{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004215484,0.002359624,0.004892739,0.0005920062,0.0007344441,0.002023821,0.001721765,0.004261768,0.0118638],"category_scores_gemma":[0.01394424,0.001149718,0.003383743,0.0005922655,0.002180716,0.002717367,0.0008351922,0.005459733,0.001356892],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001270668,"about_ca_system_score_gemma":0.001645371,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001188079,"about_ca_topic_score_gemma":0.001295701,"domain_scores_codex":[0.9961579,0.002230754,0.0004033819,0.0007266473,0.0002478877,0.0002334478],"domain_scores_gemma":[0.9892322,0.006847974,0.002257896,0.0005328812,0.0003363884,0.0007926042],"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":[0.9785699,0.008848933,0.0002781824,0.001596435,0.001486105,0.00003081621,0.00008505456,0.0001774232,0.0005828605,0.00009900953,0.0005366235,0.007708797],"study_design_scores_gemma":[0.9294909,0.06641603,0.0008019691,0.0001451355,0.001789719,0.00001296367,0.0000535584,0.0004833253,0.0002111423,0.0002834796,0.0002843108,0.00002747543],"study_design_candidate":"randomized_trial","study_design_consensus":"randomized_trial","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9704897,0.01146213,0.001810582,0.001945172,0.002097816,0.008300344,0.001025261,0.0003186637,0.002550393],"genre_scores_gemma":[0.9789109,0.003282858,0.002967702,0.001616389,0.001108425,0.00909745,0.0005124378,0.00004517542,0.002458734],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0118638,"threshold_uncertainty_score":0.03968835,"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."}}