AI总结:根据提供的论文列表,这些研究主要围绕医学影像学、放射学以及大型语言模型(LLM)在医疗领域的应用展开。以下是整体概要:---上述论文集中探讨了放射学领域中患者风险分层、卵巢和附件病变的诊断与管理、以及大型语言模型(如O-RADS评分系统)的应用价值。具体而言,研究内容涵盖了以下关键主题: 1. **患者风险分层优化**:通过对比不同版本的O-RADS评分系统(2022版与2019版),评估其在临床实践中的有效性,旨在提高对卵巢及附件病变患者的个性化管理能力。 2. **放射学诊断技术**:深入分析了基于MRI报告的自动化计算方法,用于提升卵巢-附件病变报告和数据系统的准确性,并探索其对手术切除率的影响。 3. **大型语言模型的应用潜力**:研究了开放源代码的大型语言模型在放射学领域的潜在用途,包括自然语言处理技术如何辅助医学影像解读、报告生成及数据标准化。此外,还讨论了模型性能评估方法及其面临的挑战。 总体来看,这些论文强调了将人工智能技术与传统放射学工具相结合的重要性,以改善诊断精度、提高工作效率并优化患者结局。研究成果为未来放射学领域的发展提供了重要参考依据。--- 希望这个摘要能够满足您的需求!
Scan me!
共6篇 平均IF=15.2 (15.2-15.2)更多分析
  • 1区Q1影响因子: 15.2
    1. Accuracy of Large Language Model-based Automatic Calculation of Ovarian-Adnexal Reporting and Data System MRI Scores from Pelvic MRI Reports.
    期刊:Radiology
    日期:2025-04-01
    DOI :10.1148/radiol.241554
    Background Ovarian-Adnexal Reporting and Data System (O-RADS) for MRI helps assign malignancy risk, but radiologist adoption is inconsistent. Automatic assignment of O-RADS scores from reports could increase adoption and accuracy. Purpose To evaluate the accuracy of large language models (LLMs), after strategic optimization, for automatically calculating O-RADS scores from reports. Materials and Methods This retrospective single-center study from a large quaternary care cancer center included consecutive gadolinium chelate-enhanced pelvic MRI reports with at least one assigned O-RADS score from July 2021 to October 2023. Reports from January 2018 to October 2019 (before O-RADS MRI implementation) were randomly selected for additional testing. Reference standard O-RADS scores were determined by radiologists interpreting reports. After prompt optimization using a subset of reports, two LLM-based strategies were evaluated: few-shot learning with GPT-4 (version 0613; OpenAI) prompted with O-RADS rules ("LLM only") and a hybrid strategy leveraging GPT-4 to classify features fed into a deterministic formula ("hybrid"). Accuracy of each model and originally reported scores were calculated and compared using the McNemar test. Results A total of 284 reports from 284 female patients (mean age, 53.2 years ± 16.3 [SD]) with 372 adnexal lesions were included: 10 reports in the training set (16 lesions), 134 reports in the internal test set 1 (173 lesions; 158 O-RADS assigned), and 140 reports in internal test set 2 (183 lesions). For assigning O-RADS MRI scores, the hybrid model accuracy (97%; 168 of 173) outperformed LLM-only model (90%; 155 of 173; = .006). For lesions with an originally reported O-RADS score, hybrid model accuracy exceeded that of reporting radiologists (97% [153 of 158] vs 88% [139 of 158]; = .004). Hybrid model also outperformed LLM-only model for 183 lesions from before O-RADS implementation (95% [173 of 183] vs 87% [159 of 183], respectively; = .01). Conclusion A hybrid LLM-based application, combining LLM feature classification with deterministic elements, accurately assigned O-RADS MRI scores from report descriptions, exceeding both an LLM-only strategy and the original reporting radiologist. © RSNA, 2025
  • 1区Q1影响因子: 15.2
    2. O-RADS US Version 2022 Improves Patient Risk Stratification When Compared with O-RADS US Version 2019.
    期刊:Radiology
    日期:2025-03-01
    DOI :10.1148/radiol.242200
  • 1区Q1影响因子: 15.2
    3. Beyond Proprietary Models: The Potential of Open-Source Large Language Models in Radiology.
    期刊:Radiology
    日期:2025-04-01
    DOI :10.1148/radiol.242454
  • 1区Q1影响因子: 15.2
    4. The Radiologist at the Forefront of Management of Ovarian and Adnexal Lesions.
    期刊:Radiology
    日期:2024-10-01
    DOI :10.1148/radiol.242545
  • 1区Q1影响因子: 15.2
    5. The Ovarian-Adnexal Reporting and Data System (O-RADS) US Score Effect on Surgical Resection Rate.
    期刊:Radiology
    日期:2024-10-01
    DOI :10.1148/radiol.240044
    Background The Ovarian-Adnexal Imaging Reporting and Data System (O-RADS) US risk score can be used to accurately stratify ovarian lesions based on morphologic characteristics. However, there are no large multicenter studies assessing the potential impact of using O-RADS US version 2022 risk score in patients referred for surgery for an ovarian or adnexal lesion. Purpose To retrospectively determine the proportion of patients with ovarian or adnexal lesions without acute symptoms who may have been managed conservatively by using the O-RADS US version 2022 risk score. Materials and Methods This multicenter retrospective study included patients with ovarian cystic lesions and nonacute symptoms who underwent surgical resection after US before the introduction of O-RADS US between January 2011 and December 2014. Investigators blinded to the final diagnoses recorded lesion imaging features and O-RADS US risk scores. The frequency of malignancy and the diagnostic performance of the risk score were calculated. The Mann-Whitney test and Fisher exact test were performed, with < .05 indicating a statistically significant difference. Results A total of 377 patients with surgically resected lesions were included. Among the resected lesions, 42% (157 of 377) were assigned an O-RADS US risk score of 2. Of the O-RADS US 2 lesions, 54% (86 of 157) were nonneoplastic, 45% (70 of 157) were dermoids or other benign tumors, and less than 1% (one of 157) were malignant. Using O-RADS US 4 as the optimal threshold for malignancy prediction yielded a 94% (68 of 72) sensitivity, 64% (195 of 305) specificity, 38% (68 of 178) positive predictive value, and 98% (195 of 199) negative predictive value. Conclusion In patients without acute symptoms who underwent surgery for ovarian and adnexal lesions before the O-RADS US risk score was published, nearly half (42%) of surgically resected lesions retrospectively met the O-RADS US 2 version 2022 criteria. In these patients, imaging follow-up or conservative management could have been offered. © RSNA, 2024 See also the editorial by Fournier in this issue.
  • 1区Q1影响因子: 15.2
    6. Methodological Challenges in Evaluating Large Language Models in Radiology.
    期刊:Radiology
    日期:2024-12-01
    DOI :10.1148/radiol.241711
logo logo
$!{favoriteKeywords}