{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002883046,0.0001080743,0.0002994493,0.0002672447,0.0001955399,0.000005575838,0.0001555467,0.0001673571,0.0000265082],"category_scores_gemma":[0.001926603,0.00006987105,0.0000491486,0.001223503,0.0002269359,0.0000254739,0.00001396001,0.0004460916,0.000009089488],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001989278,"about_ca_system_score_gemma":0.001209862,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003317206,"about_ca_topic_score_gemma":0.0005807237,"domain_scores_codex":[0.9981263,0.00006161583,0.0007417977,0.000237012,0.0005104651,0.0003227964],"domain_scores_gemma":[0.9977409,0.001179819,0.0001722861,0.0004182808,0.0003590116,0.0001297337],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0007722075,0.0001892578,0.00135486,0.0008286355,0.00002483896,0.00000306038,0.001464391,0.000223526,0.0005547201,0.05757726,0.06726631,0.869741],"study_design_scores_gemma":[0.0003657035,0.002826248,0.005805228,0.001709383,0.000103096,0.00001116276,0.01129273,0.0251966,0.04373591,0.1038857,0.8047656,0.00030272],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.06329823,0.01273381,0.1032439,0.8049915,0.003557468,0.009658524,0.001553411,0.0002897136,0.0006733443],"genre_scores_gemma":[0.9926462,0.0006808186,0.0006972416,0.003207895,0.0009434262,0.0004027542,0.001343162,0.0000208172,0.00005767017],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.929348,"threshold_uncertainty_score":0.2849259,"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."}}