{"id":"W4381572165","doi":"10.2196/44331","title":"Optimizing Patient Record Linkage in a Master Patient Index Using Machine Learning: Algorithm Development and Validation","year":2023,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hamilton Health Sciences; McMaster University; Population Health Research Institute; University of Toronto","funders":"Strong; Bill and Melinda Gates Foundation; United States Agency for International Development","keywords":"Computer science; Matching (statistics); Machine learning; Population; Software; Linkage (software); Record linkage; Artificial intelligence; Interface (matter); Bayesian optimization; Set (abstract data type); Data mining; Programming language; Medicine; Operating system","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"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":[0.01243754,0.001528139,0.001278124,0.002323569,0.001019223,0.001530141,0.002903967,0.001736925,0.0022761],"category_scores_gemma":[0.03694769,0.000900014,0.001168224,0.002129853,0.0009377457,0.001423163,0.001965604,0.001970128,0.0008585815],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002447515,"about_ca_system_score_gemma":0.004264635,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01006249,"about_ca_topic_score_gemma":0.00694179,"domain_scores_codex":[0.9949857,0.002312803,0.000391695,0.001217368,0.0008403846,0.0002520901],"domain_scores_gemma":[0.9800518,0.01451172,0.001116312,0.001799815,0.002173743,0.0003466094],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006099157,0.0006625018,0.02018025,0.0003059329,0.0003512777,0.0001671609,0.0002541619,0.7175884,0.003102239,0.002129243,0.005623543,0.2490253],"study_design_scores_gemma":[0.00009920041,0.0001011638,0.001838519,0.00002441336,0.00002788579,0.00006363083,0.00003083724,0.9910143,0.003987824,0.001922641,0.0008707125,0.00001891394],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2228469,0.0006873513,0.7535523,0.0009107583,0.00009005723,0.0009584618,0.001096743,0.01781921,0.002038107],"genre_scores_gemma":[0.3194998,0.000149802,0.6757693,0.0002896171,0.00003687633,0.0009617317,0.002023709,0.0004644252,0.0008046576],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01243754,"threshold_uncertainty_score":0.06577677,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3302761360581266,"score_gpt":0.4740721920281339,"score_spread":0.1437960559700073,"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."}}