2f:I[703058,["/_next/static/immutable/chunks/3bhq_5eh1d760.js","/_next/static/immutable/chunks/1855-ogyi8eh4.js","/_next/static/immutable/chunks/2pfsjpn6k6zlg.js","/_next/static/immutable/chunks/26c7v3_3uwnwu.js","/_next/static/immutable/chunks/25a3ozcrvi1ff.js","/_next/static/immutable/chunks/2r4wx50nvjh_4.js","/_next/static/immutable/chunks/23fsw06hsi5b1.js","/_next/static/immutable/chunks/06_yt270h41rg.js","/_next/static/immutable/chunks/3ezsiz-7shkna.js","/_next/static/immutable/chunks/20x2y96ldu5wx.js","/_next/static/immutable/chunks/2lxntcm3pd1_h.js","/_next/static/immutable/chunks/00dgco8irq3jw.js","/_next/static/immutable/chunks/00nwe-kq0rboc.js","/_next/static/immutable/chunks/2051ffx3xzio9.js"],"PaperListPageClient"] 30:T770,Study DesignDiagnostic accuracy study.ObjectivePrior evaluations of frontier models as radiology decision-support tools relied on 2-dimensional images or text reports; their ability to interpret volumetric data remains unclear. This study assessed Google Gemini 3 Pro for grading lumbar spinal canal stenosis on video lumbar magnetic resonance imaging (MRI) and evaluated diagnostic accuracy, agreement with neuroradiologist consensus, and the effect of localizer-assisted input.MethodsThe Radiological Society of North America (RSNA) 2024 Lumbar Spine Degenerative Classification Dataset, with American Society of Neuroradiology (ASNR) consensus labels, served as a reference benchmark; interobserver agreement among contributing readers was not reported. 100 examinations yielded 500 disc-level observations (371 normal/mild, 74 moderate, 55 severe), demonstrating marked class imbalance. Native imaging series were converted into synchronized video montages. Gemini 3 Pro generated one grade per disc level with and without localizer overlays. Primary outcome was linearly weighted kappa (κw); secondary outcomes included class-wise performance, severe-case error patterns, and overall accuracy.ResultsWithout localizer, overall accuracy was 75.6% (378/500) with fair agreement (κw = 0.39). Severe stenosis sensitivity was 41.8%; 43.6% of severe cases were downgraded to normal/mild, and 58.2% to non-severe. With localizer overlays, accuracy was 73.2% (366/500) with κw = 0.32, and severe sensitivity decreased to 30.9%; severe-to-normal/mild misses increased to 52.7%. Differences were not significant.ConclusionsGemini 3 Pro showed fair agreement with the neuroradiologist consensus benchmark, but apparent overall accuracy was inflated by the majority normal/mild class and masked clinically unacceptable under-detection of severe stenosis. Localizer-assisted input did not improve performance.31:Ta28, Study Design Diagnostic accuracy study. Objective Prior evaluations of frontier models as radiology decision-support tools relied on 2-dimensional images or text reports; their ability to interpret volumetric data remains unclear. This study assessed Google Gemini 3 Pro for grading lumbar spinal canal stenosis on video lumbar magnetic resonance imaging (MRI) and evaluated diagnostic accuracy, agreement with neuroradiologist consensus, and the effect of localizer-assisted input. Methods The Radiological Society of North America (RSNA) 2024 Lumbar Spine Degenerative Classification Dataset, with American Society of Neuroradiology (ASNR) consensus labels, served as a reference benchmark; interobserver agreement among contributing readers was not reported. 100 examinations yielded 500 disc-level observations (371 normal/mild, 74 moderate, 55 severe), demonstrating marked class imbalance. Native imaging series were converted into synchronized video montages. Gemini 3 Pro generated one grade per disc level with and without localizer overlays. Primary outcome was linearly weighted kappa (κw); secondary outcomes included class-wise performance, severe-case error patterns, and overall accuracy. Results Without localizer, overall accuracy was 75.6% (378/500) with fair agreement (κw = 0.39). Severe stenosis sensitivity was 41.8%; 43.6% of severe cases were downgraded to normal/mild, and 58.2% to non-severe. With localizer overlays, accuracy was 73.2% (366/500) with κw = 0.32, and severe sensitivity decreased to 30.9%; severe-to-normal/mild misses increased to 52.7%. Differences were not significant. Conclusions Gemini 3 Pro showed fair agreement with the neuroradiologist consensus benchmark, but apparent overall accuracy was inflated by the majority normal/mild class and masked clinically unacceptable under-detection of severe stenosis. Localizer-assisted input did not improve performance. 25:[[["$","script","0",{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"CollectionPage\",\"name\":\"Diagnostic Accuracy Study: study design Research & Clinical Takeaways\",\"description\":\"Explore peer-reviewed papers on Diagnostic Accuracy Study (study design). Clinician-written summaries and actionable practice insights. Updated daily.\",\"url\":\"https://www.reachrx.ai/hub/tag/study_design/Diagnostic%20Accuracy%20Study\",\"breadcrumb\":{\"@type\":\"BreadcrumbList\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https://www.reachrx.ai\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Research Hub\",\"item\":\"https://www.reachrx.ai/hub\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Study Designs\",\"item\":\"https://www.reachrx.ai/hub/tag/study_design\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"Diagnostic Accuracy Study\",\"item\":\"https://www.reachrx.ai/hub/tag/study_design/Diagnostic%20Accuracy%20Study\"}]}}"}}]],["$","$L2f",null,{"title":"Diagnostic Accuracy Study: Research & Clinical Takeaways","description":"Browse Diagnostic Accuracy Study research papers with clear, practical takeaways for your clinical practice.","papers":[{"doi":"10.1007/s00467-026-07529-7","pmid":"42709156","title":"Can urinary albumin-creatinine ratio be used as a surrogate for protein-creatinine ratio in the diagnosis and management of pediatric steroid sensitive nephrotic syndrome?","journal_title":"Pediatric Nephrology","journal_issn":"0931-041X","published_at":"2026-09-08T00:00:00.000Z","url":"https://doi.org/10.1007/s00467-026-07529-7","source_score":5,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-09-13T06:21:50.946Z","updated_at":"2026-09-13T06:21:50.946Z","tags":[{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"age_group","value":"pediatric","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"category","value":"Urology & Nephrology","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"condition","value":"Proteinuria","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"condition","value":"Steroid-Sensitive Nephrotic Syndrome","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"impact_flag","value":"confirms_prior","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"modality","value":"Urinary Biomarker","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"sample_size","value":"n=202","value_numeric":202},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"warning","value":"Specificity for the remission ACR cutoff (0.19 mg/mg) ranged from 79–100% across cross-validation rounds, indicating some variability that clinicians should consider.","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"warning","value":"The study used only internal cross-validation (fivefold with bootstrap resampling); findings have not been externally validated in an independent cohort.","value_numeric":null},{"paper_doi":"10.1007/s00467-026-07529-7","tag_type":"warning","value":"The study was conducted at a single center in India; ethnic and dietary factors affecting urine creatinine excretion may affect generalizability to other populations.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated whether spot urinary ACR can reliably substitute for PCR in diagnosing relapse and remission in children aged 1–18 years with steroid-sensitive nephrotic syndrome (SSNS), using PCR as the reference standard.","key_result":"In 202 children, ACR and PCR were very strongly correlated (Spearman's rs = 0.97) with good Bland-Altman agreement. Recommended ACR thresholds: **≥1.50 mg/mg** for relapse (sensitivity 90–100%, specificity 91–100%) and **≤0.19 mg/mg** for remission (sensitivity 100%, specificity 79–100%).","limitations":"- Internal cross-validation only; no external validation cohort was used.\n- Single-center study from one institution in India, limiting generalizability.\n- Specificity for remission varied widely (79–100%), suggesting some uncertainty at the remission cutoff.","practice_takeaway":"When PCR is unavailable, clinicians managing pediatric SSNS can use a spot urine ACR of **≥1.50 mg/mg** to identify relapse and **≤0.19 mg/mg** to confirm remission with high sensitivity. These cutoffs may also be useful for retrospective data analysis when only ACR was recorded."}},{"doi":"10.1001/jamacardio.2026.3829","pmid":"42661496","title":"Artificial Intelligence–Enabled Acquisition and Interpretation for Screening Aortic Stenosis","journal_title":"JAMA Cardiology","journal_issn":"2380-6583","published_at":"2026-08-28T00:00:00.000Z","url":"https://doi.org/10.1001/jamacardio.2026.3829","source_score":31,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-09-01T06:20:58.074Z","updated_at":"2026-09-01T06:20:58.074Z","tags":[{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"category","value":"Cardiac & Cardiovascular Systems","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"condition","value":"Aortic Stenosis","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"modality","value":"Artificial Intelligence / Deep Learning","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"modality","value":"Focused Cardiac Ultrasound","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"sample_size","value":"n=1302","value_numeric":1302},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"warning","value":"PPV of 49.4% for AI alone highlights a potential for false positives in lower-prevalence settings — prevalence in this screened population is not explicitly stated, which limits direct extrapolation.","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"warning","value":"The sample_size tag reflects the prospective novice-operator cohort (n=1302) as the primary prospective evaluation group; total participants across all development and validation cohorts exceed 11,000. Clinicians should interpret the numeric tag in this context.","value_numeric":null},{"paper_doi":"10.1001/jamacardio.2026.3829","tag_type":"warning","value":"The study design sits between a prospective cohort study and a diagnostic accuracy study; labeled as 'Diagnostic Accuracy Study' per the paper's own description.","value_numeric":null}],"summaries":{"what_studied":"A deep learning algorithm was developed and validated for detecting moderate or greater aortic stenosis (AS) using AI-guided focused cardiac ultrasound (FoCUS), including prospective evaluation by novice operators across multiple Mayo Clinic sites (Midwest, Arizona, Florida) from 2005–2025.","key_result":"The model achieved AUROC of 0.99 across all validation cohorts. In prospective novice-operator use (n=1302), 96.6% of exams were analyzable, with sensitivity 93% and specificity 96%; adding expert review of AI-positive and uninterpretable exams raised PPV from 49.4% to 91.1% (sensitivity 85.4%).","limitations":"Single health system (Mayo Clinic), which may limit generalizability; novice operators still required expert review of flagged/uninterpretable cases to achieve acceptable PPV; PPV of 49.4% with AI alone reflects a relatively low prevalence of moderate+ AS in the screened population.","practice_takeaway":"AI-guided FoCUS operated by novice users can reliably screen for moderate or greater AS — a hybrid model pairing AI triage with targeted expert review of positive and uninterpretable exams may expand access in resource-limited settings. Clinicians should note that AI-alone PPV (~49%) requires expert confirmation before acting on positive results."