{"id":"W4411857731","doi":"10.1186/s12911-025-03061-0","title":"Evaluating the performance of artificial intelligence-based speech recognition for clinical documentation: a systematic review","year":2025,"lang":"en","type":"review","venue":"BMC Medical Informatics and Decision Making","topic":"Electronic Health Records Systems","field":"Health Professions","cited_by":60,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; McGill University Health Centre","funders":"","keywords":"Documentation; Health informatics; Computer science; Artificial intelligence; Natural language processing; Speech recognition; Medicine; Public health; Pathology","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":["metaresearch"],"category_scores_codex":[0.04014418,0.0003685089,0.00344426,0.0002660239,0.0006476867,0.00003943795,0.0005994607,0.0007267076,0.000250193],"category_scores_gemma":[0.02799014,0.0002067515,0.0005339955,0.000659879,0.0001082937,0.0001429158,0.0002023691,0.001327393,0.00009737581],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001844204,"about_ca_system_score_gemma":0.007951264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004736826,"about_ca_topic_score_gemma":0.00003687932,"domain_scores_codex":[0.9838175,0.002671278,0.0112058,0.0002768763,0.001489964,0.000538609],"domain_scores_gemma":[0.9163532,0.07742577,0.004796159,0.0006333083,0.0006151382,0.0001764182],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"systematic_review","study_design_scores_codex":[0.00001857383,0.00001005005,0.000001567954,0.4956455,0.00002008622,1.15015e-7,0.00003478949,6.505562e-7,1.674167e-10,0.00006088409,0.0003104992,0.5038972],"study_design_scores_gemma":[0.0001759153,0.0002805482,1.317788e-7,0.8520691,0.001056786,0.00001022826,0.0003018666,0.1385738,2.334989e-8,0.0005291474,0.00683934,0.000163022],"study_design_candidate":"systematic_review","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00004086238,0.8014576,0.1858322,0.00002207699,0.001102662,0.01135495,0.00002983278,0.0000273797,0.0001324592],"genre_scores_gemma":[0.000002477258,0.9171008,0.0786253,0.001408678,0.0002938501,0.002429282,0.00008818741,0.00002302855,0.00002838257],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.5037342,"threshold_uncertainty_score":0.9976727,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.534462879473,"score_gpt":0.6514211515766116,"score_spread":0.1169582721036116,"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."}}