tertiary teaching hospital in Singapore. PARTICIPANTS:The study includes consecutive patients diagnosed with advanced cancer who were referred to a palliative care unit between 2013 and 2015 (N=840). Data were randomly split into training (n=560) and validation (n=280) sets. RESULTS:743 (88.5%) patients died with a mean follow-up of 97.0 days (SD 174.0). Cox regression modelling was used to build a prognostic model, cross-validating with six randomly split dataset pairs. Predictor variables for the model included functional status (Palliative Performance Scale, PPS V.2), symptoms (Edmonton Symptom Assessment System, ESASr), clinical assessment (eg, the number of organ systems with metastasis, serum albumin and total white cell count level) and patient demographics. The area under the receiver operating characteristic curve using the final averaged prognostic model was between 0.69 and 0.75. Our model classified patients into three prognostic groups, with a median survival of 79.0 days (IQR 175.0) for the low-risk group (0-1.5 points), 42.0 days (IQR 75.0) for the medium-risk group (2.0-5.5 points), and 15.0 days (IQR 28.0) for the high-risk group (6.0-10.5 points). CONCLUSIONS:PROgnostic Model for Advanced Cancer (PRO-MAC) takes into account patient and disease-related factors and identify high-risk patients with 90-day mortality. PPS V.2 and ESASr are important predictors. PRO-MAC will help physicians identify patients earlier for supportive care, facilitating multidisciplinary, shared decision-making.
Machine learning-based random forest for predicting decreased quality of life in thyroid cancer patients after thyroidectomy.
Liu Yong Hong,Jin Jian,Liu Yun Jiang
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
OBJECTIVE:Decreased quality of life (QoL) in thyroid cancer patients after thyroidectomy is a common, but there is a lack of predictive methods for decreased QoL. This study aimed to construct a machine learning-based random forest for predicting decreased QoL in thyroid cancer patients 3 months after thyroidectomy. MATERIALS AND METHODS:Two hundred and eighty-six thyroid cancer patients after thyroidectomy were enrolled in this prospective cross-sectional study from November 2018 to June 2019, and were randomly assigned to training and validation cohorts at a ratio of 7:3. The European Organization for Research and Treatment of Cancer quality of life questionnaire version 3 (EORTC QLQ-C30) questionnaire was used to assess the QoL 3 months after thyroidectomy, and decreased QoL was defined as EORTC QLQ-C30 < 60 points. The random forest model was constructed for predicting decreased QoL in thyroid cancer patients after thyroidectomy. RESULTS:The mean QoL 3 months after thyroidectomy was 65.93 ± 9.00 with 21.33% (61/286) decreased QoL. The main manifestation is fatigue in symptom scales and social functioning dysfunction in functional scales. The top seven most important indices affecting QoL were clinical stage, marital status, histological type, age, nerve injury symptom, economic income and surgery type. For random forest prediction model, the areas under the curve in the training and validation courts were 0.834 and 0.897, respectively. CONCLUSION:The present study demonstrated that random forest model for predicting decreased QoL in thyroid cancer patients 3 months after thyroidectomy displayed relatively high accuracy. These findings should be applied clinically to optimise health care.
Edmonton symptom assessment scale as a prognosticative indicator in patients with advanced cancer.
Zeng Liang,Zhang Liying,Culleton Shaelyn,Jon Florencia,Holden Lori,Kwong Justin,Khan Luluel,Tsao May,Danjoux Cyril,Sahgal Arjun,Barnes Elizabeth,Chow Edward
Journal of palliative medicine
BACKGROUND:Few studies incorporate patient self-assessment scales in prognostic models of survival prediction. The Edmonton Symptom Assessment Scale (ESAS) is commonly used as a symptom screening tool in cancer patients. OBJECTIVE:The goal of this study was to evaluate the prognostic value of the ESAS for survival prediction in the advanced cancer population. MATERIALS AND METHODS:Patients completed the ESAS and demographic information prior to palliative radiotherapy consultation and at follow-up at the Odette Cancer Centre between 1999 and 2009. Generalized estimating equation (GEE) methodology was applied to analyze ESAS trends within the last months of life. One-way analysis of variance (ANOVA) with repeated measurements was used to characterize trends between time periods. RESULTS:ESAS records (2377) from 808 patients (433 male and 375 female) were included in this cohort. Median age was 68 years (range 32-95) with median Karnofsky performance status (KPS) of 60 (range 10-100). Primary cancer sites were of the lung (36%), breast (20%), and prostate (19%). All nine ESAS symptoms significantly deteriorated in the last 4 weeks immediately before death when compared with those scores in the preceding months. At one week prior to death, the worst ESAS symptoms experienced by patients were fatigue, appetite, and well-being with mean scores of 7.4, 6.9, and 6.1, respectively. CONCLUSIONS:All ESAS scores significantly worsened in the last 4 weeks prior to death compared with those in the previous months. Sudden deterioration of the global ESAS symptoms may predict impending death. Future studies on a prognostic model should incorporate both ESAS symptom severity and trends.
The impact of symptom interference using the MD Anderson Symptom Inventory-Brain Tumor Module (MDASI-BT) on prediction of recurrence in primary brain tumor patients.
Armstrong Terri S,Vera-Bolanos Elizabeth,Gning Ibrahima,Acquaye Alvina,Gilbert Mark R,Cleeland Charles,Mendoza Tito
BACKGROUND:Tumor grade, age, extent of resection, and performance status are established prognostic factors for survival in primary brain tumor (PBT) patients. Development of disease-related symptoms is predictive of tumor recurrence in other cancers but has not been reported in the PBT population. METHODS:A cross-sectional sample of 294 PBT patients participated. Progression was based on the radiologist report of the magnetic resonance imaging (MRI). The relation of clinical variables (age, extent of resection, tumor grade, and Karnofsky performance status [KPS]) and MD Anderson Symptom Inventory-Brain Tumor Module (MDASI-BT) mean symptom and interference subscales with progression was examined using logistic regression. RESULTS:The study enrolled more men (60%, n = 175); median age was 46 years. The majority had less than a gross total resection (n = 186, 64%), and a good KPS (KPS ≥ 90) (N = 208). The majority had a grade 3 or 4 tumor (n = 199) and 24% of patients had recurrence. Tumor grade and activity-related interference were significantly related to progression. Patients with tumor grade 4 were 2.4 times more likely to have recurrence (95% CI, 1.2-5.; P < .015). Patients with significant (ratings of ≥ 5) activity-related interference were 3.8 times more likely to have recurrence (95% CI, 2.14-6.80; P < .001). Mean activity-related score was 4.8 for those with progression on MRI and 2.2 for those with stable disease. CONCLUSIONS:Significant activity-related interference and tumor grade were associated with recurrence but not KPS, age, or extent of resection. These results provide preliminary support for the use of symptom interference in assessment of disease status. Because the authors used a cross-sectional sample, future studies evaluating change over time are needed.
Breast Cancer Patients' Depression Prediction by Machine Learning Approach.
One of the most common cancer in females is breasts cancer. This cancer can has high impact on the women including health and social dimensions. One of the most common social dimension is depression caused by breast cancer. Depression can impairs life quality. Depression is one of the symptom among the breast cancer patients. One of the solution is to eliminate the depression in breast cancer patients is by treatments but these treatments can has different unpredictable impacts on the patients. Therefore it is suitable to develop algorithm in order to predict the depression range.
Patient-reported symptom distress, and most bothersome issues, before and during cancer treatment.
Hong Fangxin,Blonquist Traci M,Halpenny Barbara,Berry Donna L
Patient related outcome measures
INTRODUCTION:Frequently reported symptoms and treatment side effects may not be the most bothersome issues to patients with cancer. The purpose of this study was to investigate patient-reported symptom distress and bothersome issues among participants with cancer. METHODS:Participants completed the Symptom Distress Scale-15 before treatment (T1) and during cancer treatment (T2) and reported up to two most bothersome issues among symptoms rated with moderate-to-severe distress. We compared symptom ratings and perceived bother and explored two approaches predicting patients' most bothersome issues: worst absolute symptom score or worst change from pretreatment. RESULTS:Significantly, (P≤0.0002) more patients reported moderate-to-severe distress at T2 for eight of 13 symptoms. At T1, 81% of patients reported one and 56% reported multiple symptoms with moderate-to-severe distress, while at T2, 89% reported one and 69% reported multiple symptoms with moderate-to-severe distress. Impact on sexual activity/interest, pain, fatigue, and insomnia were the most prevalent symptoms with moderate-to-severe distress. Fatigue, pain, and insomnia were perceived most often as bothersome. When one symptom was rated moderate-to-severe, predictive accuracy of the absolute score was 46% and 48% (T1 & T2) and 38% with the change score (T2-T1). When two or more symptoms were rated moderate-to-severe, predictive accuracy of the absolute score was 76% and 79% (T1 & T2) and 70% with the change score (T2-T1). CONCLUSION:More patients experienced moderate-to-severe symptom distress after treatment initiation. Patient identification of bothersome issues could not be assumed based on prevalence of symptoms reported with moderate-to-severe distress. The absolute symptom distress scores identified patients' most bothersome issues with good accuracy, outperforming change scores.
Prediction of chemotherapy-induced nausea and vomiting from patient-reported and genetic risk factors.