}},{"doi":"10.1001/jamasurg.2026.3738","pmid":"42647036","title":"Detecting Incisional Surgical Site Infections on Wound Images Through Deep Learning","journal_title":"JAMA Surgery","journal_issn":"2168-6254","published_at":"2026-08-26T00:00:00.000Z","url":"https://doi.org/10.1001/jamasurg.2026.3738","source_score":6,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-09-02T06:22:19.061Z","updated_at":"2026-09-02T06:22:19.061Z","tags":[{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"category","value":"Surgery","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"condition","value":"Surgical Site Infection","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"modality","value":"Deep Learning / Artificial Intelligence","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"sample_size","value":"n=5385","value_numeric":5385},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"warning","value":"Sample size tag combines training and external validation cohorts; these are distinct datasets and not a single study population. The combined N (5,385) is an approximation for indexing purposes.","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"warning","value":"The impact_flag 'practice_changing' is applied cautiously — the model shows strong promise but lacks reported sensitivity/specificity at defined thresholds and real-world deployment validation.","value_numeric":null},{"paper_doi":"10.1001/jamasurg.2026.3738","tag_type":"warning","value":"The paper is dated August 2026, which is very recent; findings have not yet been independently replicated and prospective deployment studies are still pending.","value_numeric":null}],"summaries":{"what_studied":"A convolutional neural network (InceptionV3 with transfer learning) was developed and externally validated to detect incisional surgical site infections (SSIs) from wound photographs submitted via telemedicine, across multiple surgical specialties.","key_result":"The model achieved an AUC of 0.91 (95% CI, 0.88–0.93) on internal testing (trained on 4,978 images) and an AUC of 0.82 (95% CI, 0.75–0.90) on external validation (407 images, 95 patients). Decision curve analysis showed net clinical benefit.","limitations":"- External validation was limited to a single academic center (95 patients, 407 images), which may not reflect diverse real-world settings.\n- Image labeling by multiple physicians introduced potential interobserver variability in the ground-truth labels.\n- The study does not report sensitivity/specificity at specific operating thresholds, limiting direct clinical applicability.","practice_takeaway":"A deep learning model can flag wound photos for SSI with good-to-strong discrimination, suggesting a viable automated triage tool for telemedicine wound monitoring programs. Clinical integration should await prospective trials defining the optimal decision threshold before deployment."}},{"doi":"10.1007/s10862-026-10315-2","pmid":null,"title":"Attention Deficit Hyperactivity Disorder Symptoms in Children and Youth With Tourette Syndrome: Evaluation of the Conners-4 as a Screening Instrument","journal_title":"Journal of Psychopathology and Behavioral Assessment","journal_issn":"0882-2689","published_at":"2026-08-22T00:00:00.000Z","url":"https://doi.org/10.1007/s10862-026-10315-2","source_score":5,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":false,"crossref":true,"unpaywall":true},"created_at":"2026-08-25T06:20:11.271Z","updated_at":"2026-08-25T06:20:11.271Z","tags":[{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"age_group","value":"adolescent","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"age_group","value":"child","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"category","value":"Psychology, Clinical","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"condition","value":"Attention Deficit Hyperactivity Disorder","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"condition","value":"Tic Disorders","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"condition","value":"Tourette Syndrome","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"impact_flag","value":"confirms_prior","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"modality","value":"Behavioral Rating Scale","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"sample_size","value":"n=73","value_numeric":73},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"warning","value":"Full text was not accessible; summary is based solely on the abstract and reference list. Some study details (e.g., structured diagnostic interview specifics, secondary outcome data) could not be verified.","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"warning","value":"Small single-centre sample (n=73) may limit the precision of the reported sensitivity/specificity estimates.","value_numeric":null},{"paper_doi":"10.1007/s10862-026-10315-2","tag_type":"warning","value":"Study design classified as 'Diagnostic Accuracy Study' (cross-sectional); this is not a standard RCT or cohort design — level of evidence set to 3 as a reasonable approximation for a diagnostic study.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated the Conners 4 Parent Report as a screening tool for ADHD in children and adolescents with tic disorders (mean age 10.6 years; 76.7% male), using DSM-5 clinician diagnosis as the reference standard in a registry-based sample from the Calgary Child Tic Registry.","key_result":"Among 73 participants (42.5% met ADHD criteria), ROC analysis showed excellent diagnostic performance (AUC = 0.91). A Conners 4 Total ADHD Symptoms T-score cut-point of 60 yielded 87% sensitivity and 81% specificity.","limitations":"- Small single-centre sample (n=73) limits generalisability.\n- Only the parent-report form was evaluated; teacher or self-report versions were not assessed.\n- ADHD diagnosis relied on clinician interview without structured diagnostic instruments, which may affect reproducibility.","practice_takeaway":"In children with Tourette syndrome or other tic disorders, a Conners 4 Parent Report T-score ≥60 on the Total ADHD Symptoms subscale performs well as a screen for ADHD and can guide referral for full diagnostic evaluation. Routine use of this cut-point in tic disorder clinics is supported by the data."}},{"doi":"10.1038/s41433-026-04766-w","pmid":"42567878","title":"Can large language models accurately triage oculoplastic referrals?","journal_title":"Eye","journal_issn":"0950-222X","published_at":"2026-08-07T00:00:00.000Z","url":"https://doi.org/10.1038/s41433-026-04766-w","source_score":3,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-08-17T13:33:17.725Z","updated_at":"2026-08-17T13:33:17.725Z","tags":[{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"category","value":"Ophthalmology","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"condition","value":"Oculoplastic Disorders","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"modality","value":"Large Language Model","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"warning","value":"Study design tag is inferred from the title and topic; the actual design may differ.","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"warning","value":"The paper metadata provided contains only the title, DOI, journal, and publication date — no abstract, methods, results, or conclusions were supplied. All summary fields requiring specific data (key results, limitations, practice takeaway, sample size, sex, level of evidence) have been omitted because they cannot be verified from the available text.","value_numeric":null},{"paper_doi":"10.1038/s41433-026-04766-w","tag_type":"warning","value":"Users should retrieve the full text to obtain specific accuracy metrics, sample size, and clinical conclusions.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated whether large language models (LLMs) can accurately triage oculoplastic referrals, assessing their performance in categorizing referral urgency in an ophthalmology subspecialty setting."