PLOS digital health最新文献

{"title":"Monitoring influenza-like symptoms in the UK through participatory surveillance: Insights from FluSurvey over two winter seasons (2023-24 and 2024-25).","authors":"Rebecca E Green, Jonathon Mellor, Christopher Rawlinson, Elizabeth Waller, Nurin Abdul Aziz, Conall H Watson, Gavin Dabrera","doi":"10.1371/journal.pdig.0001714","DOIUrl":"10.1371/journal.pdig.0001714","url":null,"abstract":"<p><p>FluSurvey is a participatory surveillance system used to monitor trends in influenza and other respiratory viruses through weekly symptom surveys among the UK population. We aimed to characterise the wider impact of \"influenza-like illnesses\" (ILI) among FluSurvey participants and assess correlations of ILI with other established influenza surveillance systems. We included data reported by FluSurvey participants over the 2023-24 and 2024-25 winter seasons. Using weekly symptoms surveys, we derived ILI episodes and estimated the proportion reporting healthcare service use, medication use, impact on daily life, absenteeism and use of tests. We applied existing methodologies (omitting first report and weighting to the age-sex structure of England) and assessed cross-correlations of weekly FluSurvey ILI rates with the national surveillance of GP ILI consultations, influenza hospital admissions, and influenza PCR test positivity at time lags of up to +/-2 weeks. There were 3057 participants over two winter seasons (N2023-24 = 2540, 63% female, mean age 60 years; N2024-25 = 2273, 64% female, mean age 61 years). Of 1868 ILI episodes, only a minority contacted healthcare services (14%, most frequently visiting the GP). A large proportion of episodes reported medication use (89%), impact on daily life (75%) and missing school or work (47%). Testing frequency notably declined in the 2024-25 season compared to 2023-24. FluSurvey ILI rates were strongly correlated with other influenza surveillance, predominantly leading GP ILI consultations (max r = 0.73), coinciding with influenza hospital admissions (max r = 0.88) and lagging influenza test positivity (max r = 0.88). The majority of ILI reported to FluSurvey do not contact healthcare due to symptoms but experienced wider impacts on daily life. FluSurvey ILI corresponds well with other national influenza surveillance and provides broader context on community illness, supplementing the monitoring of influenza activity for public health response.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001714"},"PeriodicalIF":7.7,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13649004/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139768","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnostic performance of the ECG-SMART AI score for detecting angiographically significant coronary artery disease in patients undergoing elective coronary angiography.","authors":"Dania A Bani Hani, Jafar A Alshraideh, Hamza Alduraidi, Akram Saleh, Nathan T Riek, Karam Daoud, Rui Qi Ji, Samir Saba, Clifton W Callaway, Salah Al-Zaiti","doi":"10.1371/journal.pdig.0001701","DOIUrl":"10.1371/journal.pdig.0001701","url":null,"abstract":"<p><p>The 12-lead ECG is central to the evaluation of patients with suspected coronary artery disease (CAD) but has limited diagnostic accuracy for identifying significant obstructive disease prior to elective coronary angiography. We previously developed an artificial intelligence-enhanced ECG (AI-ECG) model for detecting acute coronary occlusion. Whether this model generalizes to identifying angiographically significant CAD in symptomatic patients undergoing elective evaluation remains unknown. In this prospective cohort study, symptomatic outpatients undergoing elective coronary angiography at a tertiary center received standard 12-lead ECGs before the procedure. ECGs were analyzed offline using a previously validated AI-ECG model originally developed for acute coronary occlusion. Patients were categorized as low or intermediate-to-high risk using predefined thresholds. The primary outcome was angiographically significant CAD (≥70% stenosis in a major epicardial vessel or ≥50% stenosis of the left main coronary artery). Among 363 patients (age 58.9 ± 11 years; 55.1% male), 36.1% had angiographically significant CAD. The AI-ECG classified 56% as low risk and 44% as intermediate-to-high risk. After adjustment for conventional clinical characteristics, intermediate-to-high risk remained independently associated with significant CAD (OR 3.12, 95% CI 1.88-5.20; p < 0.001). Model discrimination was good (AUROC 0.79, 95% CI 0.74-0.84), with 60% precision at 80% recall. In summary, the ECG-SMART AI score demonstrated good diagnostic performance for identifying angiographically significant CAD in symptomatic patients undergoing elective coronary angiography. These findings support the generalizability of a previously developed AI-ECG model beyond acute coronary occlusion and suggest that AI-enhanced ECG analysis may complement existing clinical assessment by providing additional noninvasive information to refine pretest probability during elective evaluation.