{"id":"W4400251873","doi":"10.1016/j.jacr.2024.06.018","title":"The Impact of Large Language Model-Generated Radiology Report Summaries on Patient Comprehension: A Randomized Controlled Trial","year":2024,"lang":"en","type":"article","venue":"Journal of the American College of Radiology","topic":"Radiology practices and education","field":"Medicine","cited_by":25,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Department of Surgery, University of Manitoba; University of Vermont","keywords":"Randomized controlled trial; Medical physics; Comprehension; Medicine; Computer science; Radiology; Internal medicine; Programming language","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003051264,0.0002030686,0.002687642,0.0002279333,0.0001219924,0.000007859313,0.0002358523,0.0001054434,0.0000282352],"category_scores_gemma":[0.003660088,0.00008353104,0.001356225,0.000362626,0.001079529,0.0000528965,0.00003884709,0.000523563,9.138065e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001495436,"about_ca_system_score_gemma":0.001251442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006239101,"about_ca_topic_score_gemma":0.0000048376,"domain_scores_codex":[0.9956757,0.002043661,0.001587311,0.0001752035,0.0002475929,0.0002704991],"domain_scores_gemma":[0.9915211,0.004680694,0.002918861,0.0004295366,0.0003581971,0.00009166542],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"randomized_trial","study_design_gemma":"randomized_trial","study_design_scores_codex":[0.9493626,0.0002924544,0.0002652055,0.00001649156,0.003330356,0.0003486944,0.0006185783,0.00354741,0.005672071,0.001097856,0.03510902,0.0003392749],"study_design_scores_gemma":[0.8847866,0.0148896,0.001087618,0.0002250427,0.002383033,0.03420632,0.002254735,0.05598053,0.000639885,0.0006577915,0.002672668,0.0002161805],"study_design_candidate":"randomized_trial","study_design_consensus":"randomized_trial","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9839597,0.005700562,0.0001197052,0.006564386,0.001947555,0.001360464,0.00004902412,0.000009217057,0.0002894214],"genre_scores_gemma":[0.9972879,0.001234077,0.0002971119,0.0002532907,0.0004727712,0.00003186539,0.000008123156,0.00001806417,0.0003967463],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06457599,"threshold_uncertainty_score":0.4381732,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01341502571706059,"score_gpt":0.3384312915663722,"score_spread":0.3250162658493116,"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."}}