Puri Sonam,Hyland Kelly A,Weiss Kristine Crowe,Bell Gillian C,Gray Jhanelle E,Kim Richard,Lin Hui-Yi,Hoogland Aasha I,Gonzalez Brian D,Nelson Ashley M,Kinney Anita Y,Fischer Stacy M,Li Daneng,Jacobsen Paul B,McLeod Howard L,Jim Heather S L
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
PURPOSE:Chemotherapy-induced nausea and vomiting (CINV) is common among cancer patients. Early identification of patients at risk for CINV may help to personalize anti-emetic therapies. To date, few studies have examined the combined contributions of patient-reported and genetic risk factors to CINV. The goal of this study was to evaluate these risk factors. METHODS:Prior to their first chemotherapy infusion, participants completed demographic and risk factor questionnaires and provided a blood sample to measure genetic variants in ABCB1 (rs1045642) and HTR3B (rs45460698) as well as CYP2D6 activity score. The M.D. Anderson Symptom Inventory was completed at 24 h and 5-day post-infusion to assess the severity of acute and delayed CINV, respectively. RESULTS:Participants were 88 patients (55% female, M = 60 years). A total of 23% experienced acute nausea and 55% delayed nausea. Younger age, history of pregnancy-related nausea, fewer hours slept the night prior to infusion, and variation in ABCB1 were associated with more severe acute nausea; advanced-stage cancer and receipt of highly emetogenic chemotherapy were associated with more severe delayed nausea (p values < 0.05). In multivariable analyses, ABCB1 added an additional 5% predictive value beyond the 13% variance explained by patient-reported risk factors. CONCLUSIONS:The current study identified patient-reported and genetic factors that may place patients at risk for acute nausea despite receipt of guideline-consistent anti-emetic prophylaxis. Additional studies examining other genetic variants are needed, as well as the development of risk prediction models including both patient-reported and genetic risk factors.
Improving breast cancer care coordination and symptom management by using AI driven predictive toolkits.
Moser E C,Narayan Gayatri
Breast (Edinburgh, Scotland)
Integrated breast cancer care is complex, marked by multiple hand-offs between primary care and specialists over an extensive period of time. Communication is essential for treatment compliance, lowering error and complication risk, as well as handling co-morbidity. The director role of care, however, becomes often unclear, and patients remain lost across departments. Digital tools can add significant value to care communication but need clarity about the directives to perform in the care team. In effective breast cancer care, multidisciplinary team meetings can drive care planning, create directives and structured data collection. Subsequently, nurse navigators can take the director's role and become a pivotal determinant for patient care continuity. In the complexity of care, automated AI driven planning can facilitate their tasks, however, human intervention stays needed for psychosocial support and tackling unexpected urgency. Care allocation of patients across centres, is often still done by hand and phone demanding time due to overbooked agenda's and discontinuous system solutions limited by privacy rules and moreover, competition among providers. Collection of complete outcome information is limited to specific collaborative networks today. With data continuity over time, AI tools can facilitate both care allocation and risk prediction which may unveil non-compliance due to local scarce resources, distance and costs. Applied research is needed to bring AI modelling into clinical practice and drive well-coordinated, patient-centric cancer care in the complex web of modern healthcare today.
Artificial neural networks for simultaneously predicting the risk of multiple co-occurring symptoms among patients with cancer.
Xuyi Wenhui,Seow Hsien,Sutradhar Rinku
Patients with cancer often exhibit multiple co-occurring symptoms which can impact the type of treatment received, recovery, and long-term health. We aim to simultaneously predict the risk of three symptoms: severe pain, moderate-severe depression, and poor well-being in order to flag patients who may benefit from pre-emptive early symptom management. This was a retrospective population-based cohort study of adults diagnosed with cancer between 2008 and 2015. We developed and tested an Artificial Neural Network (ANN) model to predict the risk of multiple co-occurring symptoms within 6 months after diagnosis. The ANN model derived from a training cohort was assessed on an independent test cohort for model performance based on sensitivity, specificity, accuracy, AUC, and calibration. The mutually exclusive training and test cohorts consisted of 35,606 and 10,498 patients, respectively. The area under the curve for the risk of experiencing severe pain, moderate-severe depression, and poor well-being were 71%, 73%, and 70%, respectively. Patient characteristics at highest risk of simultaneously experiencing these three symptoms included: those with lung cancer, late stage cancer, existing chronic conditions such as osteoarthritis, mood disorder, hypertension, diabetes, and coronary disease. Patients with over a 40% risk of severe pain also had over a 70% risk of depression, and over a 55% risk of poor well-being. Our ANN model was able to simultaneously predict the risk of pain, depression, and lack of well-being. Accurate prediction of future symptom burden can serve as an early indicator tool so that providers can implement timely interventions for symptom management, ultimately improving cancer care and quality of life.
Finding Colon Cancer- and Colorectal Cancer-Related Microbes Based on Microbe-Disease Association Prediction.
Chen Yu,Sun Hongjian,Sun Mengzhe,Shi Changguo,Sun Hongmei,Shi Xiaoli,Ji Binbin,Cui Jinpeng
Frontiers in microbiology
Microbes are closely associated with the formation and development of diseases. The identification of the potential associations between microbes and diseases can boost the understanding of various complex diseases. Wet experiments applied to microbe-disease association (MDA) identification are costly and time-consuming. In this manuscript, we developed a novel computational model, NLLMDA, to find unobserved MDAs, especially for colon cancer and colorectal carcinoma. NLLMDA integrated negative MDA selection, linear neighborhood similarity, label propagation, information integration, and known biological data. The Gaussian association profile (GAP) similarity of microbes and GAPs similarity and symptom similarity of diseases were firstly computed. Secondly, linear neighborhood method was then applied to the above computed similarity matrices to obtain more stable performance. Thirdly, negative MDA samples were selected, and the label propagation algorithm was used to score for microbe-disease pairs. The final association probabilities can be computed based on the information integration method. NLLMDA was compared with the other five classical MDA methods and obtained the highest area under the curve (AUC) value of 0.9031 and 0.9335 on cross-validations of diseases and microbe-disease pairs. The results suggest that NLLMDA was an effective prediction method. More importantly, we found that Acidobacteriaceae may have a close link with colon cancer and may densely associate with colorectal carcinoma.
Machine Learning-Based Short-Term Mortality Prediction Models for Patients With Cancer Using Electronic Health Record Data: Systematic Review and Critical Appraisal.
JMIR medical informatics
BACKGROUND:In the United States, national guidelines suggest that aggressive cancer care should be avoided in the final months of life. However, guideline compliance currently requires clinicians to make judgments based on their experience as to when a patient is nearing the end of their life. Machine learning (ML) algorithms may facilitate improved end-of-life care provision for patients with cancer by identifying patients at risk of short-term mortality. OBJECTIVE:This study aims to summarize the evidence for applying ML in ≤1-year cancer mortality prediction to assist with the transition to end-of-life care for patients with cancer. METHODS:We searched MEDLINE, Embase, Scopus, Web of Science, and IEEE to identify relevant articles. We included studies describing ML algorithms predicting ≤1-year mortality in patients of oncology. We used the prediction model risk of bias assessment tool to assess the quality of the included studies. RESULTS:We included 15 articles involving 110,058 patients in the final synthesis. Of the 15 studies, 12 (80%) had a high or unclear risk of bias. The model performance was good: the area under the receiver operating characteristic curve ranged from 0.72 to 0.92. We identified common issues leading to biased models, including using a single performance metric, incomplete reporting of or inappropriate modeling practice, and small sample size. CONCLUSIONS:We found encouraging signs of ML performance in predicting short-term cancer mortality. Nevertheless, no included ML algorithms are suitable for clinical practice at the current stage because of the high risk of bias and uncertainty regarding real-world performance. Further research is needed to develop ML models using the modern standards of algorithm development and reporting.
Patient-reported symptoms and problems at admission to specialized palliative care improved survival prediction in 30,969 cancer patients: A nationwide register-based study.
Hansen Maiken B,Nylandsted Lone Ross,Petersen Morten A,Adsersen Mathilde,Rojas-Concha Leslye,Groenvold Mogens
BACKGROUND:Large, nationally representative studies of the association between quality of life and survival time in cancer patients in specialized palliative care are missing. AIM:The aim of this study was to investigate whether symptoms/problems at admission to specialized palliative care were associated with survival and if the symptoms/problems may improve prediction of death within 1 week and 1 month, respectively. SETTING/PARTICIPANTS:All cancer patients who had filled in the EORTC QLQ-C15-PAL at admission to specialized palliative care in Denmark in 2010-2017 were included through the Danish Palliative Care Database. Cox regression was used to identify clinical variables (gender, age, type of contact (inpatient vs outpatient), and cancer site) and symptoms/problems significantly associated with survival. To test whether symptoms/problems improved survival predictions, the overall accuracy (area under the receiver operating characteristic curve) for different prediction models was compared. The validity of the prediction models was tested with data on 5,508 patients admitted to palliative care in 2018. RESULTS:The study included 30,969 patients with an average age of 68.9 years; 50% were women. Gender, age, type of contact, cancer site, and most symptoms/problems were significantly associated with survival time. The predictive value of symptoms/problems was trivial except for physical function, which clearly improved the overall accuracy for 1-week and 1-month predictions of death when added to models including only clinical variables. CONCLUSION:Most symptoms/problems were significantly associated with survival and mainly physical function improved predictions of death. Interestingly, the predictive value of physical function was the same as all clinical variables combined (in hospice) or even higher (in palliative care teams).
Frailty Index for prediction of surgical outcome in ovarian cancer: Results of a prospective study.
Inci Melisa Guelhan,Anders Louise,Woopen Hannah,Richter Rolf,Guzel Duygu,Armbrust Robert,Sehouli Jalid
BACKGROUND:Complete macroscopic tumor resection is the strongest prognostic factor for patients with ovarian cancer, which requires complex surgery for achievement. Based on the mostly advanced tumor stage and high symptom burden many patients are classified as frail which may limit optimal surgical outcome. Aim of this study is to evaluate the predictive ability of Frailty Index for surgical outcomes in patients with ovarian cancer. METHODS:This prospective study enrolled patients with ovarian cancer undergoing cytoreductive surgery. We classified frailty proposed by Mitnitski et al. regarding the cumulative deficit model of frailty. Utilizing Receiver Operator Characteristic (ROC) analysis and logistic regression, we determined predictive clinical factors for severe postoperative complications. The Kaplan-Meier method and log-rank test were used for overall survival analysis. RESULTS:Out of f 144 enrolled patients, the overall prevalence of frailty based on a Frailty Index >0.26 and Frailty Index >0.15 was 33% and 74%, respectively. The logistic regression shows that frail patients with a Frailty Index >0.26 (Odds ratio (OR): 3.64, 95% CI: 1.34-9.85, p = 0.01), ECOG PS > 1 (OR 6.33, 95% CI:1.31-30.51, p = 0.02) and high surgical complexity score (OR 8.86, 95% CI:1.88-41.76, p = 0.006) had a significant higher risk for severe postoperative complications. According to multivariable cox regression Frailty Index >0.15 (hazard ratio (HR) (HR 1.87, 95% CI: 1.01-3.47, p = 0.048), residual tumor <1 cm (HR 2.75, 95%CI: 1.53-4.99, p = 0.001), residual tumor >1 cm (HR 5.00, 95% CI: 2.74-9.13, p < 0.001) and albumin<35.5 g/dl (HR 1.92, 95% CI: 1.08-3.43, p = 0.03) resulted as significant parameters for poor overall survival. Resulted as significant parameters for poor overall survival. CONCLUSION:Next to surgical complexity score, ECOG PS > 1 and recurrent surgery, Frailty Index >0.26 is associated with severe postoperative complications in patients with ovarian cancer. Besides tumor residuals and low albumin levels a Frailty Index >0.15 predicts poor survival.
