Diagnosis最新文献

{"title":"Semantic distance for hypothesis generation in differential diagnosis: a framework to support the pivot and cluster strategy.","authors":"Sho Isoda, Tadayuki Hashimoto, Taro Shimizu, Masahide Koda, Masasi Hattori, Tomio Suzuki","doi":"10.1515/dx-2026-0124","DOIUrl":"https://doi.org/10.1515/dx-2026-0124","url":null,"abstract":"<p><p>The pivot and cluster strategy (PCS) is an emerging approach to diagnostic hypothesis generation. In this strategy, clinicians generate an initial intuitive diagnosis (the pivot) along with a cluster of diseases with similar clinical representations for systematic consideration. Although PCS was proposed to prevent diagnostic error, a knowledge gap remains in its practical implementation: existing differential diagnosis lists rely primarily on individual expertise, lacking reproducibility and scalability. We have previously proposed that deliberately exploring diseases at a moderate semantic distance, that is, diseases moderately similar to the pivot, may foster diagnostic creativity. Building on that proposal, this paper specifies it as a rank-defined, auditable research programme, drawing on two established methods: ranking candidate diagnoses by embedding-based similarity, and building disease vectors from clinical presentation, not disease names. Each disease is represented as a disease concept vector and positioned within a high-dimensional semantic space (the disease map). Diseases with similar clinical features are thereby intended to lie close together, a testable design rather than an established property. Candidate diagnoses are ranked by cosine similarity to a given pivot disease. Because training is stochastic and corpus-dependent, the ranking is auditable rather than strictly reproducible. Taking a clinician-generated pivot as input, the framework supports bounded exploration of moderate-distance candidates; it does not generate the pivot. The disease map is offered as a research agenda rather than a validated tool, and we outline diagnosis-referenced evaluation of map-level retrieval validity, followed by prospective testing of its effects on clinicians' hypothesis generation and diagnostic performance.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149170479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Establishing gender- and age-specific reference ranges for eight tumor markers in adults based on real-world data from Southern China.","authors":"Yanting Li, Jinyu Zhao, Huidi Peng, Yunyun Ye, Yongtang Liu, Ting Li, Xi Peng, Zeyi Cai, Yikun Yang, Shengjin Cui","doi":"10.1515/dx-2026-0109","DOIUrl":"https://doi.org/10.1515/dx-2026-0109","url":null,"abstract":"<p><strong>Objectives: </strong>To establish and validate age- and sex-specific reference intervals (RIs) for eight common tumor markers using large-scale real-world data (RWD) and to compare the performance of five indirect statistical algorithms for RI derivation.</p><p><strong>Methods: </strong>This retrospective study included 404,347 individuals undergoing health examinations at The University of Hong Kong-Shenzhen Hospital between October 2022 and December 2024. Distribution characteristics and age-related trends were evaluated using kernel density estimation and generalized additive models. After age and sex stratification, RIs for AFP, CA125, CA15-3, CA19-9, CEA, fPSA, tPSA, and the fPSA/tPSA ratio were established using the Tukey, Z-score, Gaussian mixture model (GMM), isolation forest (IF), and local outlier factor (LOF) methods. Performance was assessed by bias ratios and validated using an independent cohort.</p><p><strong>Results: </strong>Most tumor markers exhibited markedly right-skewed distributions, whereas the fPSA/tPSA ratio showed only moderate skewness. AFP and CEA showed negligible sex differences. Age-specific stratification was identified for male AFP, female CA125 and CA15-3, CA19-9, CEA, and prostate-related markers. Tukey and Z-score methods showed the greatest agreement and achieved validation pass rates of ≥90 % across all demographic and analyte subgroups. In contrast, GMM and IF failed validation in 11 subgroups, and LOF failed in 6 subgroups, mainly because of the over-exclusion of high-end values.</p><p><strong>Conclusions: </strong>Large-scale RWD enabled robust age- and sex-specific RIs for eight tumor markers. For highly right-skewed tumor marker data, the Tukey and Z-score methods outperformed GMM, IF, and LOF, providing reliable and practical approaches for clinical RI establishment.