Journal of Medical Systems最新文献

{"title":"Dissecting HealthBench: Disease Spectrum, Clinical Diversity, and Data Insights from Multi-Turn Clinical AI Evaluation Benchmark.","authors":"Jialin Liu, Siru Liu","doi":"10.1007/s10916-025-02232-w","DOIUrl":"10.1007/s10916-025-02232-w","url":null,"abstract":"<p><p>HealthBench is an open-source, large-scale benchmark consisting of 5,000 multi-turn clinical conversations evaluated against 48,562 criteria developed by clinicians. Recognized as a significant advancement in assessing realistic artificial intelligence (AI) models, HealthBench deserves further exploration. In this article, we systematically analyze the benchmark's disease spectrum, diagnostic and therapeutic focuses, and demographic diversity. We evaluate its representativeness and strengths, as well as the essential limitations that AI researchers and clinicians should consider when using it for realistic model evaluations.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"100"},"PeriodicalIF":5.7,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12304011/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144731835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantum Oncology: The Applications of Quantum Computing in Cancer Research.","authors":"Jiaran Li, Li Li, Ning Liu, Fuhao Xu, Tao Hu, Shuanghu Yuan","doi":"10.1007/s10916-025-02215-x","DOIUrl":"https://doi.org/10.1007/s10916-025-02215-x","url":null,"abstract":"<p><p>A global technological race is underway to develop increasingly powerful and precise quantum computers. As a transformative computing paradigm, quantum computing offers the potential for exponentially accelerating specific algorithms, thereby providing the necessary computational power to process vast amounts of data. In light of the challenges classical computing faces with the complexity and volume of oncology data, we introduce the concept of \"quantum oncology\" and explore its potential applications throughout the cancer care continuum. Additionally, we address several challenges and potential solutions for integrating quantum computing into oncology research. By illuminating these issues, we aim to deepen the understanding of quantum computing's potential in oncology and advocate for multidisciplinary collaboration to propel the advancement of precision oncology. CLINICAL TRIAL NUMBER: Not applicable.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"99"},"PeriodicalIF":3.5,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144690541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Why Clinical Trials Will Fail to Ensure Safe AI.","authors":"David P W Rastall, Mohamed Rehman","doi":"10.1007/s10916-025-02231-x","DOIUrl":"10.1007/s10916-025-02231-x","url":null,"abstract":"<p><p>Recent reports have raised concerns about emergent behaviors in next-generation artificial intelligence (AI) models. These systems have been documented selectively adapting their behaviors during testing to falsify experimental outcomes and bypass regulatory oversight. This phenomenon-alignment faking-represents a fundamental challenge to medical AI safety. Regulatory strategies have largely adapted established protocols like clinical trials and medical device approval frameworks, but for next-generation AI these approaches may fail. This paper introduces alignment faking to a medical audience and critically evaluates how current regulatory tools are inadequate for advanced AI systems. We propose continuous logging through \"AI SOAP notes\" as a first step toward transparent and accountable AI functionality in clinical settings.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"98"},"PeriodicalIF":3.5,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144649703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolving Blood Pressure Estimation: From Feature Analysis to Image-Based Deep Learning Models.","authors":"Vishal Singh Roha, Rahul Ranjan, Mehmet Rasit Yuce","doi":"10.1007/s10916-025-02228-6","DOIUrl":"10.1007/s10916-025-02228-6","url":null,"abstract":"<p><p>Traditional cuffless blood pressure (BP) estimation methods often require collecting physiological signals, such as electrocardiogram (ECG) and photoplethysmography (PPG), from two distinct body sites to compute metrics like pulse transit time (PTT) or pulse arrival time (PAT). While these metrics strongly correlate with BP, their reliance on multiple signal sources and susceptibility to noise from modern wearable devices present significant challenges. Addressing these limitations, we propose an innovative framework that requires only PPG signals from a single body site, leveraging advancements in artificial intelligence and computer vision. Our approach employs images of PPG signals, along with their first (vPPG) and second (aPPG) derivatives, for enhanced BP estimation. ResNet-50 is utilized to extract features and identify regions within the PPG, vPPG, and aPPG images that correlate strongly with BP. These features are further refined using multi-head cross-attention (MHCA) mechanism, enabling efficient information exchange across the modalities derived from ResNet-50 outputs, thereby improving estimation accuracy. The framework is validated on three distinct datasets, demonstrating superior performance compared to traditional PAT and PTT-based methods. Furthermore, it adheres to stringent medical standards, such as those defined by the Association for the Advancement of Medical Instrumentation (AAMI) and the British Hypertension Society (BHS), ensuring clinical reliability. By reducing the need for multiple signal sources and incorporating cutting-edge AI techniques, this framework represents a significant advancement in non-invasive BP monitoring, offering a more practical and accurate alternative to traditional methodologies.