{"id":"W4409605504","doi":"10.2196/70924","title":"Using a Large Language Model for Breast Imaging Reporting and Data System Classification and Malignancy Prediction to Enhance Breast Ultrasound Diagnosis: Retrospective Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"AI in cancer detection","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"BI-RADS; Preprint; Malignancy; Breast imaging; Ultrasound; Medicine; Computer science; Radiology; Mammography; Breast cancer; Pathology; Cancer; World Wide Web; Internal medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0196924,0.0008493747,0.0006107873,0.002196593,0.0005239137,0.001381462,0.001270243,0.000753816,0.001353409],"category_scores_gemma":[0.0620262,0.0006539367,0.00198495,0.001479756,0.0007331829,0.00178959,0.001237537,0.00149001,0.0005266311],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001795151,"about_ca_system_score_gemma":0.001742914,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007339629,"about_ca_topic_score_gemma":0.00596554,"domain_scores_codex":[0.9898577,0.006212227,0.0008037655,0.001315805,0.001395463,0.0004149823],"domain_scores_gemma":[0.9348348,0.04422516,0.008094409,0.006638492,0.004914928,0.001292294],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0004827958,0.0002607884,0.9910883,0.00002767515,0.0002144289,0.0001353802,0.0001556882,0.001673438,0.0001379427,0.00006722462,0.0002356944,0.005520557],"study_design_scores_gemma":[0.0001750626,0.003127446,0.7714307,0.0001338117,0.001188281,0.002543206,0.001368663,0.216073,0.001450734,0.0008200328,0.001569997,0.0001190744],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9922748,0.0001629573,0.00647553,0.0001326648,0.00001847772,0.000151039,0.0004834127,0.0000506625,0.0002505065],"genre_scores_gemma":[0.9958768,0.00004410771,0.00311622,0.00004356554,0.00001478108,0.0001173908,0.0006659181,0.00001721015,0.0001039435],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0196924,"threshold_uncertainty_score":0.1041446,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03103692471142467,"score_gpt":0.3650558444411161,"score_spread":0.3340189197296914,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}