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149165829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Visible reasoning: turning AI consultation into diagnostic learning.","authors":"Taro Shimizu","doi":"10.1515/dx-2026-0106","DOIUrl":"https://doi.org/10.1515/dx-2026-0106","url":null,"abstract":"<p><p>AI consultation is now common in diagnostic work, but it is still judged mainly by whether it improves the immediate answer. In this paper, I propose that AI consultation can also support clinician learning when the clinician's diagnostic judgment and the AI response are recorded in the same format and revisited after follow-up. I describe this proposed approach as visible reasoning, a two-stage process. In Stage 1, before viewing the AI output, the clinician records a brief account of the diagnostic judgment and a ranked differential diagnosis. The AI then responds in the same format, and the clinician compares the two records. In Stage 2, after meaningful follow-up becomes available, the clinician returns to the saved records to examine what mattered, where the two judgments converged or diverged, and what lesson and cue should be carried forward to similar cases, while recording the source and certainty of the final diagnosis or outcome. A worked example shows how differences in feature weighting can change the next action and generate a concrete, clinician-specific lesson. In this way, AI consultation becomes not only a decision aid but also a reusable record for learning.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149105505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Presentation patterns for initial incorrect diagnosis in diagnostic errors.","authors":"Yukinori Harada, Kotaro Kunitomo, Masayuki Amano, Madoka Hiraga, Kosuke Ishizuka, Sho Isoda, Mamoru Komatsu, Tetsu Sakamoto, Yuya Wada, Tetsuro Aita, Mizuki Sakairi, Kiyoshi Shikino, Taku Harada, Ren Kawamura, Taiju Miyagami, Daiki Morioka, Naomichi Uemura, Satoshi Watanuki, Tomoyasu Matsubara, Rikiya Watanabe, Taro Shimizu","doi":"10.1515/dx-2026-0120","DOIUrl":"https://doi.org/10.1515/dx-2026-0120","url":null,"abstract":"<p><strong>Objectives: </strong>The typicality of clinical presentations for the initial (incorrect) diagnosis in cases of diagnostic error remains poorly understood. We conducted this study to systematically characterize these clinical presentation patterns in diagnostic-error cases using the PSUC classification.</p><p><strong>Methods: </strong>We retrospectively applied the PSUC classification (Primary, Suggestive, Uncommon, Chameleon features) to the initial diagnoses of cases from a dataset of diagnostic-error case reports. Cases with a single explicitly named initial working diagnosis were included. Classification was performed by 20 reviewers using calibrated, pair-based consensus. We summarized the distribution of initial-diagnosis PSUC patterns and the commonality of the initial and final diagnoses.</p><p><strong>Results: </strong>We included 306 cases in this study. In most error cases, the presentation carried the Primary features of the initial (incorrect) diagnosis (258 cases, 84.3 %); however, only 14.1 % were purely typical (Primary features alone present), and 64.0 % of cases with Primary features also carried at least one Uncommon or Chameleon feature. Secondarily, the proportion of common diseases decreased from 52.9 % at the initial diagnosis to 16.1 % at the final diagnosis, and the most frequent commonality transition was from a common initial diagnosis to an uncommon final diagnosis (43.0 %).</p><p><strong>Conclusions: </strong>Most diagnostic errors in this dataset occurred in cases that carried Primary features of the wrong diagnosis, suggesting that the superficial resemblance to the wrong diagnosis largely drives them. Because disconfirmatory features coexisted with Primary features in most cases, diagnostic-safety strategies may benefit from directing attention to these discriminating features even when the initial diagnosis appears superficially typical.