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001701"},"PeriodicalIF":7.7,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13649019/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Screening for patients at risk for cardiac amyloidosis via electronic health records: A multicenter machine learning development and validation study.","authors":"Clemens P Spielvogel, David Kersting, David Haberl, Maximilian Autherith, Laurenz Hauptmann, Josef Yu, Juliane Hennenberg, Kilian Kluge, Kim Moon, Stephan Settelmeier, Jing Ning, Katarina Kumpf, Markus Köfler, Felix Hofer, Katharina Mascherbauer, Julia Mascherbauer, Andreas Kammerlander, Tatjana Traub-Weidinger, Gregor Kasprian, Tienush Rassaf, Jens Kleesiek, Ken Herrmann, Philipp Bartko, Christian Hengstenberg, Marcus Hacker, Raffaella Calabretta, Christian Nitsche","doi":"10.1371/journal.pdig.0001637","DOIUrl":"10.1371/journal.pdig.0001637","url":null,"abstract":"<p><p>Timely detection is crucial to improve outcomes in patients with cardiac amyloidosis (CA) by initiation of guideline-directed disease-modifying treatments. Although confirmatory bone scintigraphy is highly accurate for CA detection, identifying at-risk patients for referral remains challenging. This study aimed to develop and validate a machine learning model, Amylo-Detect, using structured multimodal electronic health record (EHR) data to guide referrals for confirmatory scintigraphy and monoclonal protein testing. Consecutive all-comer patients (n = 11,616) referred for bone scintigraphy at the Vienna General Hospital (2010-2023) were retrospectively included. Patients referred before August 2020 formed the development cohort. The remaining patients comprised the internal validation cohort. External validation was performed at the University Hospital Essen (n = 1,521). Amylo-Detect was trained using 50 routinely available parameters to predict CA-suggestive uptake (Perugini grade ≥2) and compared with an existing score and clinical routine. High-grade uptake was present in 388 patients (3.0%). Amylo-Detect demonstrated excellent performance in development (AUC 0.93), independent internal validation (AUC 0.91), and external validation cohort (AUC 0.91), outperforming existing scoring systems and clinical routine. Results were consistent across subgroups, even when crucial predictors were missing. Of the 42/388 (10.8%) patients missed in clinical routine, 12/42 (29%) were additionally detected by Amylo-Detect. The model further conveyed significant prognostic value for mortality and heart failure hospitalization. We present Amylo-Detect, a validated EHR-based tool for CA risk prediction, available as a web app, allowing application and further evaluation. By improving timely detection and referral, Amylo-Detect may help address diagnostic delays, pending prospective evaluation.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001637"},"PeriodicalIF":7.7,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13649026/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effectiveness of the electronic Community Health Information System on maternal continuum of care completion in rural East Shewa Zone, Eastern Ethiopia: A prospective controlled quasi-experimental study.","authors":"Serbesa Dereje Degaga, Solomon Shiferaw, Girma Taye Aweke","doi":"10.1371/journal.pdig.0001775","DOIUrl":"10.1371/journal.pdig.0001775","url":null,"abstract":"<p><p>Maternal mortality in Ethiopia remains high, and strengthening the continuum of maternal care is essential for reducing preventable deaths. The electronic Community Health Information System was introduced to improve service delivery through decision support, client tracking, and electronic referrals, but evidence of its effect in routine settings is limited. This study assessed the effect of the eCHIS's Reproductive, Maternal, Newborn, Child Health and Nutrition module intervention on completion of the maternal continuum of care. A prospective controlled pre-post quasi-experimental study design was conducted in Primary Health Care Units. Baseline data were collected from January 5 to February 5, 2025, enrolling pregnant women with twenty six weeks of gestation or earlier. Participants were followed for nine months, and end-line data collected that year. Maternal care outcomes came from antenatal, delivery, and postnatal registers, with verification interviews when records were incomplete. The primary outcome was completion of the maternal continuum of care. Intervention effects were estimated using a Difference in Differences fixed effects model with cluster robust standard errors. Follow-up was completed for 409 women in each group (97.5%). Baseline continuum of care completion was comparable across groups. By endline, completion increased from 19.6% to 47.7% in the intervention group versus 19.1% to 23.7% in controls. The eCHIS-RMNCHN intervention produced a 23.5% point improvement in full continuum of care completion (β = 0.235; 95% CI: [0.155-0.314]), with significant gains in ANC4+ and ANC4 + combined with SBA. Implementation of the eCHIS-RMNCHN module intervention was associated with substantial improvements in maternal continuum-of-care completion and key intermediate indicators. Strategies guided by the CFIR ERIC framework helped address local barriers to digital system use. Sustained implementation, leadership engagement, technical support, community involvement, and continuous performance monitoring are essential for maintaining these improvements and informing scale up efforts.