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"97"},"PeriodicalIF":3.5,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12241270/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144591464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using Performance Frontiers To Evaluate Non-OR Anesthesia (NORA) Efficiency.","authors":"Justin S Routman, Erik J Zhang, Jonathan D Blocker, Juhan Paiste, Mitchell H Tsai","doi":"10.1007/s10916-025-02229-5","DOIUrl":"10.1007/s10916-025-02229-5","url":null,"abstract":"<p><strong>Introduction: </strong>In high-cost, high-revenue operating room (OR) suites, dashboards displaying key performance indicators are commonplace to optimize efficiency. Given the significant successes attained, further gains may risk compromising safety. In contrast, challenges unique to non-operating room anesthesia (NORA) sites have hindered operational efficiency. Existing productivity evaluation frameworks often fall short in guiding strategic and tactical improvements in NORA delivery. Performance frontiers have proven effective in evaluating OR systems, but their application to NORA remains unexplored. This study applies performance frontiers to assess NORA site efficiency and formulates potential operational strategies.</p><p><strong>Methods: </strong>We evaluated anesthesia billing records at our primary hospital from 1 April 2022 to 30 March 2023. Cases from operating room and NORA locations were included, except for sites with irregular volume or financial arrangements. We included only non-holiday weekdays, defining NORA block time as 7 AM to 5 PM. For each room, we calculated under-utilized (time with no anesthesia billing) and over-utilized minutes (time billed outside of NORA block hours). Data for each location were plotted as rolling 4-week sums, normalized to scheduled NORA block time. Performance frontiers were then developed and plotted.</p><p><strong>Results: </strong>Over 246 non-holiday weekdays, 42,424 cases had billable minutes during NORA block time, comprising 20,003 (47.2%) NORA cases and 22,421 (52.8%) OR cases. Performance frontiers revealed significant variability, with nonparametric tests confirming statistical significance and non-equivalence.</p><p><strong>Discussion: </strong>Performance frontiers reveal substantial efficiency variability across NORA sites, underscoring the need for targeted interventions. Some sites matched OR efficiency levels, while others showed substantial differences, particularly those with high variability and urgency. Efficient sites can leverage performance frontiers to optimize resource allocation, while inefficient locations may benefit from a shared anesthesia resource pool for real-time resource allocation. Performance frontiers provide a novel approach for operational leaders to make more effective strategic decisions.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"96"},"PeriodicalIF":3.5,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12238198/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144591465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unveiling Bias in Distilled Artificial Intelligence Models: Ethical and Clinical Impacts on Decision-Making and Medical Auditing.","authors":"Gerson Hiroshi Yoshinari Júnior, Luciano Magalhães Vitorino","doi":"10.1007/s10916-025-02230-y","DOIUrl":"https://doi.org/10.1007/s10916-025-02230-y","url":null,"abstract":"","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"95"},"PeriodicalIF":3.5,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144584152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the Performance of ChatGPT on Board-Style Examination Questions in Ophthalmology: A Meta-Analysis.","authors":"Jiawen Wei, Xiaoyan Wang, Mingxue Huang, Yanwu Xu, Weihua Yang","doi":"10.1007/s10916-025-02227-7","DOIUrl":"10.1007/s10916-025-02227-7","url":null,"abstract":"<p><p>To review empirical research on ChatGPT's accuracy in answering ophthalmology board-style examination questions up to March 2025 and to analyze the effects of GPT versions, question types, language differences, and ophthalmology topics on accuracy. A search was conducted in PubMed, Web of Science, Embase, Scopus, and the Cochrane Library in March 2025. Two authors extracted data and independently assessed study quality. Accuracy rates were calculated with Stata 17.0. GPT-4 had an integrated accuracy of 73%, higher than GPT-3.5's 54%. It scored 77% in text and 55% in image tasks. GPT-4's accuracy was 73% in English-speaking countries and 71% in non-English ones. In ophthalmology, General Medicine achieved the highest accuracy (80%), while Clinical Optics had the lowest performance (55%). GPT-4 outperforms GPT-3.5, but its image processing capability needs further validation. Performance varies by language and topic, suggesting the need for more research on cross-linguistic efficacy and error analysis.