Development and validation of a prediction model of poor performance status and severe symptoms over time in cancer patients (PROVIEW+).
BACKGROUND:Predictive cancer tools focus on survival; none predict severe symptoms. AIM:To develop and validate a model that predicts the risk for having low performance status and severe symptoms in cancer patients. DESIGN:Retrospective, population-based, predictive study. SETTING/PARTICIPANTS:We linked administrative data from cancer patients from 2008 to 2015 in Ontario, Canada. Patients were randomly selected for model derivation (60%) and validation (40%). Using the derivation cohort, we developed a multivariable logistic regression model to predict the risk of an outcome at 6 months following diagnosis and recalculated after each of four annual survivor marks. Model performance was assessed using discrimination and calibration plots. Outcomes included low performance status (i.e. 10-30 on Palliative Performance Scale), severe pain, dyspnea, well-being, and depression (i.e. 7-10 on Edmonton Symptom Assessment System). RESULTS:We identified 255,494 cancer patients (57% female; median age of 64; common cancers were breast (24%); and lung (13%)). At diagnosis, the predicted risk of having low performance status, severe pain, well-being, dyspnea, and depression in 6-months is 1%, 3%, 6%, 13%, and 4%, respectively for the reference case (i.e. male, lung cancer, stage I, no symptoms); the corresponding discrimination for each outcome model had high AUCs of 0.807, 0.713, 0.709, 0.790, and 0.723, respectively. Generally these covariates increased the outcome risk by >10% across all models: lung disease, dementia, diabetes; radiation treatment; hospital admission; pain; depression; transitional performance status; issues with appetite; or homecare. CONCLUSIONS:The model accurately predicted changing cancer risk for low performance status and severe symptoms over time.
Comorbid insomnia among breast cancer survivors and its prediction using machine learning: a nationwide study in Japan.
Ueno Taro,Ichikawa Daisuke,Shimizu Yoichi,Narisawa Tomomi,Tsuji Katsunori,Ochi Eisuke,Sakurai Naomi,Iwata Hiroji,Matsuoka Yutaka J
Japanese journal of clinical oncology
OBJECTIVE:Insomnia is an increasingly recognized major symptom of breast cancer which can seriously disrupt the quality of life during and many years after treatment. Sleep problems have also been linked with survival in women with breast cancer. The aims of this study were to estimate the prevalence of insomnia in breast cancers survivors, clarify the clinical characteristics of their sleep difficulties and use machine learning techniques to explore clinical insights. METHODS:Our analysis of data, obtained in a nationwide questionnaire survey of breast cancer survivors in Japan, revealed a prevalence of suspected insomnia of 37.5%. With the clinical data obtained, we then used machine learning algorithms to develop a classifier that predicts comorbid insomnia. The performance of the prediction model was evaluated using 8-fold cross-validation. RESULTS:When using optimal hyperparameters, the L2 penalized logistic regression model and the XGBoost model provided predictive accuracy of 71.5 and 70.6% for the presence of suspected insomnia, with areas under the curve of 0.76 and 0.75, respectively. Population segments with high risk of insomnia were also extracted using the RuleFit algorithm. We found that cancer-related fatigue is a predictor of insomnia in breast cancer survivors. CONCLUSIONS:The high prevalence of sleep problems and its link with mortality warrants routine screening. Our novel predictive model using a machine learning approach offers clinically important insights for the early detection of comorbid insomnia and intervention in breast cancer survivors.
Dietary patterns and severity of symptom with the risk of esophageal squamous cell carcinoma and its histological precursor lesions in China: a multicenter cross-sectional latent class analysis.
Zang Zhaoping,Liu Yong,Wang Jialin,Liu Yuqin,Zhang Shaokai,Zhang Yongzhen,Zhang Liwei,Zhao Deli,Liu Fugang,Chao Lina,Wang Xinzheng,Zhang Chunli,Song Guohui,Zhang Zhiyi,Li Youpeng,Yan Zheng,Wen Yongxiu,Ge Yinyin,Niu Chen,Feng Wei,Nakyeyune Rena,Shen Yi,Shao Yi,Guo Xiuhua,Yang Aiming,Liu Fen,Wang Guiqi
BACKGROUND:Dietary patterns and symptoms research among Chinese with esophageal squamous cell carcinoma (ESCC) and its precursor lesions is limited, especially as it relates to multiple food consumption and multiple co-occurring symptoms. The aim of our study was to identify the dietary patterns and severity of symptom classes with the risk of esophageal squamous cell carcinoma and its histological precursor lesions, and develop a risk prediction model for different stages of esophageal disease. METHODS:We analyzed data from a multicenter cross-sectional study carried out in ESCC high incidence areas between 2017 and 2018, which included 34,707 individuals aged 40-69 years. Dietary patterns and severity of symptom classes were derived by applying a latent class analysis (LCA). A multiple logistic regression model was used to derive the odds ratio (ORs) and corresponding 95% confidence intervals (CIs) for ESCC and the different stages of esophageal disease according to the dietary patterns and severity of symptom classes identified. We built the risk prediction model by using a nomogram. RESULTS:We identified five dietary patterns and three severity of symptom classes. The dietary patterns were classified as follows: "Healthy", "Western", "Lower consumers-combination", "Medium consumers-combination" and "Higher consumers-combination" patterns based on the intake of foods such as red meat, vegetables and fruits. The severity of symptoms was categorized into "Asymptomatic", "Mild symptoms" and "Overt symptoms" classes based on health-related symptoms reported by the participants. Compared to the "Healthy" pattern, the other four patterns were all associated with an increased risk of esophageal disease. Similarly, the other two symptom classes present different degrees of increased risk of esophageal disease compared to the "Asymptomatic". The nomograms reflect the good predictive ability of the model. CONCLUSION:Among individuals aged 40-69 years in high incidence regions of upper gastrointestinal cancer, the results supplied that subjects with diets rich in livestock and poultry meat and low in fruits and vegetables and subjects with typical symptoms were at increased ESCC risk. The findings highlight the importance of considering food and symptom combinations in cancer risk evaluation.
Predictors of Symptom-Specific Treatment Response to Dietary Interventions in Irritable Bowel Syndrome.
Colomier Esther,Van Oudenhove Lukas,Tack Jan,Böhn Lena,Bennet Sean,Nybacka Sanna,Störsrud Stine,Öhman Lena,Törnblom Hans,Simrén Magnus
(1) Background: Predictors of dietary treatment response in irritable bowel syndrome (IBS) remain understudied. We aimed to investigate predictors of symptom improvement during the low FODMAP and the traditional IBS diet for four weeks. (2) Methods: Baseline measures included faecal Dysbiosis Index, food diaries with daily energy and FODMAP intake, non-gastrointestinal (GI) somatic symptoms, GI-specific anxiety, and psychological distress. Outcomes were bloating, constipation, diarrhea, and pain symptom scores treated as continuous variables in linear mixed models. (3) Results: We included 33 and 34 patients on the low FODMAP and traditional IBS diet, respectively. Less severe dysbiosis and higher energy intake predicted better pain response to both diets. Less severe dysbiosis also predicted better constipation response to both diets. More severe psychological distress predicted worse bloating response to both diets. For the different outcomes, several differential predictors were identified, indicating that baseline factors could predict better improvement in one treatment arm, but worse improvement in the other treatment arm. (4) Conclusions: Psychological, nutritional, and microbial factors predict symptom improvement when following the low FODMAP and traditional IBS diet. Findings may help individualize dietary treatment in IBS.
A Machine Learning-Based Approach for the Prediction of Acute Coronary Syndrome Requiring Revascularization.
Noh Yung-Kyun,Park Ji Young,Choi Byoung Geol,Kim Kee-Eung,Rha Seung-Woon
Journal of medical systems
The aim of this study is to predict acute coronary syndrome (ACS) requiring revascularization in those patients presenting early-stage angina-like symptom using machine learning algorithms. We obtained data from 2344 ACS patients, who required revascularization and from 3538 non-ACS patients. We analyzed 20 features that are relevant to ACS using standard algorithms, support vector machines and linear discriminant analysis. Based on feature pattern and filter characteristics, we analyzed and extracted a strong prediction function out of the 20 selected features. The obtained prediction functions are relevant showing the area under curve of 0.860 for the prediction of ACS that requiring revascularization. Some features are missing in many data though they are considered to be very informative; it turned out that omitting those features from the input and using more data without those features for training improves the prediction accuracy. Additionally, from the investigation using the receiver operating characteristic curves, a reliable prediction of 2.60% of non-ACS patients could be made with a specificity of 1.0. For those 2.60% non-ACS patients, we can consider the recommendation of medical treatment without risking misdiagnosis of the patients requiring revascularization. We investigated prediction algorithm to select ACS patients requiring revascularization and non-ACS patients presenting angina-like symptoms at an early stage. In the future, a large cohort study is necessary to increase the prediction accuracy and confirm the possibility of safely discriminating the non-ACS patients from the ACS patients with confidence.
The relation between perceived injustice and symptom severity in individuals with major depression: A cross-lagged panel study.