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001775"},"PeriodicalIF":7.7,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13645055/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149129491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data driven approach to identify vulnerable plaques in carotid atherosclerosis using computed tomography imaging and computational fluid dynamics: A pilot study.","authors":"Shaokai Zheng, David Bervini, Hugo Besson, Sofía González Martínez, Mirjam R Heldner, Mandy Müller, Thomas Meinel, Johannes Kaesmacher, Andreas Harloff, Dominik Obrist","doi":"10.1371/journal.pdig.0001762","DOIUrl":"10.1371/journal.pdig.0001762","url":null,"abstract":"<p><p>The current clinical gold standard for identifying vulnerable plaque is plaque Magnetic Resonance Imaging (MRI). However, owing to its limited availability, many patients do not have access. This pilot study aims to explore whether computational fluid dynamics (CFD) modelling, established using computed tomography angiography (CTA) and neurovascular ultrasound measurements (NVUS), as part of current clinical routine, can be used to triage vulnerable plaque in patients with carotid atherosclerotic stenosis. Nine patients who underwent evaluation for carotid endarterectomy between January 2021 and March 2023 were retrospectively included. Carotid geometries were segmented from CTA and reconstructed for CFD simulations using patient-specific NVUS-derived boundary conditions. Geometric and novel hemodynamic features were extracted. A logistic regression (LR) and a support vector machine (SVM) classifier with leave-one-out cross-validation and embedded feature selection were trained to identify vulnerable plaque, assessed according to the Plaque-Reporting and Data System based on multi-modal plaque MRI findings. Model performance was evaluated using area under the receiver operating characteristic curve (ROC-AUC), precision-recall AUC, confusion matrix, and other related metrics. Eighteen carotid arteries (9/18 were vulnerable) were analyzed. The classifiers achieved mean ROC-AUC of 0.77 and 0.78 for the LR and SVM, respectively. Key risk factors identified by both classifiers included mean helicity (OR: 0.5, 95%CI: 0.42, 0.58; mean feature coefficient: -0.27, 95%CI: -0.39, -0.15), kurtosis/skewness of helicity, and skewness of streamwise vorticity. Patient-level and sub-cohort analyses also support these findings. This pilot study demonstrated that combining NVUS and CTA with CFD may facilitate the identification of vulnerable plaques. Higher order statistics of intravascular hemodynamics may serve as promising surrogate features of plaque vulnerability.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001762"},"PeriodicalIF":7.7,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13645035/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149129453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital solutions for tackling mental health in cancer patients and their families: The ALTHEA project.","authors":"Ilaria Durosini, Dario Monzani, Veronica Coppini, Maria Vittoria Ferrari, Chiara Bovolenta, Asia Grigis, Giuseppe Curigliano, Paola Zagami, Elisabetta Munzone, Ida Minchella, Francesca Magnoni, Fedro Alessandro Peccatori, Isabelle Scholl, Marcel Meyerheim, Mieke Van Hemelrijck, Hannah Gsell, Lucija Gosak, Laura Ciria-Suarez, Karl Làvo, Norbert Graf, Marta Pérez-Campdepadrós, Marie Lange, Elisabeth Andritsch, Birute Brasiuniene, Andreas Charalambous, Nataliia Dovganych, Iva Kirac, Haridimos Kondylakis, Kostyantyn Kovalyov, Cristian Ochoa-Arnedo, Manolis Tsiknakis, Béata Juzyna, Ines Vaz-Luis, Philip E V Van Kerrebroeck, Nicole Erickson, Margarita Tareyn, Norbert Couespel, Clémentine Rialland, Luigi Grassi, Denis Horgan, Saska Benedicic Tomat, Begonya Nafria Escalera, Delia Nicoara, Claudio Lombardo, Gabriella Pravettoni","doi":"10.1371/journal.pdig.0001675","DOIUrl":"10.1371/journal.pdig.0001675","url":null,"abstract":"<p><p>Despite advances in cancer treatment, the growing global cancer burden underscores the need for comprehensive and supportive care to address the needs of patients and their families. However, access to psycho-oncological support remains often limited across healthcare systems. This could be related to factors such as resource constraints and insufficient integration of psychological care into routine oncological pathways. Individual-level barriers also exist, including low awareness of available services, stigma, and heterogeneity in patients' perceived needs and help-seeking attitudes. Nevertheless, psychological burden can significantly influence patients' experience of the treatment pathway, influencing cognitive clarity, perceived self-efficacy, communication, risk perception, decision-making, and all areas of daily life. In this context, digital mental health solutions have emerged as promising tools to enhance accessibility, continuity of care, and personalisation of support. In this manuscript, we aim to describe the structure of the ALTHEA project (\"tAckling menTal Health cancer patients and their families: digital solutions for bEtter cAre\"), which focuses on the development of a novel web-based digital platform targeting cancer patients and their informal caregivers' mental health across different stages of the care pathway. ALTHEA is distinguished by the presence of screening, classification into different support profiles, and tailored interventions within a single platform, as well as by its co-creation approach involving patients and stakeholders. In this line, the ALTHEA project represents a timely opportunity to integrate digital health solutions into psycho-oncology care, improving mental health outcomes, enhancing patient empowerment, and promoting a more holistic approach to cancer care.