{"id":"W4406833305","doi":"10.2196/57275","title":"Large Language Model Approach for Zero-Shot Information Extraction and Clustering of Japanese Radiology Reports: Algorithm Development and Validation","year":2025,"lang":"en","type":"article","venue":"JMIR Cancer","topic":"Radiology practices and education","field":"Medicine","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Cluster analysis; Computer science; Information extraction; Artificial intelligence; Machine learning; Natural language processing; Data mining; Information retrieval","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":[0.004858026,0.001564851,0.001345689,0.002204108,0.001261439,0.001501179,0.002545134,0.001870134,0.001660108],"category_scores_gemma":[0.01053611,0.0005469095,0.001636117,0.001365387,0.00063464,0.00164057,0.00169418,0.002509653,0.001090206],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002051557,"about_ca_system_score_gemma":0.003086884,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02064329,"about_ca_topic_score_gemma":0.02168537,"domain_scores_codex":[0.9979582,0.0007739221,0.0002110088,0.0006140076,0.0002715112,0.0001713913],"domain_scores_gemma":[0.9945207,0.003439604,0.0002898243,0.0003939261,0.001196752,0.0001591727],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008345917,0.0009925935,0.008873768,0.0004301066,0.000509011,0.000401688,0.0007077591,0.3227082,0.009021322,0.003461015,0.007775716,0.6442842],"study_design_scores_gemma":[0.00002404924,0.00004247796,0.0005542089,0.00000835084,0.00002770163,0.00003410239,0.00008806944,0.9961112,0.001735405,0.0009881533,0.0003747049,0.00001146993],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1525632,0.001097009,0.8346871,0.0006616804,0.0001342433,0.0006596365,0.0008739656,0.007879929,0.00144329],"genre_scores_gemma":[0.4050306,0.000308729,0.5872131,0.0003907877,0.00006778959,0.001003604,0.003945352,0.0002583427,0.001781758],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02064329,"threshold_uncertainty_score":0.04104626,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02792160276951978,"score_gpt":0.3638044400820019,"score_spread":0.3358828373124821,"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."}}