Sullivan Michael J L,Adams Heather,Yamada Keiko,Kubota Yasuhiko,Ellis Tamra,Thibault Pascal
Journal of affective disorders
BACKGROUND:Perceived injustice has been associated with problematic recovery outcomes in individuals with debilitating health conditions. However, the relation between perceived injustice and recovery outcomes has not been previously examined in individuals with debilitating mental health conditions. The present study examined the relation between perceived injustice and symptom severity in individuals undergoing treatment for Major Depressive Disorder (MDD). METHODS:The study sample consisted of 253 work-disabled individuals with MDD who were referred to an occupational rehabilitation service. Participants completed measures of depressive symptom severity, perceived injustice, catastrophic thinking, pain and occupational disability at three time-points (pre-, mid- and post-treatment) during a 10-week behavioural activation intervention. RESULTS:Regression analysis on baseline data revealed that perceived injustice contributed significant variance to the prediction of depressive symptom severity, beyond the variance accounted for by time since diagnosis, pain severity and catastrophic thinking. Prospective analyses revealed that early treatment reductions in perceived injustice predicted late treatment reductions in depressive symptom severity. LIMITATIONS:The study sample consisted of work-disabled individuals with MDD who had been referred to an occupational rehabilitation service. This selection bias has implications for the generalizability of findings. CONCLUSION:The findings suggest that perceived injustice is a determinant of symptom severity in individuals with MDD. The inclusion of techniques designed to reduce perceived injustice might augment positive treatment outcomes for individuals receiving treatment for MDD.
A symptom-based approach in predicting ECT outcome in depressed patients employing MADRS single items.
Carstens Luisa,Hartling Corinna,Stippl Anna,Domke Ann-Kathrin,Herrera-Mendelez Ana Lucia,Aust Sabine,Gärtner Matti,Bajbouj Malek,Grimm Simone
European archives of psychiatry and clinical neuroscience
Establishing symptom-based predictors of electroconvulsive therapy (ECT) outcome seems promising, however, findings concerning the predictive value of distinct depressive symptoms or subtypes are limited; previous factor-analytic approaches based on the Montgomery-Åsberg Depression Rating Scale (MADRS) remained inconclusive, as proposed factors varied across samples. In this naturalistic study, we refrained from these previous factor-analytic approaches and examined the predictive value of MADRS single items and their change during the course of ECT concerning ECT outcome. We used logistic and linear regression models to analyze MADRS data routinely assessed at three time points in 96 depressed psychiatric inpatients over the course of ECT. Mean age was 53 years (SD 14.79), gender ratio was 58:38 (F:M), baseline MADRS score was M = 30.20 (SD 5.42). MADRS single items were strong predictors of ECT response, remission and overall symptom reduction, especially items 1 (apparent sadness), 2 (reported sadness) and 8 (inability to feel), assessing affective symptoms. Strongest effects were found for regression models including item 2 (reported sadness) with up to 80% correct prediction of ECT outcome. ROC analyses were performed to estimate the optimal cut-point for treatment response. MADRS single items during the course of ECT might pose simple, reliable, time- and cost-effective predictors of ECT outcome. More severe affective symptoms of depression at baseline and a stronger reduction of these affective symptoms during the course of ECT seem to be positively associated with ECT outcome. Precise cut-off values for clinical use were proposed. Generally, these findings underline the benefits of a symptom-based approach in depression research and treatment in addition to depression sum-scores and generalized diagnoses.
Machine Learning-Based Definition of Symptom Clusters and Selection of Antidepressants for Depressive Syndrome.
Kim Il Bin,Park Seon-Cheol
Diagnostics (Basel, Switzerland)
The current polythetic and operational criteria for major depression inevitably contribute to the heterogeneity of depressive syndromes. The heterogeneity of depressive syndrome has been criticized using the concept of language game in Wittgensteinian philosophy. Moreover, "a symptom- or endophenotype-based approach, rather than a diagnosis-based approach, has been proposed" as the "next-generation treatment for mental disorders" by Thomas Insel. Understanding the heterogeneity renders promise for personalized medicine to treat cases of depressive syndrome, in terms of both defining symptom clusters and selecting antidepressants. Machine learning algorithms have emerged as a tool for personalized medicine by handling clinical big data that can be used as predictors for subtype classification and treatment outcome prediction. The large clinical cohort data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D), Combining Medications to Enhance Depression Outcome (CO-MED), and the German Research Network on Depression (GRND) have recently began to be acknowledged as useful sources for machine learning-based depression research with regard to cost effectiveness and generalizability. In addition, noninvasive biological tools such as functional and resting state magnetic resonance imaging techniques are widely combined with machine learning methods to detect intrinsic endophenotypes of depression. This review highlights recent studies that have used clinical cohort or brain imaging data and have addressed machine learning-based approaches to defining symptom clusters and selecting antidepressants. Potentially applicable suggestions to realize machine learning-based personalized medicine for depressive syndrome are also provided herein.
Nomogram for predicting symptom severity during radiation therapy for head and neck cancer.
Sheu Tommy,Fuller Clifton David,Mendoza Tito R,Garden Adam S,Morrison William H,Beadle Beth M,Phan Jack,Frank Steven J,Hanna Ehab Y,Lu Charles,Cleeland Charles S,Rosenthal David I,Gunn G Brandon
Otolaryngology--head and neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery
OBJECTIVES:Radiation therapy (RT), with or without chemotherapy, can cause significant acute toxicity among patients treated for head and neck cancer (HNC), but predicting, before treatment, who will experience a particular toxicity or symptom is difficult. We created and evaluated 2 multivariate models and generated a nomogram to predict symptom severity during RT based on a patient-reported outcome (PRO) instrument, the MD Anderson Symptom Inventory-Head and Neck Module (MDASI-HN). STUDY DESIGN:This was a prospective, longitudinal, questionnaire-based study. SETTING:Tertiary cancer care center. SUBJECTS AND METHODS:Subjects were 264 patients with HNC (mostly oropharyngeal) who had completed the MDASI-HN before and during therapy. Pretreatment variables were correlated with MDASI-HN symptom scores during therapy with multivariate modeling and then were correlated with the composite MDASI-HN score during week 5 of therapy. RESULTS:A multivariate model incorporating pretreatment PROs better predicted MDASI-HN symptom scores during treatment than did a model based on clinical variables and physician-rated patient performance status alone (Akaike information criterion = 1442.5 vs 1459.9). In the most parsimonious model, pretreatment MDASI-HN symptom severity (P < .001), concurrent chemotherapy (P = .006), primary tumor site (P = .016), and receipt of definitive (rather than adjuvant) RT (P = .044) correlated with MDASI-HN symptom scores during week 5. That model was used to construct a nomogram. CONCLUSION:Our model demonstrates the value of incorporating baseline PROs, in addition to disease and treatment characteristics, to predict patient symptom burden during therapy. Although additional investigation and validation are required, PRO-inclusive prediction tools can be useful for improving symptom interventions and expectations for patients being treated for HNC.
Symptom presence versus symptom intensity in understanding the severity of depression: Implications for documentation in electronic medical records.
Zimmerman Mark,Balling Caroline,Chelminski Iwona,Dalrymple Kristy
Journal of affective disorders
BACKGROUND:Data mining efforts have been applied to research data bases to develop statistical models for predicting outcomes. Electronic medical records have the potential to enable efforts to apply statistical techniques to mine large clinical data bases. Of course, such prediction algorithms will only be as good as the data that is available to input. The question that we address in the present report from the Rhode Island Methods to Improve Diagnostic Assessment and Services (MIDAS) project is how much information might be gained from dimensional ratings of symptom severity over and above that which is accounted for when determining symptom presence. Such results could have implications for how medical record documentation should be established. METHODS:Patients were evaluated with a semi-structured interview, and the presence of each symptom of major depressive disorder (MDD) was recorded. Patients were also rated on the Clinical Global Index of Severity (CGI-S). RESULTS:A multiple regression analysis entering the presence of MDD symptoms as predictors of the CGI had a cumulative R of 0.26. A multiple regression analysis entering all symptom severity ratings as predictors of the CGI had a cumulative R of 0.40. LIMITATIONS:The study was based on patients presenting for outpatient treatment to a single clinical practice. Symptoms that are not diagnostic criteria for MDD were not examined. DISCUSSION:Research institutions interested in using data mining statistical approaches of electronic medical records should consider having the clinicians rate whether symptoms are mild, moderate or severe and not just whether they are present or absent.
A Validated Model to Predict Postoperative Symptom Severity After Mandibular Third Molar Removal.
Qiao Feng,Huang Xiaohuan,Li Bolong,Dong Rui,Huang Xin,Sun Jun
Journal of oral and maxillofacial surgery : official journal of the American Association of Oral and Maxillofacial Surgeons
PURPOSE:The individualized prediction of postoperative symptom severity is essential for selecting interventions after mandibular third molar (M3M) removal. The purpose of the present study was to develop and validate a nomogram for personal prediction of postoperative symptom severity. MATERIALS AND METHODS:A prospective cohort study was performed in the Stomatology Hospital of Tianjin Medical University. The sample was divided into training and testing data sets by time. The demographic, anatomic, radiographic, and operative variables were recorded. The self-reported postoperative symptom severity was recorded and defined as the primary outcome variable. Stepwise forward algorithms were applied to informative predictors based on Akaike's information criterion. Multivariable logistic regression analysis was used to develop the nomogram. An independent testing data set was used to validate the nomogram. Receiver operating characteristic curves and the Hosmer-Lemeshow test were used to assess model performance. P < .05 was considered to indicate statistical significance. RESULTS:The sample included 321 subjects who had undergone M3M removal. An independent validation data set included 103 consecutive patients. The median operation time was 15.0 minutes (interquartile range, 8.3 to 21.6 minutes) in the training data set (n = 218). Patients with serious postoperative symptoms accounted for 48.6 and 47.6% of the training and testing data sets, respectively. Gender, age, smoking status, operation time, Pell-Gregory ramus classification, and preoperative symptoms were identified as predictors and assembled into the nomogram. The area under curve demonstrated adequate discrimination in the validation data set (0.69; 95% confidence interval, 0.59 to 0.80). The nomogram was well calibrated, with a Hosmer-Lemeshow χ statistic of 6.33 (P = .78) in the testing data set. The confusion matrix was also summarized, and the accuracy was 63.3 and 65.1% in the training and testing data set, respectively. CONCLUSIONS:The present study has proposed an effective nomogram with potential application in facilitating the individualized prediction of postoperative symptom severity after M3M removal.
