{"id":"W4379163058","doi":"10.2196/preprints.48995","title":"BERT-Based Neural Network for Inpatient Fall Detection From Electronic Medical Records: Retrospective Cohort Study (Preprint)","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; Alberta Health Services","funders":"","keywords":"Machine learning; Artificial intelligence; Computer science; Artificial neural network; Electronic health record; Preprint; Medical record; Diagnosis code; Inpatient care; Medicine; Medical emergency; Natural language processing; Health care; World Wide Web; Internal medicine","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.003136609,0.0004498417,0.0002915079,0.0008290416,0.0005616155,0.0008310798,0.0009576652,0.0005541666,0.00232221],"category_scores_gemma":[0.010068,0.0004768887,0.000844968,0.001166157,0.0002777989,0.0006301692,0.0006231204,0.001044773,0.0006056444],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002054635,"about_ca_system_score_gemma":0.002072287,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.115205,"about_ca_topic_score_gemma":0.1337518,"domain_scores_codex":[0.9989412,0.0002659456,0.0001154393,0.0002935681,0.0002379456,0.0001459845],"domain_scores_gemma":[0.9959907,0.0009995507,0.00078966,0.0007933216,0.00104832,0.000378432],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003583316,0.00008479467,0.9945141,0.00002064569,0.0001425951,0.00005896122,0.00007939368,0.0002855872,0.00007878004,0.00006086616,0.001090454,0.003225386],"study_design_scores_gemma":[0.00007693958,0.0003217488,0.9888005,0.00004360535,0.0002489421,0.0002763736,0.000350862,0.008591956,0.0001623859,0.0001780869,0.0009201014,0.00002836664],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9946151,0.0001480967,0.001236643,0.0001328632,0.00002172144,0.00007443981,0.003403835,0.00002128313,0.0003459521],"genre_scores_gemma":[0.9927522,0.0001339514,0.001631147,0.00008145349,0.00002467053,0.0001327599,0.004619666,0.00001377912,0.0006105069],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.115205,"threshold_uncertainty_score":0.2290689,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02140620981354922,"score_gpt":0.304735484503432,"score_spread":0.2833292746898828,"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."}}