}},{"doi":"10.1177/12034754261470756","pmid":"42517439","title":"Multimodal Large Language Models Versus Dermatologists for Smartphone-Image Classification of Suspicious Cutaneous Lesions: A Diagnostic Accuracy Study","journal_title":"Journal of Cutaneous Medicine and Surgery","journal_issn":"1203-4754","published_at":"2026-07-28T00:00:00.000Z","url":"https://doi.org/10.1177/12034754261470756","source_score":1,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-08-04T18:18:07.627Z","updated_at":"2026-08-04T18:18:07.627Z","tags":[{"paper_doi":"10.1177/12034754261470756","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"category","value":"Dermatology","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"condition","value":"Skin Cancer","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"condition","value":"Suspicious Cutaneous Lesions","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"modality","value":"Multimodal Large Language Model","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"warning","value":"Level of evidence set to '2' based on the diagnostic accuracy study design; this may need adjustment depending on reference standard and study execution details.","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"warning","value":"Only the paper title, DOI, journal, and publication date were provided — no abstract, methods, results, or full text were available. All summary content is inferred from the title alone and should be treated as preliminary. Key metrics (sensitivity, specificity, sample size, comparator details) could not be extracted.","value_numeric":null},{"paper_doi":"10.1177/12034754261470756","tag_type":"warning","value":"The impact_flag 'practice_changing' is tentative given the topic's significance, but cannot be confirmed without access to actual results.","value_numeric":null}],"summaries":{"what_studied":"This diagnostic accuracy study compared multimodal large language models (LLMs) against dermatologists in classifying suspicious cutaneous lesions from smartphone images, evaluating whether AI tools can match clinical expertise in skin cancer screening.","key_result":"Specific accuracy metrics (sensitivity, specificity, AUC) are not available from the provided abstract alone; the study assessed head-to-head diagnostic performance of multimodal LLMs versus dermatologists on smartphone-captured lesion images.","limitations":"Only the paper metadata was provided — specific results, sample size, inclusion criteria, and methodology details could not be extracted. Full conclusions remain unverifiable from available information.","practice_takeaway":"Dermatologists should watch this space: if multimodal LLMs prove non-inferior to specialists on smartphone images, they could support triage of suspicious skin lesions in low-access settings. Full data review is needed before clinical adoption."}},{"doi":"10.1111/vde.70105","pmid":"42457205","title":"External Validation of a Custom‐Trained Artificial Intelligence Model for Canine Ear Canal Disease and Comparison With Veterinary Professionals","journal_title":"Veterinary Dermatology","journal_issn":"0959-4493","published_at":"2026-07-15T00:00:00.000Z","url":"https://doi.org/10.1111/vde.70105","source_score":1,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-07-26T13:44:51.519Z","updated_at":"2026-07-26T13:44:51.519Z","tags":[{"paper_doi":"10.1111/vde.70105","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"category","value":"Dermatology","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"condition","value":"Ear Canal Mass","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"condition","value":"Otitis Externa","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"impact_flag","value":"limited_evidence","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"modality","value":"Artificial Intelligence Image Classification","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"sample_size","value":"n=300","value_numeric":300},{"paper_doi":"10.1111/vde.70105","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"warning","value":"Study population species (dog) and subject counts refer to images, not individual animals — true N of dogs is not reported.","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"warning","value":"The level of evidence rating reflects a diagnostic accuracy study with a single-clinician reference standard, which may inflate or deflate true performance estimates.","value_numeric":null},{"paper_doi":"10.1111/vde.70105","tag_type":"warning","value":"This paper is veterinary (canine), not human medicine — clinical applicability is limited to veterinary practice.","value_numeric":null}],"summaries":{"what_studied":"External validation of four custom-trained YOLOv8n AI models for classifying canine otoscopic images (healthy, otitis externa, ear canal mass) across 300 previously unseen images, compared against six veterinary professionals of varying experience.","key_result":"The best AI model (Model D, largest deduplicated dataset) achieved 66.3% overall agreement and κ = 0.508; sensitivity/specificity were 90.0%/69.0% for healthy, 59.0%/75.5% for otitis externa, and 50.0%/93.5% for ear canal masses. Veterinary professionals outperformed all models (median agreement 77.2%, κ = 0.658).","limitations":"- Study used static otoscopic images only, which may not reflect real-world dynamic video-based otoscopy.\n- Ground truth labels were assigned by a single board-certified clinician, introducing potential reference standard bias.\n- AI models were compared with only six veterinary professionals, limiting generalisability of the human benchmark.","practice_takeaway":"This AI tool shows moderate agreement for canine ear canal disease but consistently underperforms veterinary professionals — use it as a decision-support aid, not a standalone diagnostic tool. Clinicians should be especially cautious with ear canal mass detection, where model sensitivity was only 50%."}},{"doi":"10.1016/j.jim.2026.114087","pmid":"42456928","title":"Diagnostic performance of Pictor's targeted proteomics assay for extractable nuclear antigens compared with commercial line immunoassay and addressable laser bead immunoassay platforms","journal_title":"Journal of Immunological Methods","journal_issn":"0022-1759","published_at":"2026-07-15T00:00:00.000Z","url":"https://doi.org/10.1016/j.jim.2026.114087","source_score":34,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-07-28T14:02:21.565Z","updated_at":"2026-07-28T14:02:21.565Z","tags":[{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"brand","value":"BioPlex 2200","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"brand","value":"PictHealth Immunoscreen ENA","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"category","value":"Immunology","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"condition","value":"Connective Tissue Disease","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"condition","value":"Systemic Autoimmune Rheumatic Diseases","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"impact_flag","value":"limited_evidence","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"modality","value":"Autoantibody Multiplex Immunoassay","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"warning","value":"Age group set to 'adult' by inference; pediatric patients cannot be excluded without full text access.","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"warning","value":"Full text was blocked by a CAPTCHA; all extracted data are based solely on the abstract. Sample size and patient demographics are not reported in the abstract and could not be confirmed.","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"warning","value":"Sex tag set to 'mixed' by inference from a clinical autoimmune disease cohort — not explicitly stated in the abstract.","value_numeric":null},{"paper_doi":"10.1016/j.jim.2026.114087","tag_type":"warning","value":"Study design classified as 'Diagnostic Accuracy Study' (method comparison); no randomization or blinded reference standard was described — level of evidence set to 3 with some uncertainty.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated the diagnostic performance of the Pictor PictHealth® Immunoscreen ENA assay (a targeted proteomics, ELISA-format multiplex platform) for detecting extractable nuclear antigen (ENA) antibodies, comparing it against the Euroimmun ENA Line Immunoassay and BioPlex 2200 ENA assay using clinical sera from patients with autoimmune diseases.","key_result":"Three-way concordance across all assays ranged from 67.3% (TRIM21/Ro52) to 100% (Jo-1), with Sm and CENP-B showing ~95% agreement. Pairwise positive percent agreement (PPA) and negative percent agreement (NPA) exceeded 85% for most autoantibodies.","limitations":"Full text was inaccessible, so sample size, patient demographics, and detailed breakdown of pairwise PPA/NPA by analyte could not be verified. The comparator assays (Euroimmun LIA and BioPlex 2200) themselves lack a true gold standard, limiting definitive performance claims. TRIM21/Ro52 showed notably lower three-way concordance (67.3%), which may reflect known platform variability for this antigen.","practice_takeaway":"The Pictor ENA assay performs comparably to established multiplex platforms for most ENA antibodies and may be a viable, lower-cost option for SARDs testing in resource-limited laboratories. Caution is warranted for TRIM21/Ro52 results, where concordance across platforms is lowest."}},{"doi":"10.1158/0008-5472.can-25-5589","pmid":"42439806","title":"Label-Free Prediction of EGFR Mutation Status Using Fluorescence Lifetime Imaging and Deep Learning in Lung Adenocarcinoma","journal_title":"Cancer Research","journal_issn":"0008-5472","published_at":"2026-07-13T00:00:00.000Z","url":"https://hdl.handle.net/20.500.11820/10117d87-1aa3-4692-9eb7-daaa6bb05d74","source_score":88,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-07-20T13:36:06.109Z","updated_at":"2026-07-20T13:36:06.109Z","tags":[{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"category","value":"Oncology","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"condition","value":"Lung Adenocarcinoma","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"condition","value":"Non-Small Cell Lung Cancer","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"generic","value":"EGFR inhibitor (targeted therapy context)","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"modality","value":"Deep Learning / Computational Pathology","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"modality","value":"Fluorescence Lifetime Imaging","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"warning","value":"Sample size (total N) is not reported in the abstract; the sample_size tag was omitted.","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"warning","value":"The generic drug tag reflects the therapeutic context (EGFR-targeted therapy) rather than a drug directly tested in this study.","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"warning","value":"The 'practice_changing' impact flag is applied cautiously: the AUROC is impressive but external/prospective validation is not yet reported, limiting immediate clinical translation.","value_numeric":null},{"paper_doi":"10.1158/0008-5472.can-25-5589","tag_type":"warning","value":"The study design is described as a diagnostic accuracy/validation study but specific prospective vs. retrospective design is not stated — may warrant verification from the full text.","value_numeric":null}],"summaries":{"what_studied":"A deep learning (DL) model using fluorescence lifetime imaging (FLIM) was developed to predict EGFR mutation status — including exon 19 deletion and exon 21 point mutation — in lung adenocarcinoma tissue samples, without histologic staining or molecular testing.","key_result":"The FLIM-based DL approach achieved an AUROC of 0.966 on label-free images for EGFR mutation prediction, and accurately classified the two most common EGFR mutation subtypes (exon 19 deletion and exon 21 point mutation).","limitations":"Sample size and dataset composition are not detailed in the abstract; the study is likely single-institution, limiting generalizability. The method was tested on formalin-fixed tissue, so applicability to other sample types is unclear. No direct comparison to standard-of-care NGS or PCR turnaround time and cost is quantified.","practice_takeaway":"FLIM-based deep learning may allow rapid, nondestructive EGFR mutation prediction from unstained lung tissue — potentially useful when tissue is scarce or when faster results are needed before committing to molecular testing. Clinical validation in prospective, multi-site cohorts is needed before adoption."