Combination of four clinical indicators predicts the severe/critical symptom of patients infected COVID-19.
Sun Liping,Song Fengxiang,Shi Nannan,Liu Fengjun,Li Shenyang,Li Ping,Zhang Weihan,Jiang Xiao,Zhang Yongbin,Sun Lining,Chen Xiong,Shi Yuxin
Journal of clinical virology : the official publication of the Pan American Society for Clinical Virology
BACKGROUND:Despite the death rate of COVID-19 is less than 3%, the fatality rate of severe/critical cases is high, according to World Health Organization (WHO). Thus, screening the severe/critical cases before symptom occurs effectively saves medical resources. METHODS AND MATERIALS:In this study, all 336 cases of patients infected COVID-19 in Shanghai to March 12th, were retrospectively enrolled, and divided in to training and test datasets. In addition, 220 clinical and laboratory observations/records were also collected. Clinical indicators were associated with severe/critical symptoms were identified and a model for severe/critical symptom prediction was developed. RESULTS:Totally, 36 clinical indicators significantly associated with severe/critical symptom were identified. The clinical indicators are mainly thyroxine, immune related cells and products. Support Vector Machine (SVM) and optimized combination of age, GSH, CD3 ratio and total protein has a good performance in discriminating the mild and severe/critical cases. The area under receiving operating curve (AUROC) reached 0.9996 and 0.9757 in the training and testing dataset, respectively. When the using cut-off value as 0.0667, the recall rate was 93.33 % and 100 % in the training and testing datasets, separately. Cox multivariate regression and survival analyses revealed that the model significantly discriminated the severe/critical cases and used the information of the selected clinical indicators. CONCLUSION:The model was robust and effective in predicting the severe/critical COVID cases.
Machine Learning Models Identify Multimodal Measurements Highly Predictive of Transdiagnostic Symptom Severity for Mood, Anhedonia, and Anxiety.
Mellem Monika S,Liu Yuelu,Gonzalez Humberto,Kollada Matthew,Martin William J,Ahammad Parvez
Biological psychiatry. Cognitive neuroscience and neuroimaging
BACKGROUND:Insights from neuroimaging-based biomarker research have not yet translated into clinical practice. This translational gap may stem from a focus on diagnostic classification, rather than on prediction of transdiagnostic psychiatric symptom severity. Currently, no transdiagnostic, multimodal predictive models of symptom severity that include neurobiological characteristics have emerged. METHODS:We built predictive models of 3 common symptoms in psychiatric disorders (dysregulated mood, anhedonia, and anxiety) from the Consortium for Neuropsychiatric Phenomics dataset (N = 272), which includes clinical scale assessments, resting-state functional magnetic resonance imaging (MRI), and structural MRI measures from patients with schizophrenia, bipolar disorder, and attention-deficit/hyperactivity disorder and healthy control subjects. We used an efficient, data-driven feature selection approach to identify the most predictive features from these high-dimensional data. RESULTS:This approach optimized modeling and explained 65% to 90% of variance across the 3 symptom domains, compared to 22% without using the feature selection approach. The top performing multimodal models retained a high level of interpretability that enabled several clinical and scientific insights. First, to our surprise, structural features did not substantially contribute to the predictive strength of these models. Second, the Temperament and Character Inventory scale emerged as a highly important predictor of symptom variation across diagnoses. Third, predictive resting-state functional MRI connectivity features were widely distributed across many intrinsic resting-state networks. CONCLUSIONS:Combining resting-state functional MRI with select questions from clinical scales enabled high prediction of symptom severity across diagnostically distinct patient groups and revealed that connectivity measures beyond a few intrinsic resting-state networks may carry relevant information for symptom severity.
Development and validation of multivariable prediction models of remission, recovery, and quality of life outcomes in people with first episode psychosis: a machine learning approach.
Leighton Samuel P,Upthegrove Rachel,Krishnadas Rajeev,Benros Michael E,Broome Matthew R,Gkoutos Georgios V,Liddle Peter F,Singh Swaran P,Everard Linda,Jones Peter B,Fowler David,Sharma Vimal,Freemantle Nicholas,Christensen Rune H B,Albert Nikolai,Nordentoft Merete,Schwannauer Matthias,Cavanagh Jonathan,Gumley Andrew I,Birchwood Max,Mallikarjun Pavan K
The Lancet. Digital health
BACKGROUND:Outcomes for people with first-episode psychosis are highly heterogeneous. Few reliable validated methods are available to predict the outcome for individual patients in the first clinical contact. In this study, we aimed to build multivariable prediction models of 1-year remission and recovery outcomes using baseline clinical variables in people with first-episode psychosis. METHODS:In this machine learning approach, we applied supervised machine learning, using regularised regression and nested leave-one-site-out cross-validation, to baseline clinical data from the English Evaluating the Development and Impact of Early Intervention Services (EDEN) study (n=1027), to develop and internally validate prediction models at 1-year follow-up. We assessed four binary outcomes that were recorded at 1 year: symptom remission, social recovery, vocational recovery, and quality of life (QoL). We externally validated the prediction models by selecting from the top predictor variables identified in the internal validation models the variables shared with the external validation datasets comprised of two Scottish longitudinal cohort studies (n=162) and the OPUS trial, a randomised controlled trial of specialised assertive intervention versus standard treatment (n=578). FINDINGS:The performance of prediction models was robust for the four 1-year outcomes of symptom remission (area under the receiver operating characteristic curve [AUC] 0·703, 95% CI 0·664-0·742), social recovery (0·731, 0·697-0·765), vocational recovery (0·736, 0·702-0·771), and QoL (0·704, 0·667-0·742; p<0·0001 for all outcomes), on internal validation. We externally validated the outcomes of symptom remission (AUC 0·680, 95% CI 0·587-0·773), vocational recovery (0·867, 0·805-0·930), and QoL (0·679, 0·522-0·836) in the Scottish datasets, and symptom remission (0·616, 0·553-0·679), social recovery (0·573, 0·504-0·643), vocational recovery (0·660, 0·610-0·710), and QoL (0·556, 0·481-0·631) in the OPUS dataset. INTERPRETATION:In our machine learning analysis, we showed that prediction models can reliably and prospectively identify poor remission and recovery outcomes at 1 year for patients with first-episode psychosis using baseline clinical variables at first clinical contact. FUNDING:Lundbeck Foundation.
Evaluation of the prediction of CoVID-19 recovered and unrecovered cases using symptoms and patient's meta data based on support vector machine, neural network, CHAID and QUEST Models.
Al-Najjar D,Al-Najjar H,Al-Rousan N
European review for medical and pharmacological sciences
OBJECTIVE:This paper aims to develop four prediction models for recovered and unrecovered cases using descriptive data of patients and symptoms of CoVID-19 patients. The developed prediction models aim to extract the important variables in predicting recovered cases by using the binary values for recovered cases. MATERIALS AND METHODS:The data were collected from different countries all over the world. The input of the prediction model contains 28 symptoms and four variables of the patient's information. Symptoms of COVID-19 include a high fever, low fever, sore throat, cough, and so on, where patient metadata includes Province, county, sex, and age. The dataset contains 1254 patients with 664 recovered cases. To develop prediction models, four models are used including neural network, support vector machine, CHAID, and QUEST models. To develop prediction models, the dataset is divided into train and test datasets with splitting ratios equal to 70%, and 30%, respectively. RESULTS:The results showed that the neural network model is the most effective model in developing COVID-19 prediction with the highest performance metrics using train and test datasets. The results found that recovered cases are associated with the place of the patients mainly, province of the patient. Besides the results showed that high fever is not strongly associated with recovered cases, where cough and low fever are strongly associated with recovered cases. In addition, the country, sex, and age of the patients have higher importance than other patient's symptoms in COVID-19 development. CONCLUSIONS:The results revealed that the prediction models of the recovered COVID-19 cases can be effectively predicted using patient characteristics and symptoms, besides the neural network model is the most effective model to create a COVID -19 prediction model. Finally, the research provides empirical evidence that recovered cases of COVID-19 are closely related to patients' provinces.
Multimodal Learning and Intelligent Prediction of Symptom Development in Individual Parkinson's Patients.
Przybyszewski Andrzej W,Kon Mark,Szlufik Stanislaw,Szymanski Artur,Habela Piotr,Koziorowski Dariusz M
Sensors (Basel, Switzerland)
We still do not know how the brain and its computations are affected by nerve cell deaths and their compensatory learning processes, as these develop in neurodegenerative diseases (ND). Compensatory learning processes are ND symptoms usually observed at a point when the disease has already affected large parts of the brain. We can register symptoms of ND such as motor and/or mental disorders (dementias) and even provide symptomatic relief, though the structural effects of these are in most cases not yet understood. It is very important to obtain early diagnosis, which can provide several years in which we can monitor and partly compensate for the disease's symptoms, with the help of various therapies. In the case of Parkinson's disease (PD), in addition to classical neurological tests, measurements of eye movements are diagnostic. We have performed measurements of latency, amplitude, and duration in reflexive saccades (RS) of PD patients. We have compared the results of our measurement-based diagnoses with standard neurological ones. The purpose of our work was to classify how condition attributes predict the neurologist's diagnosis. For n = 10 patients, the patient age and parameters based on RS gave a global accuracy in predictions of neurological symptoms in individual patients of about 80%. Further, by adding three attributes partly related to patient 'well-being' scores, our prediction accuracies increased to 90%. Our predictive algorithms use rough set theory, which we have compared with other classifiers such as Naïve Bayes, Decision Trees/Tables, and Random Forests (implemented in KNIME/WEKA). We have demonstrated that RS are powerful biomarkers for assessment of symptom progression in PD.