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001675"},"PeriodicalIF":7.7,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13645018/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149129483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Regulating the drafting fiction: A qualitative content analysis of official guidance and regulator documents on AI scribe adoption and use in healthcare.","authors":"Samuel Oluwatobi Atiku","doi":"10.1371/journal.pdig.0001776","DOIUrl":"10.1371/journal.pdig.0001776","url":null,"abstract":"<p><p>Artificial intelligence scribes use automatic speech recognition and generative language models to document clinical consultations, but official expectations for safe adoption are scattered across privacy, professional, medical device, safety and health-system guidance. We conducted a qualitative content analysis of 74 unique English-language official documents issued by regulators, government bodies, medical colleges and health-system organisations. Documents were identified through a staged grey literature search and coded using a hybrid deductive and inductive framework. We developed the analytic concept of a \"drafting fiction\" to describe a recurrent cross-document pattern in which artificial intelligence-generated clinical text remains provisional until clinician review and authentication. Across guidelines, transparency is common, consent is uneven, verification is treated as the central safeguard in direct AI-scribe guidance, privacy governance extends from secure storage to voice identifiability, and medical device classification shifts when transcription becomes summarisation. Organisational safety cases, vendor due diligence and data protection assessments narrow product-level uncertainty, but they do not define the encounter-level support needed for clinicians to detect hallucination, omission, automation bias or model drift. The study concludes that artificial intelligence scribe governance is taking shape as a sociotechnical architecture, although it relies on an underspecified human control assumption. Future guidance should define reasonable verification standards, allocate protected review time, resource consent refusal pathways and require monitoring of performance across accents, language and speech differences.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001776"},"PeriodicalIF":7.7,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13641376/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"When the algorithm passes the test but fails the patient.","authors":"Alex Mirugwe","doi":"10.1371/journal.pdig.0001774","DOIUrl":"10.1371/journal.pdig.0001774","url":null,"abstract":"<p><p>Tens of thousands of AI diagnostic tools have been authorised for clinical diagnosis worldwide, justified primarily by benchmark accuracy (performance on a fixed, curated test dataset), a number that says little about what the tool will do at the bedside. The field rests on three flawed foundations: datasets that do not reflect the populations served, development pipelines that exclude clinicians, and regulatory pathways that mistake on-paper accuracy for evidence of benefit. Closing the gap requires global coordination and the political will to treat clinical validation as a prerequisite rather than an afterthought.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001774"},"PeriodicalIF":7.7,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13641339/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Naïve Bayes is an interpretable and predictive machine learning algorithm in predicting osteoporotic hip fracture in-hospital mortality compared to other machine learning algorithms.","authors":"Jo-Wai Douglas Wang","doi":"10.1371/journal.pdig.0001771","DOIUrl":"10.1371/journal.pdig.0001771","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.1371/journal.pdig.0000529.].</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001771"},"PeriodicalIF":7.7,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13641337/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Entity-centric evaluation of large language model responses for medical question-answering tasks.","authors":"Yi Liu, Vijaya B Kolachalama","doi":"10.1371/journal.pdig.0001752","DOIUrl":"10.1371/journal.pdig.0001752","url":null,"abstract":"<p><p>Large language models (LLMs) are increasingly evaluated for clinical question-answering (QA) tasks, yet benchmark accuracy alone provides limited assurance of clinical reliability. Existing evaluation metrics often fail to capture whether model reasoning preserves patient-specific context and diagnostic intent, while many require external references or manual annotation, limiting scalability and real-world applicability. To address this gap, this study proposes EntQA, a reference-free, entity-centric metric that evaluates how well LLM-generated responses retain clinically relevant biomedical concepts from patient backgrounds and diagnostic questions. Across five medical QA benchmarks and seven Qwen 2.5 Instruct models (0.5B-72B parameters), EntQA showed consistently positive associations with model accuracy and scaling, outperforming conventional overlap- and embedding-based metrics that frequently exhibited weak or negative correlations. Group-level correlations with accuracy reached Spearman ρ =0.9286, while correlations with model scale reached ρ = 0.252. These findings suggest that EntQA provides a scalable and interpretable framework for assessing clinical fidelity and reasoning quality in healthcare LLMs without requiring gold standard references or external evidence.</p>","PeriodicalId":74465,"journal":{"name":"PLOS digital health","volume":"5 10","pages":"e0001752"},"PeriodicalIF":7.7,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13641718/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}