{"id":"W4312036666","doi":"10.2196/preprints.44331","title":"Optimizing Patient Record Linkage in a Master Patient Index Using Machine Learning: Algorithm Development and Validation (Preprint)","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hamilton Health Sciences; McMaster University; Population Health Research Institute; University of Toronto","funders":"","keywords":"Computer science; Matching (statistics); Record linkage; Population; Software; Machine learning; Linkage (software); Interface (matter); Artificial intelligence; Set (abstract data type); Data mining; Programming language; Medicine; Operating system","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":["metaepi_narrow","open_science","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.005329957,0.0004283621,0.0006329089,0.001021737,0.0003136504,0.0008581755,0.0008090555,0.0001931109,0.002463829],"category_scores_gemma":[0.0006795306,0.0003827459,0.000113218,0.0005250188,0.00005195279,0.0003690624,0.01196875,0.001185256,0.00004611078],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004958296,"about_ca_system_score_gemma":0.0001849917,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001363297,"about_ca_topic_score_gemma":0.0003128739,"domain_scores_codex":[0.992633,0.001155114,0.002036447,0.001672287,0.00207453,0.0004285534],"domain_scores_gemma":[0.9972724,0.0004256201,0.0009918343,0.0009973553,0.0001567246,0.0001560623],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004627971,0.0002250374,0.01408916,0.00005902915,0.00005251629,0.0000513101,0.01266206,0.07284756,0.000008300683,0.00008901961,0.0001501346,0.8997196],"study_design_scores_gemma":[0.0009884603,0.0002418745,0.004139515,0.0002955887,0.00003563958,0.000009511429,0.01992134,0.6369343,0.0007026095,0.005282512,0.3300926,0.001356008],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.6575446,0.0001726222,0.33611,0.0003042965,0.001105391,0.001543618,0.00006717451,0.0001005623,0.003051754],"genre_scores_gemma":[0.4539741,0.0001561819,0.5415155,0.0008520401,0.00005738005,0.0003799467,0.0006750537,0.0000721861,0.002317615],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8983636,"threshold_uncertainty_score":0.9998624,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1697825093542959,"score_gpt":0.3599210885867782,"score_spread":0.1901385792324823,"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."}}