Prediction of specific depressive symptom clusters in youth with epilepsy: The NDDI-E-Y versus Neuro-QOL SF.
Kellermann Tanja S,Mueller Martina,Carter Emma G,Brooks Byron,Smith Gigi,Kopp Olivia J,Wagner Janelle L
OBJECTIVE:Proper assessment and early identification of depressive symptoms are essential to initiate treatment and minimize the risk for poor outcomes in youth with epilepsy (YWE). The current study examined the predictive utility of the Neurological Disorders Depression Inventory-Epilepsy for Youth (NDDI-E-Y) and the Neuro-QOL Depression Short Form (Neuro-QOL SF) in explaining variance in overall depressive symptoms and specific symptom clusters on the gold standard Children's Depression Inventory-2 (CDI-2). METHODS:Cross-sectional study examining 99 YWE (female 68, mean age 14.7 years) during a routine epilepsy visit, who completed self-report measures of depressive symptoms, including the NDDI-E-Y, CDI-2, and the Neuro-QOL SF. Caregivers completed a measure of seizure severity. All sociodemographic and medical information was evaluated through electronic medical record review. RESULTS:After accounting for seizure and demographic variables, the NDDI-E-Y accounted for 45% of the variance in the CDI-2 Total score and the CDI-2 Ineffectiveness subscale. Furthermore, the NDDI-E-Y predicted CDI-2 Total scores and subscales similarly, with the exception of explaining significantly more variance in the CDI-2 Ineffectiveness subscale compared to the Negative Mood subscale. The NDDI-E-Y explained greater variance compared to Neuro-QOL SF across the Total (48% vs. 37%) and all CDI-2 subscale scores; however, the NDDI-E-Y emerged as a stronger predictor of only CDI-2 Ineffectiveness. Both the NDDI-E-Y and Neuro-QOL SF accounted for the lowest amount of variance in CDI-2 Negative Mood. Sensitivity was poor for the Neuro-QOL SF in predicting high versus low CDI-2 scores. SIGNIFICANCE:The NDDI-E-Y has strong psychometrics and can be easily integrated into routine epilepsy care for quick, brief screening of depressive symptoms in YWE.
Identifying the Symptom Severity in Obsessive-Compulsive Disorder for Classification and Prediction: An Artificial Neural Network Approach.
Shahzad Mirza Naveed,Suleman Muhammad,Ahmed Mirza Ashfaq,Riaz Amna,Fatima Khadija
The present study is aimed at identifying the most prominent determinants of OCD along with their strength to classify the OCD patients from healthy controls. The data for this cross-sectional study were collected from 200 diagnosed OCD patients and 400 healthy controls. The respondents were selected through purposive sampling and interviewed by using the Y-BOCS scale with the addition of a factor, worth of an individual in his family. The validity and reliability of data were assessed through Cronbach's alpha and confirmatory factor analysis. Artificial Neural Network (ANN) modeling was adopted to determine threatening determinants along with their strength to predict OCD in an individual. The results of ANN modeling depicted 98% accurate classification of OCD patients from healthy controls. The most contributing factors in determining the OCD patients according to normalized importance were the contamination and cleaning (100%); symmetric and perfection (72.5%); worth of an individual in the family (71.1%); aggressive, religious, and sexual obsession (50.5%); high-risk assessment (46.0%); and somatic obsessions and checking (24.0%).
Predicting Clinically Relevant Patient-Reported Symptom Improvement After Carpal Tunnel Release: A Machine Learning Approach.
Hoogendam Lisa,Bakx Jeanne A C,Souer J Sebastiaan,Slijper Harm P,Andrinopoulou Eleni-Rosalina,Selles Ruud W,
BACKGROUND:Symptom improvement is an important goal when considering surgery for carpal tunnel syndrome. There is currently no prediction model available to predict symptom improvement for patients considering a carpal tunnel release (CTR). OBJECTIVE:To predict using a model the probability of clinically relevant symptom improvement at 6 mo after CTR. METHODS:We split a cohort of 2119 patients who underwent a mini-open CTR and completed the Boston Carpal Tunnel Questionnaire preoperatively and 6 mo postoperatively into training (75%) and validation (25%) data sets. Patients who improved more than the minimal clinically important difference of 0.8 at the Boston Carpal Tunnel Questionnaire-symptom severity scale were classified as "improved." Logistic regression, random forests, and gradient boosting machines were considered to train prediction models. The best model was selected based on discriminative ability (area under the curve) and calibration in the validation data set. This model was further assessed in a holdout data set (N = 397). RESULTS:A gradient boosting machine with 5 predictors was chosen as optimal trade-off between discriminative ability and the number of predictors. In the holdout data set, this model had an area under the curve of 0.723, good calibration, sensitivity of 0.77, and specificity of 0.55. The positive predictive value was 0.50, and the negative predictive value was 0.81. CONCLUSION:We developed a prediction model for clinically relevant symptom improvement 6 mo after a CTR, which required 5 patient-reported predictors (18 questions) and has reasonable discriminative ability and good calibration. The model is available online and might help shared decision making when patients are considering a CTR.
Symptom severity prediction from neuropsychiatric clinical records: Overview of 2016 CEGS N-GRID shared tasks Track 2.
Filannino Michele,Stubbs Amber,Uzuner Özlem
Journal of biomedical informatics
The second track of the CEGS N-GRID 2016 natural language processing shared tasks focused on predicting symptom severity from neuropsychiatric clinical records. For the first time, initial psychiatric evaluation records have been collected, de-identified, annotated and shared with the scientific community. One-hundred-ten researchers organized in twenty-four teams participated in this track and submitted sixty-five system runs for evaluation. The top ten teams each achieved an inverse normalized macro-averaged mean absolute error score over 0.80. The top performing system employed an ensemble of six different machine learning-based classifiers to achieve a score 0.86. The task resulted to be generally easy with the exception of two specific classes of records: records with very few but crucial positive valence signals, and records describing patients predominantly affected by negative rather than positive valence. Those cases proved to be very challenging for most of the systems. Further research is required to consider the task solved. Overall, the results of this track demonstrate the effectiveness of data-driven approaches to the task of symptom severity classification.
Combined symptom index and second-generation multivariate biomarker test for prediction of ovarian cancer in patients with an adnexal mass.
Urban Renata R,Pappas Todd C,Bullock Rowan G,Munroe Donald G,Bonato Vinicius,Agnew Kathy,Goff Barbara A
OBJECTIVE:To assess the performance of a symptom index (SI) and multivariate biomarker panel in the identification of ovarian cancer in women presenting for surgery with an adnexal mass. STUDY DESIGN:Prospective study of patients seen at a tertiary medical center. Following consent, patients completed an SI and preoperative serum was collected for individual markers (CA 125) and a second-generation FDA-cleared biomarker test (MIA2G). Results for the SI and MIA2G were correlated with operative findings and surgical pathology. Logistic regression modeling was performed to assess the interaction of the SI with MIA2G to determine the risk of malignancy (ROM). RESULTS:Of the 218 patients enrolled, the mean age was 53.6 years (range 18-86). One-hundred and forty-seven patients (67.4%) were postmenopausal. Sixty-four patients (29.4%) had epithelial ovarian cancer or fallopian tube cancer (EOC/FTC) and 17 (7.8%) had borderline ovarian tumors. A positive SI or MIA2G correctly identified 96.1% of patients with EOC/FTC. Using logistic regression, we found that both SI and MIA2G score were significantly associated with ROM (p < 0.001). In a simulation with disease prevalence set at 5%, patients with a negative SI and a MIA2G score of 6 had a ROM of 1.8% whereas patients with the same MIA2G and positive SI had a 10.5% ROM, nearly a 6-fold higher risk. CONCLUSIONS:The combination of a patient-reported symptom index and refined biomarker panel allows for improved accuracy in the assessment for ovarian cancer in patients with an adnexal mass. This strategy could offer a personalized approach to addressing ROM to triage patients with an adnexal mass to appropriate care.
Patient-Reported Symptoms Improve Performance of Risk Prediction Models for Emergency Department Visits Among Patients With Cancer: A Population-Wide Study in Ontario Using Administrative Data.
Sutradhar Rinku,Rostami Mehdi,Barbera Lisa
Journal of pain and symptom management
CONTEXT:Prior work shows measurements of symptom severity using the Edmonton Symptom Assessment System (ESAS) which are associated with emergency department (ED) visits in patients with cancer; however, it is not known if symptom severity improves the ability to predict ED visits. OBJECTIVES:To determine whether information on symptom severity improves the ability to predict ED visits among patients with cancer. METHODS:This was a population-based study of patients who were diagnosed with cancer and had at least one ESAS assessment completed between 2007 and 2015 in Ontario, Canada. After splitting the cohort into training and test sets, two ED visit risk prediction models using logistic regression were developed on the training cohort, one without ESAS and one with ESAS. The predictive performance of each risk model was assessed on the test cohort and compared with respect to area under the curve and calibration. RESULTS:The full cohort consisted of 212,615 unique patients with a total of 1,267,294 ESAS assessments. The risk prediction model including ESAS was superior in sensitivity, specificity, accuracy, and discrimination. The area under the curve was 73.7% under the model with ESAS, whereas it was 70.1% under the model without ESAS. The model with ESAS was also better calibrated. This improvement in calibration was particularly noticeable among patients in the higher deciles of predicted risk. CONCLUSION:This study demonstrates the importance of incorporating symptom measurements when developing an ED visit risk calculator for patients with cancer. Improved predictive models for ED visits using measurements of symptom severity may serve as an important clinical tool to prompt timely interventions by the cancer care team before an ED visit is necessary.
Multifactorial prediction of depression diagnosis and symptom dimensions.