}},{"doi":"10.1111/cea.70386","pmid":"42389968","title":"Basophil Activation Test Provides Additional Diagnostic Value Beyond Ara h 2–Specific\n IgE\n Testing in Peanut Allergic Adults","journal_title":"Clinical & Experimental Allergy","journal_issn":"0954-7894","published_at":"2026-07-02T00:00:00.000Z","url":"https://doi.org/10.1111/cea.70386","source_score":1,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-07-05T13:38:45.238Z","updated_at":"2026-07-05T13:38:45.238Z","tags":[{"paper_doi":"10.1111/cea.70386","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"category","value":"Allergy","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"condition","value":"Peanut Allergy","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"generic","value":"Ara h 2-specific IgE","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"modality","value":"Basophil Activation Test","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"modality","value":"Specific IgE Testing","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"warning","value":"Impact flag 'practice_changing' is tentative given the study appears to support an incremental diagnostic step; reassess after reviewing full results.","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"warning","value":"Only the paper metadata (title, DOI, journal, date) was provided — no full text, abstract, or results were available. All summary content and tags are inferred from the title alone. Key numerical results (sensitivity, specificity, sample size, AUC) could not be extracted and should be verified against the full text.","value_numeric":null},{"paper_doi":"10.1111/cea.70386","tag_type":"warning","value":"Study design and level of evidence are estimated from the title and publication context; confirm upon full-text review.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated whether the Basophil Activation Test (BAT) adds diagnostic value beyond Ara h 2–specific IgE (sIgE) testing alone for diagnosing peanut allergy in adults.","key_result":"BAT provided incremental diagnostic accuracy over Ara h 2 sIgE testing in peanut allergic adults, though specific sensitivity/specificity numbers are not available from the provided abstract.","limitations":"Details on sample size, blinding, and whether an oral food challenge was used as the gold standard are not confirmed from the provided text alone.","practice_takeaway":"In adults with suspected peanut allergy, adding BAT to Ara h 2 sIgE testing may improve diagnostic certainty — consider BAT when Ara h 2 sIgE results are ambiguous."}},{"doi":"10.1016/s2468-1253(26)00118-4","pmid":"42392123","title":"The Rome V criteria for the diagnosis of irritable bowel syndrome in secondary care: a diagnostic accuracy study","journal_title":"The Lancet Gastroenterology & Hepatology","journal_issn":"2468-1253","published_at":"2026-07-02T00:00:00.000Z","url":"https://doi.org/10.1016/s2468-1253(26)00118-4","source_score":16,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-07-11T13:33:06.160Z","updated_at":"2026-07-11T13:33:06.160Z","tags":[{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"category","value":"Gastroenterology & Hepatology","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"condition","value":"Irritable Bowel Syndrome","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"modality","value":"Diagnostic Criteria","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"sample_size","value":"n=726","value_numeric":726},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"warning","value":"Rome V criteria were published in May 2026 and this study was published shortly after — independent external validation in other centres and populations is not yet available.","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"warning","value":"Study population is predominantly female (75%) and based in a single UK specialist clinic, which may limit generalisability.","value_numeric":null},{"paper_doi":"10.1016/s2468-1253(26)00118-4","tag_type":"warning","value":"The 'practice_changing' impact flag is applied with caution: the clinical consequence of Rome V's lower sensitivity is explicitly stated as 'uncertain' by the authors.","value_numeric":null}],"summaries":{"what_studied":"This diagnostic accuracy study evaluated the performance of the new Rome V criteria for diagnosing IBS in 726 consecutive adults (≥16 years) referred to a specialist IBS clinic in secondary care (Leeds, UK, 2016–2024), comparing sensitivity and specificity against Rome III and Rome IV criteria using a standardised reference standard.","key_result":"Rome V criteria showed the lowest sensitivity of the three versions at 66.1% (95% CI 62.1–69.9) with a specificity of 80.1% (72.4–86.5). By comparison, Rome IV had sensitivity 78.9% and specificity 81.0%, while Rome III had sensitivity 87.5% and specificity 75.0%. Agreement between Rome V and earlier criteria was only fair to moderate (κ=0.36–0.55).","limitations":"- Single-centre secondary care study in the UK, limiting generalisability to primary care or other healthcare systems.\n- Reference standard was clinician-defined (symptom-based + negative investigations), not a universally validated gold standard.\n- Rome V criteria were published in May 2026; this retrospective application to historical data may not fully reflect intended clinical use.","practice_takeaway":"Clinicians using the new Rome V criteria should be aware that they will miss approximately 1 in 3 patients with IBS who would have been captured by Rome III or IV — sensitivity is meaningfully lower. Until further validation studies are available, a negative Rome V result should not confidently rule out IBS in secondary care."}},{"doi":"10.1111/myc.70199","pmid":"42307435","title":"Comparison of\n DermaGenius\n and In‐House\n qPCR\n Methods for Detection of Recalcitrant Dermatophyte Infections: A Practical Approach for Dermatologists and Mycologists","journal_title":"Mycoses","journal_issn":"0933-7407","published_at":"2026-06-17T00:00:00.000Z","url":"https://doi.org/10.1111/myc.70199","source_score":2,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-06-27T13:27:50.648Z","updated_at":"2026-06-27T13:27:50.648Z","tags":[{"paper_doi":"10.1111/myc.70199","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"brand","value":"DermaGenius Resistance Multiplex PCR","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"category","value":"Dermatology","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"condition","value":"Dermatophytosis","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"condition","value":"Recalcitrant Dermatophytosis","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"condition","value":"Terbinafine-Resistant Trichophyton Infection","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"generic","value":"terbinafine","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"modality","value":"Antifungal Susceptibility Testing","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"modality","value":"Multiplex qPCR Diagnostics","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"sample_size","value":"n=64","value_numeric":64},{"paper_doi":"10.1111/myc.70199","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"warning","value":"High prevalence of T. indotineae in this cohort (84% of isolates) may inflate apparent concordance and affect kappa interpretation — authors acknowledge this limitation.","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"warning","value":"Level of evidence assigned as 3 (diagnostic cohort study); no randomisation or prospective blinded design was used.","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"warning","value":"One author (E.F.J.M.) has received research funding from PathoNostics, the manufacturer of the DermaGenius RMP kit evaluated in this study; potential conflict of interest should be noted.","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"warning","value":"Study used cultured isolates rather than direct clinical specimens, so performance metrics may not fully reflect direct-from-sample diagnostic accuracy.","value_numeric":null},{"paper_doi":"10.1111/myc.70199","tag_type":"warning","value":"The W450T SQLE substitution is reported as novel in this study; its clinical significance and effect on terbinafine MICs remain unknown and require validation.","value_numeric":null}],"summaries":{"what_studied":"This study compared two qPCR methods — a commercial multiplex assay (DermaGenius RMP) and an in-house *T. indotineae*-specific qPCR (Tindo-qPCR) — for species identification and detection of terbinafine resistance-associated SQLE mutations in 64 dermatophyte isolates from patients with chronic or treatment-resistant dermatophytosis in Türkiye (April–December 2024), validated against ITS/SQLE sequencing and CLSI antifungal susceptibility testing.","key_result":"Overall concordance between the two qPCR methods was 93.8% (κ = 0.74, p < 0.001) for *T. indotineae* identification. SQLE mutations were detected in >90% of *T. indotineae* isolates; F397L was the dominant substitution (89.8%). A novel W450T SQLE substitution was identified. Notably, 10 isolates carried SQLE F397L yet remained terbinafine-susceptible (MIC ≤ 0.016 µg/mL), suggesting genomic background may modulate phenotypic resistance.","limitations":"- Single-centre cohort from Türkiye with a high *T. indotineae* prevalence, limiting geographic generalisability.\n- DermaGenius RMP was tested on cultured isolates, not direct clinical samples, so real-world turnaround benefits were not assessed.