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149105221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of a new non-competitive chemiluminescent immunoassay for 25-OH vitamin D measurement.","authors":"Jinqing Zhong, Binning Huang, Aihua Lin, Geng Wu, Lili Zhang, Xiaowei Yu, Shun Zhuang, Daying Zheng, Huangxing Chen, Zhongying Zhang","doi":"10.1515/dx-2026-0094","DOIUrl":"https://doi.org/10.1515/dx-2026-0094","url":null,"abstract":"<p><strong>Objectives: </strong>The increasing demand for vitamin D measurement over the past few decades has prompted manufacturers to develop automated chemiluminescence immunoassay (CLIA), including the noncompetitive anti-immune complex assay (NACA) system for the measurement of small molecules. This study aims to evaluate the performance of the Snibe-NACA 25-OH Vitamin D assay for measuring total 25-Hydroxy Vitamin D assay [25(OH)D] levels.</p><p><strong>Methods: </strong>A total of 200 serum samples were collected from a clinical laboratory for 25(OH)D assay to evaluate the performance of the Snibe-NACA 25-OH Vitamin D assay using a double-blind method. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used as a reference method and the results were compared with those obtained from four other chemiluminescence competitive 25(OH)D assays.</p><p><strong>Results: </strong>The Snibe-NACA 25-OH Vitamin D assay demonstrated satisfactory analytical performance with optimal precision (repeatability CV: 1.13-2.75 %;intra-laboratory CV: 1.59-2.58 %) and relatively small bias (-3.84 % mean relative bias compared with the LC-MS/MS),excellent agreement with LC-MS/MS, as evidenced by Passing-Bablok regression (R²: 0.9245, slope: 0.99, 95 % CI: 0.96-1.03; intercept: -1.24, 95 % CI: -3.49 to 0.65) and high correlation (Spearman's ρ=0.969; ICC=0.9584). Notably, the assay showed the highest concordance with LC-MS/MS in the classification of vitamin D status, achieving an overall agreement of 87.0 %, with particularly strong concordance in identifying deficiency (97.8 %).</p><p><strong>Conclusions: </strong>The Snibe-NACA 25-OH Vitamin D assay demonstrated good performance and compared favorably with both LC-MS/MS and other competitive assays, providing evidence of its high reliability in clinical diagnostic practice.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149007969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnosis-related handoffs for inter-facility transfers to the pediatric intensive care unit.","authors":"Irene Pantekidis, Poornima Pandiyan, Hardeep Singh, Christopher P Landrigan, Christina L Cifra","doi":"10.1515/dx-2026-0083","DOIUrl":"https://doi.org/10.1515/dx-2026-0083","url":null,"abstract":"<p><strong>Objectives: </strong>Inter-facility transfers to the pediatric intensive care unit (PICU) increase the risk of patient harm from communication breakdowns. We previously developed I-PASS-to-PICU, a structured inter-facility handoff program to improve information exchange during these transitions. However, it was not designed to optimize communication related to diagnosis (diagnostic handoff). This exploratory study evaluated how well I-PASS-to-PICU facilitated inter-facility diagnostic handoffs.</p><p><strong>Methods: </strong>Using mixed methods, we analyzed audio-recorded referral calls to a single PICU to characterize inter-facility diagnostic handoff communication before and after I-PASS-to-PICU implementation. Two pediatric intensivists trained to review calls indicated whether diagnosis-relevant information or activities were discussed. Discrepancies in review were resolved via consensus. Reviewers also provided qualitative observations on diagnosis-related discussions.</p><p><strong>Results: </strong>Forty-four referral calls were reviewed (16 pre-and 28 post-I-PASS-to-PICU implementation). Compared with unstructured handoffs, I-PASS-to-PICU use did not result in statistically significant differences in the communication of diagnosis-related information in inter-facility handoffs. However, we observed some absolute differences; there were more calls using I-PASS-to-PICU in which illness severity was discussed (75 % vs. 50 %) and the primary diagnosis was stated by the referring clinician (68 % vs. 44 %) or receiving PICU physician (39 % vs. 25 %). Qualitative analysis revealed that shared mental models were more effectively created when primary diagnoses and diagnostic uncertainty were discussed.</p><p><strong>Conclusions: </strong>I-PASS-to-PICU may support aspects of the inter-facility diagnostic handoff but needs further revision to facilitate consistent discussion of diagnoses and diagnostic uncertainty. Future work will focus on redesigning and evaluating I-PASS-to-PICU for improving the diagnostic handoff during inter-facility PICU transfers.