{"id":"W4319454867","doi":"10.1101/2023.02.04.23285478","title":"ChatGPT for Clinical Vignette Generation, Revision, and Evaluation","year":2023,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Clinical Reasoning and Diagnostic Skills","field":"Medicine","cited_by":82,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta Hospital; University of Alberta","funders":"","keywords":"Vignette; Rewriting; Triage; Set (abstract data type); Medicine; Perspective (graphical); Literacy; Psychology; Psychiatry; Artificial intelligence; Computer science; Social psychology; Programming language","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.006236599,0.0002152745,0.0006702234,0.00008944834,0.00007643037,0.00004968935,0.00009897412,0.0004978248,0.00008473972],"category_scores_gemma":[0.1578281,0.0001790421,0.0002618621,0.00009112646,0.00009558378,0.00001795727,0.0002052655,0.0004698093,0.00008280275],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003768114,"about_ca_system_score_gemma":0.0003475743,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001478044,"about_ca_topic_score_gemma":0.000009007306,"domain_scores_codex":[0.9973259,0.0002183381,0.0009408001,0.0008073123,0.0004943525,0.0002133111],"domain_scores_gemma":[0.9887364,0.009330944,0.000292588,0.0006902917,0.0006715475,0.0002782171],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0002109733,0.0003596323,0.4353725,0.0004621524,0.0003419702,0.00002877732,0.0002254726,0.0001156807,0.00009936231,0.0002824083,0.2939451,0.268556],"study_design_scores_gemma":[0.005359747,0.0008086019,0.8406899,0.00620951,0.002197919,0.00001844782,0.00002351828,0.1006896,0.0001319743,0.006310421,0.03701798,0.000542326],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9763666,0.002934967,0.002564113,0.01266936,0.002987727,0.00214644,0.00004578567,0.0001505215,0.0001345372],"genre_scores_gemma":[0.9576254,0.01389849,0.00913489,0.003583789,0.00932608,0.0007272733,0.002458334,0.0001450672,0.003100631],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4053174,"threshold_uncertainty_score":0.8492658,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2538464383166623,"score_gpt":0.4922916759630908,"score_spread":0.2384452376464285,"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."}}