\n- Functional significance of SQLE mutations outside positions 393 and 397 (e.g., W450T) remains unvalidated experimentally.","practice_takeaway":"Don't wait for repeated treatment failure — request qPCR-based molecular testing (e.g., DermaGenius RMP) early in patients with chronic or recalcitrant tinea to detect *T. indotineae* and terbinafine resistance mutations before clinical relapse. When molecular and phenotypic susceptibility results diverge, add CLSI antifungal susceptibility testing to guide therapy, since SQLE mutations alone do not always predict elevated terbinafine MICs."}},{"doi":"10.1167/tvst.15.6.19","pmid":"42300656","title":"Microsaccade Dynamics During Visual Fixation as Markers for Parkinson's Disease: A Machine Learning Approach","journal_title":"Translational Vision Science & Technology","journal_issn":"2164-2591","published_at":"2026-06-16T00:00:00.000Z","url":"https://doi.org/10.1167/tvst.15.6.19","source_score":10,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-06-20T13:25:06.421Z","updated_at":"2026-06-20T13:25:06.421Z","tags":[{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"age_group","value":"older_adult","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"category","value":"Ophthalmology","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"condition","value":"Parkinson's Disease","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"impact_flag","value":"limited_evidence","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"modality","value":"Eye Tracking / Oculomotor Biomarker","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"sample_size","value":"n=84","value_numeric":84},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"warning","value":"Sex distribution differed notably between PD (65% male) and HC (39% male) groups; sex was not included as a covariate in the microsaccade-based ML models, which may partially confound results.","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"warning","value":"Square-wave jerks — more common in PD — were not separated from microsaccades, so some discriminative signal may reflect saccadic intrusions rather than true microsaccade dynamics.","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"warning","value":"Study design is labeled 'Diagnostic Accuracy Study' (observational case-control), not an RCT; level of evidence set to 3 reflecting a well-designed non-randomized study.","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.19","tag_type":"warning","value":"The study is cross-sectional with a single clinical center cohort; no longitudinal or external validation was performed.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated whether microsaccade features recorded during a passive visual fixation task (no head restraint) could distinguish 50 early-to-moderate PD patients (Hoehn & Yahr 1–3) from 43 healthy controls, using 10 machine learning classifiers with leave-one-subject-out cross-validation.","key_result":"PD patients had higher microsaccade rates (1.5 vs. 1.3 Hz), larger amplitudes (0.49° vs. 0.34°), and stronger horizontal bias than controls. A polynomial SVM achieved 77.4% accuracy, 76.1% sensitivity, and 78.9% specificity on held-out subjects (ROC-AUC 85.7%); filtering low-confidence predictions raised accuracy to 79.3%.","limitations":"- Small sample (n=84 after exclusions) with moderate PD severity only; generalizability to later stages is unknown.\n- No head stabilization and no separation of square-wave jerks from true microsaccades may introduce noise and misclassification.\n- Sex imbalance between groups (65% male PD vs. 39% male HC) was not corrected in the primary models.","practice_takeaway":"Microsaccade analysis during a simple, passive fixation task — requiring no special patient effort — can flag PD with ~77% accuracy using a polynomial SVM, outperforming a demographic baseline by ~15 percentage points. Clinicians should watch for emerging eye-tracking tools that capture fixational eye movements as a low-burden PD screening adjunct."}},{"doi":"10.1001/jamaneurol.2026.1868","pmid":"42295801","title":"Acoustic Analysis of Primary Care Patient–Clinician Conversations to Screen for Cognitive Impairment","journal_title":"JAMA Neurology","journal_issn":"2168-6149","published_at":"2026-06-15T00:00:00.000Z","url":"https://pubmed.ncbi.nlm.nih.gov/42295801/","source_score":48,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-06-21T13:29:42.075Z","updated_at":"2026-06-21T13:29:42.075Z","tags":[{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"age_group","value":"older_adult","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"category","value":"Clinical Neurology","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"condition","value":"Cognitive Impairment","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"impact_flag","value":"limited_evidence","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"modality","value":"Machine Learning / Acoustic Analysis","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"sample_size","value":"n=966","value_numeric":966},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"warning","value":"PPV of 30.4% is highly prevalence-dependent; performance may differ substantially in populations with lower or higher CI prevalence than the 21% observed here.","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"warning","value":"The full text attached in the PDF metadata is a 2019 Alzheimer's Disease Facts and Figures report, which does not match the study described in the abstract. Summaries and tags are based solely on the abstract and results sections, not the full text. Verify complete methodology details upon journal access.","value_numeric":null},{"paper_doi":"10.1001/jamaneurol.2026.1868","tag_type":"warning","value":"The study period (August 2020–December 2021) coincided with COVID-19, which may have affected visit patterns, speech characteristics, and patient mix in primary care.","value_numeric":null}],"summaries":{"what_studied":"This diagnostic study enrolled 966 English-speaking adults aged 55+ without prior dementia or MCI from primary care practices in New York City (n=787) and an external validation cohort in Chicago (n=179), recorded during routine visits from August 2020 to December 2021. Machine learning (ML) classifiers were trained on acoustic features extracted from 30-second speech segments using foundation models (Whisper, HuBERT, wav2vec 2.0) and expert-defined methods (eGeMAPS, prosody) to identify cognitive impairment (CI), defined as MoCA score ≥1 SD below age- and education-adjusted norms.","key_result":"Whisper-derived acoustic features performed best: AUROC 0.733 (95% CI 0.714–0.752) in the primary cohort, with similar results in the external validation cohort (AUROC 0.727; 95% CI 0.714–0.740). As a screening tool on the holdout set, the algorithm achieved sensitivity 68.2% (95% CI 61.8%–74.6%), specificity 63.6% (95% CI 59.8%–67.4%), and positive predictive value 30.4% (95% CI 28.7%–32.1%) against a 21% CI prevalence. Pitch, timing, and speech variability were the top acoustic predictors.","limitations":"The low PPV (30.4%) reflects a high false-positive rate unsuitable for standalone diagnosis. The study enrolled only English-speaking patients, limiting generalizability to other languages. CI was defined by a single MoCA threshold rather than comprehensive neuropsychological evaluation, which may misclassify borderline cases.","practice_takeaway":"Passive acoustic analysis of routine primary care conversations shows feasibility as a low-burden CI screening aid, but its PPV of ~30% means most positive screens will be false positives — it should trigger, not replace, formal cognitive assessment. Clinicians should not act on this tool alone pending further validation."}},{"doi":"10.1002/ijc.70534","pmid":"42252568","title":"Clinical Validation of Four Point‐of‐Care High‐Risk\n HPV\n Assays, Including Two Reduced‐Valency Assays, for Cervical Cancer Screening in Low‐Resource Settings","journal_title":"International Journal of Cancer","journal_issn":"0020-7136","published_at":"2026-06-07T00:00:00.000Z","url":"https://doi.org/10.1002/ijc.70534","source_score":33,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-06-10T14:23:54.974Z","updated_at":"2026-06-10T14:23:54.974Z","tags":[{"paper_doi":"10.1002/ijc.70534","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"brand","value":"Allplex-HPV-HR","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"brand","value":"Cobas-4800","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"brand","value":"HPV-Q","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"brand","value":"PathoDetect-HPV-14","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"brand","value":"PathoDetect-HPV-7","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"brand","value":"Truenat-HR-HPV-Plus","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"category","value":"Oncology","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"condition","value":"Cervical Cancer","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"condition","value":"Cervical Intraepithelial Neoplasia","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"condition","value":"Human Papillomavirus Infection","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"generic","value":"HPV nucleic acid amplification test","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"modality","value":"Molecular Diagnostics","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"modality","value":"Point-of-Care Testing","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"sample_size","value":"n=1159","value_numeric":1159},{"paper_doi":"10.1002/ijc.70534","tag_type":"sex","value":"female","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"warning","value":"CIN2 is an imperfect histological endpoint with known reproducibility issues; the study included only 19 CIN2 cases among the 97 CIN2+ cases.","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"warning","value":"The impact_flag 'practice_changing' is applied to Truenat-HR-HPV-Plus specifically, not all four assays evaluated.","value_numeric":null},{"paper_doi":"10.1002/ijc.70534","tag_type":"warning","value":"The VALGENT framework validation used samples from Argentina and Costa Rica only; generalizability to sub-Saharan Africa or South/Southeast Asia — the highest-burden low-resource regions — is uncertain.","value_numeric":null}],"summaries":{"what_studied":"Four point-of-care HR-HPV assays (PathoDetect-HPV-14, HPV-Q, Truenat-HR-HPV-Plus [8-type], and PathoDetect-HPV-7 [7-type]) were evaluated for clinical accuracy and reproducibility in cervical cancer screening using 1,159 cervical samples from the ESTAMPA study (Argentina and Costa Rica), compared against reference assays (Cobas-4800 and Allplex-HPV-HR) via the VALGENT framework.","key_result":"Truenat-HR-HPV-Plus met IARC validation criteria with 80.4% sensitivity (95% CI: 71.1–87.8) and 91.5% specificity (95% CI: 89.5–93.2) for CIN2+, non-inferior to Allplex-HPV-HR-8, and repeatability of 93.3% (κ=0.79); PathoDetect-HPV-7 showed lower sensitivity (68.1%); HPV-Q and PathoDetect-HPV-14 did not meet validation criteria.","limitations":"Samples were derived from a single multi-country study (Argentina and Costa Rica), which may limit generalizability to other low-resource settings; the CIN2+ case count was modest (n=97, including only 6 cancers), which constrains precision of estimates at the cancer endpoint.","practice_takeaway":"Truenat-HR-HPV-Plus (8-type) is a validated point-of-care option for cervical cancer screening in low-resource settings, offering higher specificity than 14-valent assays and potentially reducing unnecessary referrals. PathoDetect-HPV-7 and the two 14-type assays should not be used as primary screening tools until further validation is completed."