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149008031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"\"Learning bystander activation\": resident-led diagnostic error case conferences and sustained reflective engagement among junior residents.","authors":"Yuya Wada, Mamoru Komatsu, Sora Takeuchi, Shunsuke Arai, Taro Shimizu","doi":"10.1515/dx-2026-0080","DOIUrl":"https://doi.org/10.1515/dx-2026-0080","url":null,"abstract":"<p><strong>Objectives: </strong>Diagnostic errors are a major patient safety concern, particularly in emergency settings where junior residents frequently perform initial evaluations. Diagnostic error case conferences are used as educational interventions, but their impact on sustained learning behaviors remains unclear. This study examined whether participation in resident-led conferences is associated with changes in reflective and self-directed learning.</p><p><strong>Methods: </strong>This single-center mixed-methods study involved junior residents participating in weekly diagnostic error case conferences. Pre- and post-intervention surveys assessed awareness of diagnostic errors and cognitive biases, and engagement in reflective and self-directed learning. Quantitative data were analyzed using Fisher's exact test, and qualitative responses were analyzed thematically.</p><p><strong>Results: </strong>Recognition of diagnostic errors increased from 62.5 to 100 % (p=0.006), and awareness during emergency practice from 58.3 to 100 % (p=0.003). The proportion reviewing more than 70 % of their cases increased from 54.2 to 87.5 % (p=0.040). Qualitative analysis identified heightened diagnostic vigilance, emotional engagement, and environment-driven reflective pressure, even among residents not directly involved in the cases.</p><p><strong>Conclusions: </strong>Resident-led diagnostic error case conferences were associated with increased reflective engagement. We propose the concept of \"learning bystander activation,\" defined as the activation of reflective and self-directed learning behaviors through exposure to a shared learning environment, even without direct involvement in the clinical case.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149012089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Plasma signals of lung tumor promotion and the ubiquitous challenge of biomarker specificity.","authors":"Eleftherios P Diamandis","doi":"10.1515/dx-2026-0114","DOIUrl":"https://doi.org/10.1515/dx-2026-0114","url":null,"abstract":"<p><p>This opinion paper examines a recently reported 14-biomarker plasma panel designed to predict lung cancer development and identify patients who might benefit from preventive canakinumab treatment. While stratifying patients using this panel and clinical risk factors improves the number needed to treat (NNT) from 1,000 down to 50, the strategy still faces major hurdles due to the high economic cost and systemic side effects of canakinumab, which limits clinical benefit to a tiny fraction of the treated group. A closer biochemical examination reveals that the panel's 14 markers-such as CEACAM5 (CEA) and WFDC2 (HE4)-are already well-documented in classic tumor marker literature and are notoriously non-specific, frequently becoming elevated across various malignancies. Furthermore, notable established markers like CYFRA 21-1 and CA125 are missing from the panel. Ultimately, because these markers lack organ specificity, they are not utilized for early detection. The logical next step for this field will be the development of a single multiplex assay to quantify all 14 proteins simultaneously.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148988524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Small bowel obstruction from signet-ring cell appendiceal carcinoma.","authors":"Colby Weil-Lonigan, Celestine He, Robin Levenson, Andrew Sanchez","doi":"10.1515/dx-2026-0125","DOIUrl":"https://doi.org/10.1515/dx-2026-0125","url":null,"abstract":"<p><strong>Objectives: </strong>To illustrate the diagnostic challenge that arises in patients presenting with symptoms due to small bowel obstruction (SBO) whose initial cross-sectional imaging does not reveal the site or etiology of obstruction, and to review the pathophysiologic mechanisms by which SBO can evade radiographic detection.