McNamara Mary E,Shumake Jason,Stewart Rochelle A,Labrada Jocelyn,Alario Alexandra,Allen John J B,Palmer Rohan,Schnyer David M,McGeary John E,Beevers Christopher G
While depression is a leading cause of disability, prior investigations of depression have been limited by studying correlates in isolation. A data-driven approach was applied to identify out-of-sample predictors of current depression from adults (N = 217) sampled on a continuum of no depression to clinical levels. The current study used elastic net regularized regression and predictors from sociodemographic, self-report, polygenic scores, resting electroencephalography, pupillometry, actigraphy, and cognitive tasks to classify individuals into currently depressed (MDE), psychiatric control (PC), and no current psychopathology (NP) groups, as well as predicting symptom severity and lifetime MDE. Cross-validated models explained 20.6% of the out-of-fold deviance for the classification of MDEs versus PC, 33.2% of the deviance for MDE versus NP, but -0.6% of the deviance between PC and NP. Additionally, predictors accounted for 25.7% of the out-of-fold variance in anhedonia severity, 65.7% of the variance in depression severity, and 12.9% of the deviance in lifetime depression (yes/no). Self-referent processing, anhedonia, and psychosocial functioning emerged as important differentiators of MDE and PC groups. Findings highlight the advantages of using psychiatric control groups to isolate factors specific to depression.
Symptom clusters and their influence on prognosis using EORTC QLQ-C15-PAL scores in terminally ill patients with cancer.
Koyama Nanako,Matsumura Chikako,Tahara Yuuna,Sako Morito,Kurosawa Hideo,Nomura Takehisa,Eguchi Yuki,Ohba Kazuki,Yano Yoshitaka
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
PURPOSE:The aims of the present study were to investigate the symptom clusters in terminally ill patients with cancer using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 15 Palliative Care (EORTC QLQ-C15-PAL), and to examine whether these symptom clusters influenced prognosis. METHODS:We analyzed data from 130 cancer patients hospitalized in the palliative care unit from June 2018 to December 2019 in an observational study. Principal component analysis was used to detect symptom clusters using the scored date of 14 items in the QLQ-C15-PAL, except for overall QOL, at the time of hospitalization. The influence of the existence of these symptom clusters and Palliative Performance Scale (PPS) on survival was analyzed by Cox proportional hazards regression analysis, and survival curves were compared between the groups with or without existing corresponding symptom clusters using the log-rank test. RESULTS:The following symptom clusters were identified: cluster 1 (pain, insomnia, emotional functioning), cluster 2 (dyspnea, appetite loss, fatigue, and nausea), and cluster 3 (physical functioning). Cronbach's alpha values for the symptom clusters ranged from 0.72 to 0.82. An increased risk of death was significantly associated with the existence of cluster 2 and poor PPS (log-rank test, p = 0.016 and p < 0.001, respectively). CONCLUSION:In terminally ill patients with cancer, three symptom clusters were detected based on QLQ-C15-PAL scores. Poor PPS and the presence of symptom cluster that includes dyspnea, appetite loss, fatigue, and nausea indicated poor prognosis.
Post-traumatic stress symptom clusters in acute whiplash associated disorder and their prediction of chronic pain-related disability.
Maujean Annick,Gullo Matthew J,Andersen Tonny Elmose,Ravn Sophie Lykkegaard,Sterling Michele
INTRODUCTION:The presence of post-traumatic stress disorder (PTSD) symptoms has been found to be associated with an increased risk of persisting neck pain and disability in motor vehicle crash (MVC) survivors with whiplash injuries. The findings are mixed as to which PTSD symptom(s) best predicts recovery in this population. OBJECTIVES:The aims were (1) to explore the factor structure of the Post-traumatic Stress Diagnostic Scale (PDS) in a sample of acute whiplash-injured individuals following a MVC and (2) to identify the PTSD-symptom clusters that best predict long-term neck pain-related disability in this population as measured by the Neck Pain Disability Index (NDI). METHODS:A sample (N = 146) of whiplash-injured individuals completed the NDI and the PDS at baseline (<1 month) and at 6 months follow-up. RESULTS:Principal component analyses generated 2 symptom clusters: re-experiencing/avoidance and hyperarousal/numbing. Nine trauma-related PTSD symptoms loaded exclusively on the re-experiencing/avoidance cluster and 7 nonspecific PTSD symptoms loaded exclusively on the hyperarousal/numbing cluster. One PTSD symptom (ie, inability to recall an important aspect of the trauma) had no salient loading on either clusters. Structural equation modelling analysis indicated that there was a significant positive relationship between the hyperarousal/numbing symptom cluster and long-term neck pain-related disability, while no significant relationship was found between the re-experiencing/avoidance symptom cluster and long-term neck pain-related disability. CONCLUSION:Given that only the hyperarousal/numbing symptom cluster predicted long-term neck pain-related disability, this finding may have implications in terms of diagnosis, assessment, and management of the psychological impact of whiplash-injured individuals following a MVC.
Predicting the course of persistent physical symptoms: Development and internal validation of prediction models for symptom severity and functional status during 2 years of follow-up.
Claassen-van Dessel Nikki,van der Wouden Johannes C,Twisk Johannes W R,Dekker Joost,van der Horst Henriëtte E
Journal of psychosomatic research
OBJECTIVE:Increased knowledge about predictors of the course of persistent physical symptoms (PPS) is needed to identify patients at risk for long-term PPS in clinical settings. Therefore, we developed prediction models for the course of PPS in terms of symptom-severity and related functional status during a 2-year follow-up period. METHODS:We used data of the PROSPECTS cohort study, consisting of 325 PPS patients from several health care settings. Symptom severity (PHQ-15), physical functioning (RAND 36 PCS) and mental functioning (RAND 36 MCS) were assessed at baseline and 6, 12 and 24 months afterwards. We applied mixed model analyses to develop prediction models for all outcomes, using all follow-up measurements. Potential predictors were based on empirical and theoretical literature and measured at baseline. RESULTS:For symptom severity, physical functioning and mental functioning we identified predictors for the adverse course of PPS included physical comorbidity, higher severity and longer duration of PPS at baseline, anxiety, catastrophizing cognitions, embarrassment and fear avoidance cognitions, avoidance or resting behaviour and neuroticism. Predictors of a favourable course included limited alcohol use, higher education, higher levels of physical and mental functioning at baseline, symptom focusing, damage cognitions and extraversion. Explained interpersonal variance for all three models varied between 70.5 and 76.0%. Performance of the models was comparable in primary and secondary/tertiary care. CONCLUSION:The presented prediction models identified several relevant demographic, medical, psychological and behavioural predictors for adverse and favourable courses of PPS. External validation of the presented models is needed prior to clinical implementation.
Modeling and prediction of clinical symptom trajectories in Alzheimer's disease using longitudinal data.
Bhagwat Nikhil,Viviano Joseph D,Voineskos Aristotle N,Chakravarty M Mallar,
PLoS computational biology
Computational models predicting symptomatic progression at the individual level can be highly beneficial for early intervention and treatment planning for Alzheimer's disease (AD). Individual prognosis is complicated by many factors including the definition of the prediction objective itself. In this work, we present a computational framework comprising machine-learning techniques for 1) modeling symptom trajectories and 2) prediction of symptom trajectories using multimodal and longitudinal data. We perform primary analyses on three cohorts from Alzheimer's Disease Neuroimaging Initiative (ADNI), and a replication analysis using subjects from Australian Imaging, Biomarker & Lifestyle Flagship Study of Ageing (AIBL). We model the prototypical symptom trajectory classes using clinical assessment scores from mini-mental state exam (MMSE) and Alzheimer's Disease Assessment Scale (ADAS-13) at nine timepoints spanned over six years based on a hierarchical clustering approach. Subsequently we predict these trajectory classes for a given subject using magnetic resonance (MR) imaging, genetic, and clinical variables from two timepoints (baseline + follow-up). For prediction, we present a longitudinal Siamese neural-network (LSN) with novel architectural modules for combining multimodal data from two timepoints. The trajectory modeling yields two (stable and decline) and three (stable, slow-decline, fast-decline) trajectory classes for MMSE and ADAS-13 assessments, respectively. For the predictive tasks, LSN offers highly accurate performance with 0.900 accuracy and 0.968 AUC for binary MMSE task and 0.760 accuracy for 3-way ADAS-13 task on ADNI datasets, as well as, 0.724 accuracy and 0.883 AUC for binary MMSE task on replication AIBL dataset.
Prediction of breast cancer-related outcomes with the Edmonton Symptom Assessment Scale: A literature review.
Milton Lauren,Behroozian Tara,Coburn Natalie,Trudeau Maureen,Razvi Yasmeen,McKenzie Erin,Karam Irene,Lam Henry,Chow Edward
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
PURPOSE:The Edmonton Symptom Assessment Scale (ESAS) is a validated tool used in patients with varied cancer diagnoses to measure patient symptoms. The present manuscript will review the literature assessing the ability of the ESAS to predict patient-related outcomes in breast cancer patients. METHODS:A literature search was conducted of Cochrane Central Register of Controlled Trials databases, Ovid MEDLINE, and Embase for English articles that investigated the use of predictive modelling with the ESAS in the breast cancer population. Study type, publication year, sample size, patient demographics, predicted outcomes, and strongest predictive factors/symptoms were summarized for each study. RESULTS:A total of nine articles were included in this review. Five articles used the ESAS in predictive models to determine patient time to death. ESAS was also used to predict emergency department visits, determine symptoms associated with decreased quality of life, and generate a Health Utility Score. Lack of appetite was the most common ESAS symptom, as it was reported in five studies to be associated with decreased survival. In four of the nine articles, an additional survey investigating physical functioning was used in combination with ESAS to strengthen the predictive models. CONCLUSIONS:Included studies support the use of ESAS in predictive models, particularly for predicting survival. Using the ESAS as a predictive tool allows for more accurate time to death predictions, potentially improving symptom management and preventing overtreatment of palliative patients near the end of life.
Intrinsic Connectivity Patterns of Task-Defined Brain Networks Allow Individual Prediction of Cognitive Symptom Dimension of Schizophrenia and Are Linked to Molecular Architecture.