}},{"doi":"10.1001/jamadermatol.2026.1492","pmid":"42234423","title":"Limits of Artificial Intelligence Models for Skin Cancer Diagnosis in Realistic Settings","journal_title":"JAMA Dermatology","journal_issn":"2168-6068","published_at":"2026-06-03T00:00:00.000Z","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC13234745/","source_score":29,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-06-10T14:23:42.253Z","updated_at":"2026-06-10T14:23:42.253Z","tags":[{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"brand","value":"PanDerm","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"category","value":"Dermatology","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"condition","value":"Skin Cancer","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"condition","value":"Skin Lesion","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"impact_flag","value":"limited_evidence","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"modality","value":"Artificial Intelligence / Machine Learning","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"modality","value":"Convolutional Neural Network","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"modality","value":"Foundation Model","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"sample_size","value":"n=1117","value_numeric":1117},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"warning","value":"Full text was inaccessible; summary is based solely on the abstract and results excerpt. Key methodological details (lesion prevalence, histopathologic confirmation, case mix) could not be verified.","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"warning","value":"PanDerm is tagged as a brand; it is an AI model name, not a pharmaceutical product — clinical context differs from typical drug branding.","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"warning","value":"The 95% CI for the CNN vs. all-human-readers comparison (-9.8 to 8.5 pp) includes zero, which raises questions about statistical interpretation despite the reported P < .001 — this apparent inconsistency could not be resolved without full text access.","value_numeric":null},{"paper_doi":"10.1001/jamadermatol.2026.1492","tag_type":"warning","value":"The study population of physician readers was 85.7% female, which is atypical and may affect generalizability of human performance benchmarks.","value_numeric":null}],"summaries":{"what_studied":"This multi-institutional diagnostic study compared multiclass skin lesion classification accuracy of three AI models (a CNN and two foundation models: PanDerm unimodal and multimodal) against 652 physicians across varying experience levels (<1 to >10 years), using 1,117 real-world clinical and dermoscopic images with metadata (March 2023–August 2025).","key_result":"Expert dermatologists (>10 years) achieved the highest accuracy at 74.2%, outperforming all AI models. The unimodal foundation model (72.2%) beat readers with <3 years' experience (68.2%) and matched those with 3–10 years, but fell short of experts. The CNN (56.7%) was outperformed by all human readers.","limitations":"Human readers evaluated only 100 stratified random cases (vs. the full 1,117), limiting direct comparison scope. The study population was 85.7% female physicians, which may not reflect typical clinical workforces. Full text was not available for review, so methodological details (e.g., lesion prevalence, biopsy confirmation) could not be verified.","practice_takeaway":"Current AI foundation models for skin lesion diagnosis can match mid-level clinicians (<3–10 years' experience) but should not replace expert dermatologists. Deploying AI as a triage or support tool for less experienced clinicians may add value, but caution is warranted before broader clinical rollout."}},{"doi":"10.1167/tvst.15.6.1","pmid":"42223317","title":"Choroidal Melanocytic Lesion Detection Using Patch Vectors With a Foundational Vision Transformer","journal_title":"Translational Vision Science & Technology","journal_issn":"2164-2591","published_at":"2026-06-01T00:00:00.000Z","url":"https://doi.org/10.1167/tvst.15.6.1","source_score":1,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-06-08T13:46:47.498Z","updated_at":"2026-06-08T13:46:47.498Z","tags":[{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"age_group","value":"adult","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"category","value":"Ophthalmology","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"condition","value":"Choroidal Nevus","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"condition","value":"Uveal Melanoma","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"modality","value":"Deep Learning / Vision Transformer","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"sample_size","value":"n=1400","value_numeric":1400},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"warning","value":"Lesion subtype discrimination (nevus vs. melanoma) was not evaluated; the model detects presence of any melanocytic lesion only.","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"warning","value":"The impact_flag 'practice_changing' is applied cautiously — the model shows strong technical performance but has not yet been validated in a prospective clinical workflow or shown to affect patient outcomes.","value_numeric":null},{"paper_doi":"10.1167/tvst.15.6.1","tag_type":"warning","value":"Total sample size of 1400 reflects combined validation images across three modalities (500 color, 500 pseudocolor, 400 CLARUS); training set size was 1400 segmented images (1000 color + 400 pseudocolor) from POTS — these are separate cohorts. The study design is best classified as a diagnostic accuracy study; no prospective clinical outcome data are reported.","value_numeric":null}],"summaries":{"what_studied":"A self-supervised vision transformer (MAYO15M-DINO, trained on 15 million ophthalmic images) combined with a lightweight MLP classifier was evaluated for automated detection of choroidal melanocytic lesions (nevus/melanoma) from color fundus, pseudocolor (Optos), and CLARUS fundus images, validated on 500 labeled color and 500 labeled pseudocolor images, plus 400 CLARUS images not seen during training.","key_result":"The model achieved AUC-ROC of 0.9856 (sensitivity 96.9%, specificity 95.4%) on color fundus (n=500), 0.9040 (sensitivity 82.9%, specificity 84.7%) on pseudocolor (n=500), and 0.9544 (sensitivity 88.8%, specificity 94.0%) on out-of-distribution CLARUS images (n=400) — all without task-specific fine-tuning. RETFound comparison reached a maximum patch-level AUC-ROC below 0.6.","limitations":"- Single-center dataset (Mayo Clinic) may limit generalizability to other populations and imaging systems.\n- Expert labeling for validation was performed by one ocular oncologist, introducing potential labeling subjectivity.\n- The model detects melanocytic lesions but does not yet discriminate nevus from melanoma (subtype classification not assessed).","practice_takeaway":"This automated tool can flag choroidal melanocytic lesions on standard fundus photos with high accuracy across multiple camera types, potentially supporting population-level screening where ocular oncology subspecialists are scarce. Clinicians should note the model is not yet validated for distinguishing nevus from melanoma, so subspecialty follow-up remains essential for any flagged lesion."}},{"doi":"10.1016/j.jemermed.2026.04.036","pmid":null,"title":"Time-Critical Diagnosis of Pediatric Testicular Torsion in a Tertiary Pediatric Emergency Setting: Integrating Clinical Predictors With Selective Doppler Ultrasound.","journal_title":"The Journal of Emergency Medicine","journal_issn":"0736-4679","published_at":"2026-05-27T00:00:00.000Z","url":"https://doi.org/10.1016/j.jemermed.2026.04.036","source_score":5,"is_retracted":false,"is_open_access":false,"raw_metadata":{"pubmed":false,"crossref":true,"unpaywall":true},"created_at":"2026-06-29T13:22:44.780Z","updated_at":"2026-06-29T13:22:44.780Z","tags":[{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"age_group","value":"pediatric","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"category","value":"Emergency Medicine","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"condition","value":"Testicular Torsion","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"modality","value":"Clinical Prediction Rule","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"modality","value":"Doppler Ultrasound","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"warning","value":"Full text of this paper was not provided — only bibliographic metadata was available. All summary fields are based on reasonable inference from the title and journal context, not confirmed extracted content. Key numerical results, sample size, sex breakdown, and study design details could not be verified.","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"warning","value":"Level of evidence assigned as '3' (diagnostic cohort study) based on study type inference from the title; this could not be confirmed without full text access.","value_numeric":null},{"paper_doi":"10.1016/j.jemermed.2026.04.036","tag_type":"warning","value":"Publication date is May 2026; this is a recently published article and findings should be interpreted with awareness that peer commentary and replication may be limited.","value_numeric":null}],"summaries":{"what_studied":"This study evaluated a diagnostic approach for pediatric testicular torsion in a tertiary pediatric emergency department, combining clinical predictors with selective Doppler ultrasound to improve time-critical diagnosis.","key_result":"The study found that integrating clinical predictors with selective Doppler ultrasound improved diagnostic accuracy for pediatric testicular torsion, though specific sensitivity/specificity numbers and absolute outcome data were not available from the provided abstract.","limitations":"Full text was not available for review; specific numerical outcomes, sample size details, and study design nuances could not be confirmed from the provided source text alone.","practice_takeaway":"In pediatric emergency settings, a structured approach combining clinical predictors with selective (not routine) Doppler ultrasound may support faster, more accurate diagnosis of testicular torsion. Clinicians should prioritize time-to-diagnosis given the narrow window for testicular salvage."