</p><p><strong>Case presentation: </strong>A man in his 80s with a remote history of intraabdominal surgeries presented with abdominal pain, bloating, and emesis. Non-contrast enhanced computed tomography (CT) imaging of the abdomen and pelvis showed distended small bowel loops with air-fluid levels, but no transition point or distal decompression. He was initially treated with isotonic fluids and a nasogastric tube to suction with resolution of symptoms, and he was discharged with a diagnosis of ileus of unclear etiology. However, approximately one month later, his symptoms recurred, and repeat CT imaging showed potential SBO at the terminal ileum. Exploratory laparotomy was performed for failure of conservative management, which ultimately revealed appendiceal mucinous adenocarcinoma with peritoneal carcinomatosis as the cause of the patient's initial presentation.</p><p><strong>Conclusions: </strong>CT findings of obstruction without an identifiable transition point may not reliably distinguish SBO from ileus, and symptom resolution with conservative management may falsely lead to an ileus diagnosis and drive premature closure. Because ileus requires an explanatory trigger, its diagnosis in the absence of one should prompt consideration of low-grade partial SBO as well as intermittent and functional pathophysiology.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148988480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Early evidence for a multi-agent AI simulator for clinical reasoning practice: performance, consistency, and challenges.","authors":"Matthew E Kelleher, Christine Y Zhou, Danielle E Weber, Seth Overla, James Bowen, Jose Generoso, Weibing Zheng, Sally A Santen, Laurah Turner","doi":"10.1515/dx-2026-0069","DOIUrl":"https://doi.org/10.1515/dx-2026-0069","url":null,"abstract":"<p><strong>Objectives: </strong>Clinical reasoning develops through repeated, deliberate practice, yet clinical and simulation environments are often limited by continuity, feedback, and scalability constraints. Large language models (LLMs) may address this by generating virtual clinical encounters accessible without live instructors or standardized patients but validity evidence remains limited. This study explored the performance of MAESSCR (Multi-Agent Educational Scenario Simulator for Clinical Reasoning), a platform where multiple LLM-based agents, each assigned a distinct role (i.e. patient, physical exam, diagnostic testing) interact with learners through a text-based encounter.</p><p><strong>Methods: </strong>In Fall of 2024, 175 second-year medical students completed three MAESSCR clinical encounters as coursework. Six clinician-educators developed a rating tool to evaluate 120 transcripts (40 randomly sampled per case) using a dichotomous (yes/no) scale across four domains: (1) realism of agent responses, (2) adherence to scripted case details, (3) platform functionality, and (4) interference with students' independent reasoning through clinical findings. Qualitative narrative review supplemented binary ratings.</p><p><strong>Results: </strong>MAESSCR followed scripted details in 92 % (110/120) of transcripts. Diagnosis-changing information occurred in 2.5 % (3/120). Unrealistic patient portrayal appeared in 12 % (14/120) of encounters, and technical issues in 17 % (20/120). AI agents most often interfered with student's clinical reasoning by interpreting findings before the students had the opportunity. This occurred in 28 % (34/120) of encounters due to history/physical exam agents and 59 % (71/120) because of diagnostics/management agents. Most disruptions were minor and unlikely to compromise the overall encounter.</p><p><strong>Conclusions: </strong>Multi-agent, LLM-based simulations offer a scalable approach to deliberate practice of clinical reasoning. Educational value depends on role stability, contextual fidelity, and learners' opportunity to interpret clinical information independently. Specific design considerations are essential to ensure AI-generated simulations support, rather than disrupt, the clinical reasoning process.</p>","PeriodicalId":11273,"journal":{"name":"Diagnosis","volume":" ","pages":""},"PeriodicalIF":1.9,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148967270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}