Chen Ji,Müller Veronika I,Dukart Juergen,Hoffstaedter Felix,Baker Justin T,Holmes Avram J,Vatansever Deniz,Nickl-Jockschat Thomas,Liu Xiaojin,Derntl Birgit,Kogler Lydia,Jardri Renaud,Gruber Oliver,Aleman André,Sommer Iris E,Eickhoff Simon B,Patil Kaustubh R
BACKGROUND:Despite the marked interindividual variability in the clinical presentation of schizophrenia, the extent to which individual dimensions of psychopathology relate to the functional variability in brain networks among patients remains unclear. Here, we address this question using network-based predictive modeling of individual psychopathology along 4 data-driven symptom dimensions. Follow-up analyses assess the molecular underpinnings of predictive networks by relating them to neurotransmitter-receptor distribution patterns. METHODS:We investigated resting-state functional magnetic resonance imaging data from 147 patients with schizophrenia recruited at 7 sites. Individual expression along negative, positive, affective, and cognitive symptom dimensions was predicted using a relevance vector machine based on functional connectivity within 17 meta-analytic task networks following repeated 10-fold cross-validation and leave-one-site-out analyses. Results were validated in an independent sample. Networks robustly predicting individual symptom dimensions were spatially correlated with density maps of 9 receptors/transporters from prior molecular imaging in healthy populations. RESULTS:Tenfold and leave-one-site-out analyses revealed 5 predictive network-symptom associations. Connectivity within theory of mind, cognitive reappraisal, and mirror neuron networks predicted negative, positive, and affective symptom dimensions, respectively. Cognitive dimension was predicted by theory of mind and socioaffective default networks. Importantly, these predictions generalized to the independent sample. Intriguingly, these two networks were positively associated with D receptor and serotonin reuptake transporter densities as well as dopamine synthesis capacity. CONCLUSIONS:We revealed a robust association between intrinsic functional connectivity within networks for socioaffective processes and the cognitive dimension of psychopathology. By investigating the molecular architecture, this work links dopaminergic and serotonergic systems with the functional topography of brain networks underlying cognitive symptoms in schizophrenia.
A Prediction Model to Prioritize Individuals for a SARS-CoV-2 Test Built from National Symptom Surveys.
Med (New York, N.Y.)
BACKGROUND:The gold standard for COVID-19 diagnosis is detection of viral RNA through PCR. Due to global limitations in testing capacity, effective prioritization of individuals for testing is essential. METHODS:We devised a model estimating the probability of an individual to test positive for COVID-19 based on answers to 9 simple questions that have been associated with SARS-CoV-2 infection. Our model was devised from a subsample of a national symptom survey that was answered over 2 million times in Israel in its first 2 months and a targeted survey distributed to all residents of several cities in Israel. Overall, 43,752 adults were included, from which 498 self-reported as being COVID-19 positive. FINDINGS:Our model was validated on a held-out set of individuals from Israel where it achieved an auROC of 0.737 (CI: 0.712-0.759) and auPR of 0.144 (CI: 0.119-0.177) and demonstrated its applicability outside of Israel in an independently collected symptom survey dataset from the US, UK, and Sweden. Our analyses revealed interactions between several symptoms and age, suggesting variation in the clinical manifestation of the disease in different age groups. CONCLUSIONS:Our tool can be used online and without exposure to suspected patients, thus suggesting worldwide utility in combating COVID-19 by better directing the limited testing resources through prioritization of individuals for testing, thereby increasing the rate at which positive individuals can be identified. Moreover, individuals at high risk for a positive test result can be isolated prior to testing. FUNDING:E.S. is supported by the Crown Human Genome Center, Larson Charitable Foundation New Scientist Fund, Else Kroener Fresenius Foundation, White Rose International Foundation, Ben B. and Joyce E. Eisenberg Foundation, Nissenbaum Family, Marcos Pinheiro de Andrade and Vanessa Buchheim, Lady Michelle Michels, and Aliza Moussaieff and grants funded by the Minerva foundation with funding from the Federal German Ministry for Education and Research and by the European Research Council and the Israel Science Foundation. H.R. is supported by the Israeli Council for Higher Education (CHE) via the Weizmann Data Science Research Center and by a research grant from Madame Olga Klein - Astrachan.
Symptom clusters in COVID-19: A potential clinical prediction tool from the COVID Symptom Study app.
Sudre Carole H,Lee Karla A,Lochlainn Mary Ni,Varsavsky Thomas,Murray Benjamin,Graham Mark S,Menni Cristina,Modat Marc,Bowyer Ruth C E,Nguyen Long H,Drew David A,Joshi Amit D,Ma Wenjie,Guo Chuan-Guo,Lo Chun-Han,Ganesh Sajaysurya,Buwe Abubakar,Pujol Joan Capdevila,du Cadet Julien Lavigne,Visconti Alessia,Freidin Maxim B,El-Sayed Moustafa Julia S,Falchi Mario,Davies Richard,Gomez Maria F,Fall Tove,Cardoso M Jorge,Wolf Jonathan,Franks Paul W,Chan Andrew T,Spector Tim D,Steves Claire J,Ourselin Sébastien
As no one symptom can predict disease severity or the need for dedicated medical support in coronavirus disease 2019 (COVID-19), we asked whether documenting symptom time series over the first few days informs outcome. Unsupervised time series clustering over symptom presentation was performed on data collected from a training dataset of completed cases enlisted early from the COVID Symptom Study Smartphone application, yielding six distinct symptom presentations. Clustering was validated on an independent replication dataset between 1 and 28 May 2020. Using the first 5 days of symptom logging, the ROC-AUC (receiver operating characteristic - area under the curve) of need for respiratory support was 78.8%, substantially outperforming personal characteristics alone (ROC-AUC 69.5%). Such an approach could be used to monitor at-risk patients and predict medical resource requirements days before they are required.
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.
TSMDA: Target and symptom-based computational model for miRNA-disease-association prediction.
Uthayopas Korawich,de Sá Alex G C,Alavi Azadeh,Pires Douglas E V,Ascher David B
Molecular therapy. Nucleic acids
The emergence of high-throughput sequencing techniques has revealed a primary role of microRNAs (miRNAs) in a wide range of diseases, including cancers and neurodegenerative disorders. Understanding novel relationships between miRNAs and diseases can potentially unveil complex pathogenesis mechanisms, leading to effective diagnosis and treatment. The investigation of novel miRNA-disease associations, however, is currently costly and time consuming. Over the years, several computational models have been proposed to prioritize potential miRNA-disease associations, but with limited usability or predictive capability. In order to fill this gap, we introduce TSMDA, a novel machine-learning method that leverages target and symptom information and negative sample selection to predict miRNA-disease association. TSMDA significantly outperforms similar methods, achieving an area under the receiver operating characteristic (ROC) curve (AUC) of 0.989 and 0.982 under 5-fold cross-validation and blind test, respectively. We also demonstrate the capability of the method to uncover potential miRNA-disease associations in breast, prostate, and lung cancers, as case studies. We believe TSMDA will be an invaluable tool for the community to explore and prioritize potentially new miRNA-disease associations for further experimental characterization. The method was made available as a freely accessible and user-friendly web interface at http://biosig.unimelb.edu.au/tsmda/.
Adolescent pain: appraisal of the construct and trajectory prediction-by-symptom between age 12 and 17 years in a Canadian twin birth cohort.
ABSTRACT:Adolescent pain is common and continues into adulthood, leading to negative long-term outcomes including substance-related morbidity: an empirical definition of its construct may inform the early detection of persistent pain trajectories. These secondary analyses of a classical twin study assessed whether headaches, back pains, abdominal pain, chest pains, stabbing/throbbing pain, and gastric pain/nausea, measured in 501 pairs across 5 waves between age 12 and 17 years, fit a unitary construct or constitute independent manifestations. We then assessed which symptoms were associated with a steady, "frequent pain" trajectory that is associated with risk for early opioid prescriptions. Item response theory results indicated that all 6 pain symptoms index a unitary construct. Binary logistic regressions identified "back pain" as the only symptom consistently associated with membership in the "frequent adolescent pain" trajectory (odds ratio: 1.66-3.38) at all 5 measurement waves. Receiver operating characteristic analyses computed the discriminating power of symptoms to determine participants' membership into the "frequent" trajectory: they yielded acceptable (0.7-0.8) to excellent (0.8-0.9) area under the curve values for all 6 symptoms. The highest area under the curve was attained by "back pain" at age 14 years (0.835); for multiple cut-off thresholds of symptom frequency, "back pain" showed good sensitivity/false alarm probability trade-offs, predominantly in the 13 to 15 years age range, to predict the "frequent pain" trajectory. These data support a unitary conceptualization and assessment of adolescent pain, which is advantageous for epidemiological, clinical, and translational purposes. Persistent back pain constitutes a sensitive indicator of a steady trajectory of adolescent pain.
Prediction of Early Symptom Remission in Two Independent Samples of First-Episode Psychosis Patients Using Machine Learning.
BACKGROUND:Validated clinical prediction models of short-term remission in psychosis are lacking. Our aim was to develop a clinical prediction model aimed at predicting 4-6-week remission following a first episode of psychosis. METHOD:Baseline clinical data from the Athens First Episode Research Study was used to develop a Support Vector Machine prediction model of 4-week symptom remission in first-episode psychosis patients using repeated nested cross-validation. This model was further tested to predict 6-week remission in a sample of two independent, consecutive Danish first-episode cohorts. RESULTS:Of the 179 participants in Athens, 120 were male with an average age of 25.8 years and average duration of untreated psychosis of 32.8 weeks. 62.9% were antipsychotic-naïve. Fifty-seven percent attained remission after 4 weeks. In the Danish cohort, 31% attained remission. Eleven clinical scale items were selected in the Athens 4-week remission cohort. These included the Duration of Untreated Psychosis, Personal and Social Performance Scale, Global Assessment of Functioning and eight items from the Positive and Negative Syndrome Scale. This model significantly predicted 4-week remission status (area under the receiver operator characteristic curve (ROC-AUC) = 71.45, P < .0001). It also predicted 6-week remission status in the Danish cohort (ROC-AUC = 67.74, P < .0001), demonstrating reliability. CONCLUSIONS:Using items from common and validated clinical scales, our model significantly predicted early remission in patients with first-episode psychosis. Although replicated in an independent cohort, forward testing between machine learning models and clinicians' assessment should be undertaken to evaluate the possible utility as a routine clinical tool.