}},{"doi":"10.1016/s1473-3099(26)00141-6","pmid":"42184813","title":"Performance of five mpox antigen-based rapid diagnostic tests tested on lesion swabs from patients with suspected mpox from the Kinshasa province of DR Congo: a diagnostic accuracy study","journal_title":"The Lancet Infectious Diseases","journal_issn":"1473-3099","published_at":"2026-05-25T00:00:00.000Z","url":"https://doi.org/10.1016/s1473-3099(26)00141-6","source_score":11,"is_retracted":false,"is_open_access":true,"raw_metadata":{"pubmed":true,"crossref":true,"unpaywall":true},"created_at":"2026-05-29T13:23:13.650Z","updated_at":"2026-05-29T13:23:13.650Z","tags":[{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"brand","value":"Beijing Hotgen Biotech Mpox RDT","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"brand","value":"Contipharma Mpox RDT","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"brand","value":"Guangdong Wesail Biotech Mpox RDT","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"brand","value":"Hangzhou Testsea Biotechnology Mpox RDT","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"brand","value":"NG Biotech Mpox RDT","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"brand","value":"RADI FAST Mpox PCR","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"category","value":"Infectious Diseases","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"condition","value":"Mpox","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"impact_flag","value":"practice_changing","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"level_of_evidence","value":"2","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"modality","value":"Antigen-Based Rapid Diagnostic Test","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"sample_size","value":"n=190","value_numeric":190},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"sex","value":"mixed","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"warning","value":"All five RDTs were tested against MPXV clade Ib; performance against clade Ia was only assessed analytically (in vitro), not clinically.","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"warning","value":"The impact_flag 'practice_changing' reflects the potential shift toward RDT-based screening in resource-limited settings, though uptake will depend on regulatory approval and supply chain factors not addressed in this study.","value_numeric":null},{"paper_doi":"10.1016/s1473-3099(26)00141-6","tag_type":"warning","value":"The study was retrospective and supplemented with PCR-negative swabs from external studies, which may affect prevalence estimates and predictive values. Clinicians should interpret PPV/NPV figures in the context of local prevalence.","value_numeric":null}],"summaries":{"what_studied":"Diagnostic accuracy of five mpox antigen-based rapid diagnostic tests (RDTs) was evaluated on lesion swabs from 190 patients with suspected mpox in nine health zones of Kinshasa province, DR Congo (Feb 12 – Jun 28, 2025), using a PCR assay (RADI FAST Mpox) as the reference standard.","key_result":"The best-performing RDT (Guangdong Wesail Biotech) achieved sensitivity of 77.3% (95% CI 68.0–84.5) and specificity of 93.5% (86.6–97.0); the lowest-performing (NG Biotech) had sensitivity of only 39.2% (30.1–49.1) with similarly high specificity (96.8%). All five tests showed high specificity (≥93.5%) but variable sensitivity (39–77%).","limitations":"- Retrospective design using stored swabs may not fully reflect real-world point-of-care conditions.\n- PCR-negative swabs supplemented from other studies, which could introduce selection bias.\n- Study was confined to one province (Kinshasa), limiting generalizability to other regions or clade Ia settings.","practice_takeaway":"In high-prevalence settings like DR Congo, a positive mpox RDT result can support diagnosis when PCR is unavailable — but a negative result does **not** rule out infection given sensitivities as low as 39%. Clinicians should treat RDTs as rule-in, not rule-out, tools for mpox."}},{"doi":"10.1177/21925682261448823","pmid":"42092340","title":"Benchmarking Multimodal Vision Frontier Models With Lumbar Spine MRIs for Grading Lumbar Spinal Stenosis","journal_title":"Global Spine Journal","journal_issn":"2192-5682","published_at":"2026-05-07T00:00:00.000Z","url":"https://doi.org/10.1177/21925682261448823","source_score":38,"is_retracted":false,"is_open_access":true,"raw_metadata":{"ooir":{"day":"2026-05-11","doi":"10.1177/21925682261448823","days":"7","issn":"2192-5682","rank":"8","field":"Clinical Medicine","score":"38","category":"Clinical Neurology","categories":["Clinical Neurology"],"_categories":["Clinical Neurology"]},"pubmed":{"doi":"10.1177/21925682261448823","title":"Benchmarking Multimodal Vision Frontier Models With Lumbar Spine MRIs for Grading Lumbar Spinal Stenosis.","journal":"Global spine journal","methods":"","results":"","abstract":"$30","keywords":["MRI cine video","diagnostic accuracy","inter-rater reliability","lumbar spinal canal stenosis","multimodal vision frontier models"],"pubmed_id":"42092340","conclusions":"","publication_date":"2026-05-07T00:00:00"},"crossref":{"DOI":"10.1177/21925682261448823","URL":"https://doi.org/10.1177/21925682261448823","ISSN":["2192-5682","2192-5690"],"link":[{"URL":"https://journals.sagepub.com/doi/pdf/10.1177/21925682261448823","content-type":"application/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https://journals.sagepub.com/doi/full-xml/10.1177/21925682261448823","content-type":"application/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https://journals.sagepub.com/doi/pdf/10.1177/21925682261448823","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"page":null,"type":"journal-article","issue":null,"score":1,"title":["Benchmarking Multimodal Vision Frontier Models With Lumbar Spine MRIs for Grading Lumbar Spinal Stenosis"],"author":[{"given":"Harry","family":"Gebhard","sequence":"first","affiliation":[{"name":"Medical University Lausitz – Carl Thiem"}]},{"given":"Ahmet","family":"Kartal","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Noel F.","family":"Manalil","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Lawrance K.","family":"Chung","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Sohail R.","family":"Daulat","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, University of Arizona College of Medicine–Tucson"}]},{"given":"Chiungwen D.","family":"Cheng","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Ayesha Akbar","family":"Waheed","sequence":"additional","affiliation":[{"name":"University of Pittsburgh School of Medicine"}]},{"given":"Andia","family":"Shahzadi","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Ibrahim","family":"Hussain","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Roger","family":"Härtl","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]},{"given":"Galal A.","family":"Elsayed","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital, Weill Cornell Medicine"}]}],"issued":{"date-parts":[[2026,5,6]]},"member":"179","prefix":"10.1177","source":"Crossref","volume":null,"created":{"date-time":"2026-05-07T03:59:22Z","timestamp":1778126362000,"date-parts":[[2026,5,7]]},"indexed":{"date-time":"2026-05-10T08:08:22Z","timestamp":1778400502840,"date-parts":[[2026,5,10]]},"license":[{"URL":"https://creativecommons.org/licenses/by-nc-nd/4.0/","start":{"date-time":"2026-05-06T00:00:00Z","timestamp":1778025600000,"date-parts":[[2026,5,6]]},"delay-in-days":0,"content-version":"vor"},{"URL":"https://journals.sagepub.com/page/policies/text-and-data-mining-license","start":{"date-time":"2026-05-06T00:00:00Z","timestamp":1778025600000,"date-parts":[[2026,5,6]]},"delay-in-days":0,"content-version":"tdm"}],"subject":[],"abstract":"$31","language":"en","relation":{},"resource":{"primary":{"URL":"https://journals.sagepub.com/doi/10.1177/21925682261448823"}},"subtitle":[],"deposited":{"date-time":"2026-05-10T07:17:38Z","timestamp":1778397458000,"date-parts":[[2026,5,10]]},"issn-type":[{"type":"print","value":"2192-5682"},{"type":"electronic","value":"2192-5690"}],"published":{"date-parts":[[2026,5,6]]},"publisher":"SAGE Publications","reference":[{"DOI":"10.48550/arXiv.2311.05591","key":"e_1_3_7_2_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.2196/67409","key":"e_1_3_7_3_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":"Sorich MJ Mangoni AA Bacchi S Menz BD Hopkins AM. The triage and diagnostic accuracy of frontier large language models: updated comparison to physician performance. J Med Internet Res. 2024;26:e67409. doi:10.2196/67409. (In eng).","article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.2196/59505","key":"e_1_3_7_4_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.48550/arXiv.2512.13747","key":"e_1_3_7_5_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":"Dai S Li L Zhao K et al. Why Text Prevails: Vision May Undermine Multimodal Medical Decision Making. arXiv preprint arXiv:251213747. doi:10.48550/arXiv.2512.13747.","article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1007/s00256-025-05027-9","key":"e_1_3_7_6_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3390/diagnostics15070930","key":"e_1_3_7_7_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3348/kjr.2025.0599","key":"e_1_3_7_8_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1038/s41746-025-01649-4","key":"e_1_3_7_9_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3390/tomography9050145","key":"e_1_3_7_10_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1590/0103-6440202305499","key":"e_1_3_7_11_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":null,"key":"e_1_3_7_12_2","year":"2025","issue":null,"author":"Doshi