{"id":"W4283012421","doi":"10.1038/s41597-022-01423-1","title":"A dataset of simulated patient-physician medical interviews with a focus on respiratory cases","year":2022,"lang":"en","type":"article","venue":"Scientific Data","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"Tellabs (Canada); University of Waterloo; Polytechnique Montréal; Western University","funders":"","keywords":"Computer science; Conversation; Artificial intelligence; Focus (optics); Natural language processing; Set (abstract data type); Converse; Machine learning; Psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001228609,0.0006775219,0.0005034858,0.00103983,0.0006825943,0.0006590692,0.001197615,0.00225145,0.004838457],"category_scores_gemma":[0.005869254,0.0001831635,0.0005421198,0.001149346,0.0004456006,0.0004671047,0.0009344594,0.0008027412,0.002620764],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001076891,"about_ca_system_score_gemma":0.001078785,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009292213,"about_ca_topic_score_gemma":0.01866661,"domain_scores_codex":[0.9981883,0.0009519371,0.0001582053,0.0003027979,0.0002568006,0.0001418719],"domain_scores_gemma":[0.9955349,0.002796391,0.0002594582,0.0004803619,0.000515009,0.0004138274],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.005045532,0.006504293,0.1811896,0.00524164,0.0007305542,0.01303061,0.008023771,0.05992502,0.009213215,0.005588054,0.5620433,0.1434644],"study_design_scores_gemma":[0.0009521241,0.003181631,0.2945705,0.001425826,0.0003520544,0.01056768,0.02124857,0.1578553,0.01256589,0.0091586,0.4876689,0.0004529832],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"empirical","genre_gemma":"dataset","genre_scores_codex":[0.6078508,0.002081299,0.01618553,0.004269673,0.0009174111,0.001605148,0.3506335,0.00148523,0.01497137],"genre_scores_gemma":[0.5901259,0.00074607,0.02354062,0.001115361,0.0003001399,0.001749254,0.3763774,0.00009163376,0.005953609],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.009292213,"threshold_uncertainty_score":0.01847625,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0875093153622667,"score_gpt":0.3520148190788424,"score_spread":0.2645055037165757,"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."}}