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"94"},"PeriodicalIF":3.5,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144564863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Simulation-Based Approach for Inpatient Capacity Management at a Hospital Dedicated for Cancer Treatment.","authors":"Anup C Mokashi, Ginger J Gardner, Adam D Klotz, Jacquelyn J Burns, Jeena L Velzen","doi":"10.1007/s10916-025-02206-y","DOIUrl":"https://doi.org/10.1007/s10916-025-02206-y","url":null,"abstract":"<p><p>This paper describes the development and application of an analytical solution to assist with inpatient flow and capacity management at Memorial Sloan Kettering Cancer Center (MSKCC) in New York City. We present a discrete-event simulation model that captures several key aspects of the complex patient flow patterns at MSKCC in the inpatient setting. The model captures the variation in admission patterns based on various patient cohorts and admit locations. The model also accounts for the variability in specialized care needs for distinct patient cohorts using categorical distributions. Durations for various patient flow states from admission till discharge are modeled as probability distributions. Key patient-and resource attributes are also incorporated to accurately capture the constraints affecting resource allocation. A comprehensive set of output metrics is used to validate the model, and to compare alternative scenarios. We present results for a scenario that tests the impact of resource allocation changes aimed at consolidating patients on certain floors based on the hospital department tasked with their inpatient care. Outputs for the scenario are compared with baseline using the following output metrics: mean bed utilization by floor, mean admit boarding times by service, proportion of home floor admissions by service, and wait times for step-down care beds. Our results show an estimated reduction in average admit wait times by 30 minutes or more across 4 inpatient services (an annual reduction of <math><mo>∼</mo></math> 116 days), with a neutral impact on other output metrics. The analysis from the scenario was utilized by hospital leadership to implement actual bed allocation changes in the hospital. The model demonstrates a structured analytical approach to evaluate the impact of strategic or tactical changes prior to implementing them in practice, specifically in an inpatient setting. It also provides the flexibility to design and test a wide variety of scenarios, and has proved its utility as a decision support tool that can be leveraged periodically by leadership at MSKCC.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"93"},"PeriodicalIF":3.5,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144528379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Addressing Model Inconsistencies and Latency Challenges in LLM-Driven Emergency Medical Documentation: A Commentary.","authors":"Zekai Yu, Siyi Liu","doi":"10.1007/s10916-025-02226-8","DOIUrl":"https://doi.org/10.1007/s10916-025-02226-8","url":null,"abstract":"","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"92"},"PeriodicalIF":3.5,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144505998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Epilepsy Prediction via Time-Frequency Features and Multi-Scale Hybrid Neural Networks.","authors":"Wenwen Chang, Bingyang Ji, Dandan Li, Lei Zhen, Yaxuan Wei, Xuan Liu, Guanghui Yan","doi":"10.1007/s10916-025-02224-w","DOIUrl":"https://doi.org/10.1007/s10916-025-02224-w","url":null,"abstract":"<p><p>The prediction of epileptic seizures heavily depends on the precise embedding and classification of complex, multi-dimensional electroencephalogram (EEG) signals. Due to individual variability and the dynamic non-linear nature of EEG signals, extracting highly discriminative spatiotemporal features is a core challenge in this field. In this study, to address this issue, we proposed a novel architecture based on the Epilepsy Prediction using Multi-Scale Hybrid Neural Network (EPM-HNN), which integrates adaptive channel weighting, multi-scale spatial feature extraction, and bidirectional temporal dependency modeling. Specifically, we incorporated a sliding window mechanism with spatiotemporal resolution into the feature extraction process, enhancing the model's sensitivity to neural dynamics across frequency bands and improving its ability to capture micro-patterns. We used the Res2Net-50 multi-scale feature extractor to enhance the convolutional neural network's capacity to process complex local micro-features, such as polyspike-and-slow-wave complexes. Additionally, we introduced Squeeze-and-Excitation Networks (SENet) to adaptively capture potential effective features between different EEG channels. This dynamic weighting mechanism based on adaptive attention demonstrates strong robustness and high generalization across individual subject data. Furthermore, we proposed a non-single-subject, non-specific cross-subject training and testing method, demonstrating its ability to combat overfitting when addressing differences in data distribution. Experiments on the CHB-MIT scalp EEG dataset achieved an overall prediction accuracy of 97.7%, validating the effectiveness of the proposed EPM-HNN architecture.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"90"},"PeriodicalIF":3.5,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144484681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}