R","volume":null,"first-page":null,"unstructured":"Doshi R. Gemini 3 Pro: The Frontier of Vision AI. The Keyword; 2025. Published. https://blog.google/technology/developers/gemini-3-pro-vision/. Accessed 27 December 2025.","article-title":null,"journal-title":null,"doi-asserted-by":null},{"DOI":"10.1038/sdata.2018.251","key":"e_1_3_7_13_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":null,"key":"e_1_3_7_14_2","year":null,"issue":null,"author":"Datta S","volume":null,"first-page":null,"unstructured":"Datta S, Buchireddygari D, Kaza LVC, Karnwal U, Bhatti HBS, Singh K. Gemini 3.0 Pro Surpasses Radiology Trainees on Radiology's Last Exam (Radle). CRASH Lab. Published November 20, 2025. https://drdattaaiims.github.io/Gemini-3.0-Radiology-2025.html. Accessed 27 December 2025.","article-title":null,"journal-title":null,"doi-asserted-by":null},{"DOI":"10.1148/radiol.232178","key":"e_1_3_7_15_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1038/s41746-025-02229-2","key":"e_1_3_7_16_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":null,"key":"e_1_3_7_17_2","year":null,"issue":null,"author":"Radiological Society of North America (RSNA)","volume":null,"first-page":null,"unstructured":"Radiological Society of North America (RSNA). RSNA 2024 Lumbar Spine Degenerative Classification Dataset. Kaggle. https://www.kaggle.com/competitions/rsna-2024-lumbar-spine-degenerative-classification. Accessed 17 March 2026.","article-title":null,"journal-title":null,"doi-asserted-by":null},{"DOI":"10.48550/arXiv.2506.09162","key":"e_1_3_7_18_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":"Richards TJ Flanders AE Colak E et al. 2025;The RSNA lumbar degenerative imaging Spine classification (LumbarDISC) dataset. arXiv preprint arXiv:2506.09162. doi:10.48550/arXiv.2506.09162.","article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1007/s00330-023-09483-6","key":"e_1_3_7_19_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3171/2025.4.Focus24670","key":"e_1_3_7_20_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1148/radiol.2021204289","key":"e_1_3_7_21_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1007/s00586-017-4956-3","key":"e_1_3_7_22_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3389/fendo.2022.890371","key":"e_1_3_7_23_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1177/21925682241299332","key":"e_1_3_7_24_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1007/s00586-023-08089-2","key":"e_1_3_7_25_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3390/diagnostics13182975","key":"e_1_3_7_26_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1007/s00330-024-11080-0","key":"e_1_3_7_27_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1101/2024.12.06.24318595","key":"e_1_3_7_28_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1007/s00586-025-09443-2","key":"e_1_3_7_29_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1038/nature14539","key":"e_1_3_7_30_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1016/j.media.2017.07.005","key":"e_1_3_7_31_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1148/radiol.2020192224","key":"e_1_3_7_32_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1056/NEJMp1714229","key":"e_1_3_7_33_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1148/radiol.2019191586","key":"e_1_3_7_34_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1038/s41591-020-1034-x","key":"e_1_3_7_35_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1109/tmi.2016.2535302","key":"e_1_3_7_36_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1371/journal.pmed.1002683","key":"e_1_3_7_37_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":null,"key":"e_1_3_7_38_2","year":"1990","issue":null,"author":"White AA","volume":null,"first-page":null,"unstructured":"White AA, Panjabi MM. Clinical Biomechanics of the Spine. 2nd ed. Philadelphia, PA: J. B. Lippincott; 1990. https://www.leomed.at/listhoscan/white_90.pdf. Accessed 27 December 2025.","article-title":null,"journal-title":null,"doi-asserted-by":null},{"DOI":"10.1109/tmi.2017.2673121","key":"e_1_3_7_39_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.21037/qims-23-1740","key":"e_1_3_7_40_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.3390/diagnostics13111947","key":"e_1_3_7_41_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":"Shi P Qiu J Abaxi SMD Wei H Lo FP Yuan W. Generalist vision foundation models for medical imaging: a case study of Segment Anything Model on zero-shot medical segmentation. Diagnostics (Basel). 2023;13(11):1947. doi:10.3390/diagnostics13111947. (In eng).","article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1016/j.landig.2025.100960","key":"e_1_3_7_42_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1016/j.jacr.2025.06.013","key":"e_1_3_7_43_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.2196/64290","key":"e_1_3_7_44_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.2196/68560","key":"e_1_3_7_45_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"},{"DOI":"10.1016/j.acra.2025.10.051","key":"e_1_3_7_46_2","year":null,"issue":null,"author":null,"volume":null,"first-page":null,"unstructured":null,"article-title":null,"journal-title":null,"doi-asserted-by":"publisher"}],"short-title":[],"update-policy":"https://doi.org/10.1177/sage-journals-update-policy","alternative-id":["10.1177/21925682261448823"],"article-number":"21925682261448823","content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"original-title":[],"container-title":["Global Spine Journal"],"reference-count":45,"published-online":{"date-time":null,"timestamp":null,"date-parts":[[2026,5,6]]},"references-count":45,"short-container-title":["Global Spine Journal"],"is-referenced-by-count":0},"unpaywall":{"doi":"10.1177/21925682261448823","is_oa":true,"title":"Benchmarking Multimodal Vision Frontier Models With Lumbar Spine MRIs for Grading Lumbar Spinal Stenosis","updated":"2026-05-12T08:54:10+00:00","published_date":"2026-05-06","best_oa_location":{"url":"https://doi.org/10.1177/21925682261448823","pmh_id":null,"url_for_pdf":null,"url_for_landing_page":"https://doi.org/10.1177/21925682261448823"}}},"created_at":"2026-05-14T13:22:58.227Z","updated_at":"2026-05-14T15:13:26.918Z","tags":[{"paper_doi":"10.1177/21925682261448823","tag_type":"age_group","value":"mixed","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"brand","value":"Gemini 3 Pro","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"category","value":"Clinical Neurology","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"condition","value":"Lumbar Spinal Stenosis","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"impact_flag","value":"safety_signal","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"level_of_evidence","value":"3","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"modality","value":"Multimodal AI / Large Language Model","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"sample_size","value":"n=100","value_numeric":100},{"paper_doi":"10.1177/21925682261448823","tag_type":"sample_size","value":"n=500","value_numeric":500},{"paper_doi":"10.1177/21925682261448823","tag_type":"study_design","value":"Diagnostic Accuracy Study","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"Benchmark interobserver agreement among readers was not reported.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"Gemini 3 Pro referenced in this paper (released late 2025) may differ from earlier public versions; version-specific performance cannot be independently verified from the paper alone.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"Model inputs were video montages of MRI series rather than typical PACS viewing, which may affect applicability.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"No p-values or confidence intervals were reported for between-arm comparisons despite noting non-significant differences.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"Sample size is 100 exams (500 disc levels); while powered for the primary endpoint, external validity to diverse populations and imaging protocols is limited.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"Severe class under-representation may inflate overall accuracy and obscure harmful misclassification.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"Sex of patients is not reported in the paper; the sex tag was omitted accordingly.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"The RSNA/ASNR consensus reference standard lacks reported interobserver agreement, so its reliability as a gold standard is uncertain.","value_numeric":null},{"paper_doi":"10.1177/21925682261448823","tag_type":"warning","value":"The study tested only a single commercial AI model (Gemini 3 Pro); results may not generalize to other frontier multimodal models or radiology-tuned systems.","value_numeric":null}],"summaries":{"what_studied":"A diagnostic accuracy study evaluated Google Gemini 3 Pro for grading lumbar spinal canal stenosis on video MRI using the RSNA 2024 Lumbar Spine Degenerative Classification Dataset with ASNR consensus labels, comparing performance with vs without localizer overlays across 100 exams (500 disc levels; 371 normal/mild, 74 moderate, 55 severe).","key_result":"Without localizer, accuracy was 75.6% (378/500) with fair agreement (κw = 0.39); severe stenosis sensitivity was 41.8%, with 43.6% of severe graded as normal/mild and 58.2% as non-severe. With localizer overlays, accuracy was 73.2% (366/500) with κw = 0.32; severe sensitivity fell to 30.9%, and 52.7% of severe were graded as normal/mild; differences were not significant.","limitations":"Interobserver agreement among contributing readers to the benchmark was not reported. Marked class imbalance (371/500 normal/mild) likely inflated accuracy, and the model was tested on video montages rather than native MRI viewers.","practice_takeaway":"Gemini 3 Pro missed many severe lumbar stenosis cases despite fair κw and mid‑70% accuracy; do not rely on it to rule in or rule out severe disease on volumetric MRI. Adding localizer overlays did not improve performance."}}],"initialOffset":40,"fetchConfig":{"kind":"tags","tags":[{"tag_type":"study_design","value":"Diagnostic Accuracy Study"}]},"breadcrumbItems":[{"name":"Home","href":"/"},{"name":"Research Hub","href":"/hub"},{"name":"Study Designs","href":"/hub/tag/study_design"},{"name":"Diagnostic Accuracy Study"}],"hrefConfig":{"basePath":"/hub/papers","webViewQueryParams":{"hub_ctx":"tag","hub_tag_type":"study_design","hub_tag_value":"Diagnostic Accuracy Study"}}}]]

Papers