{"id":"W4387966489","doi":"10.1038/s41591-023-02608-w","title":"The value of standards for health datasets in artificial intelligence-based applications","year":2023,"lang":"en","type":"article","venue":"Nature Medicine","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":271,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; Google (Canada); Vector Institute; Genome Canada; Hospital for Sick Children","funders":"National Institute on Minority Health and Health Disparities; Medical Research Council; Great Ormond Street Institute of Child Health; Oxford University Hospitals NHS Foundation Trust; Hospital for Sick Children; Linköpings Universitet; Moorfields Eye Hospital NHS Foundation Trust; University of Oxford; Nuclear Power Institute of China; University College London; University of Leeds; National Institute for Health and Care Research; Government of the United Kingdom; European Commission; University Hospitals Birmingham NHS Foundation Trust; Wellcome Trust; Birmingham Biomedical Research Centre; UK Research and Innovation; Massachusetts Institute of Technology","keywords":"Best practice; Data science; Diversity (politics); Health care; Computer science; Transparency (behavior); Stakeholder; Equity (law); Knowledge management; Set (abstract data type); Health equity; Management science; Public relations; Political science; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.6396646,0.00149622,0.003290636,0.01666534,0.006870376,0.03590678,0.01353618,0.01319431,0.002099545],"category_scores_gemma":[0.7377541,0.00257215,0.004353138,0.01769166,0.03302901,0.05290517,0.02976837,0.02626171,0.001254553],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0233656,"about_ca_system_score_gemma":0.06977214,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01004115,"about_ca_topic_score_gemma":0.007912511,"domain_scores_codex":[0.2903119,0.5148094,0.0953707,0.02000828,0.075705,0.00379462],"domain_scores_gemma":[0.09313326,0.6194893,0.03366709,0.1667943,0.08194698,0.004968985],"domain_codex":"methods","domain_gemma":"methods","domain_candidate":"methods","domain_consensus":"methods","study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001754344,0.0001979695,0.007318426,0.009019447,0.0005349944,0.0002316573,0.0174257,0.004870525,0.001233796,0.7588277,0.03521927,0.1649451],"study_design_scores_gemma":[0.0001322003,0.000152726,0.004982831,0.03245224,0.0002744543,0.0004327143,0.007164272,0.005619602,0.001884326,0.7114661,0.235143,0.0002955254],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01169669,0.02810059,0.5650133,0.3542137,0.004861261,0.003347291,0.002494711,0.00126943,0.02900311],"genre_scores_gemma":[0.1350453,0.01015855,0.8086542,0.03261628,0.001922938,0.005653492,0.004023822,0.0006766972,0.001248756],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.3603354,"threshold_uncertainty_score":0.4443577,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1310099043928624,"score_gpt":0.5244179764861818,"score_spread":0.3934080720933194,"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."}}