{"id":"W2994923354","doi":"10.1186/s12911-019-0961-2","title":"Incorporating medical code descriptions for diagnosis prediction in healthcare","year":2019,"lang":"en","type":"article","venue":"BMC Medical Informatics and Decision Making","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cisco Systems (Canada)","funders":"Nvidia; National Science Foundation","keywords":"Interpretability; Computer science; Diagnosis code; Health informatics; Code (set theory); Machine learning; Artificial intelligence; Convolutional neural network; Medical diagnosis; Source code; Embedding; Task (project management); Predictive modelling; Health care; Data mining; Population; Medicine","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":[],"consensus_categories":[],"category_scores_codex":[0.003097129,0.0001456156,0.0002949444,0.0002755327,0.0001856225,0.0001639441,0.0006188572,0.0002942912,0.00007455263],"category_scores_gemma":[0.003460666,0.0001229749,0.0000556722,0.0004144808,0.00004432407,0.0005152767,0.0004638253,0.0005321556,0.00002972401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008083582,"about_ca_system_score_gemma":0.0005843014,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005036412,"about_ca_topic_score_gemma":0.0005591725,"domain_scores_codex":[0.9967674,0.0001068142,0.001127747,0.0002434147,0.001410042,0.000344602],"domain_scores_gemma":[0.9949864,0.003829143,0.0002739403,0.0003869689,0.0001350449,0.0003885208],"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":[0.00002004697,0.00003108556,0.3479552,0.0005753803,0.000003043013,0.000004604227,0.0006740265,0.001673128,9.41215e-8,0.03319658,0.000700407,0.6151664],"study_design_scores_gemma":[0.0007375356,0.0001296205,0.007905776,0.001417275,0.000001577913,0.00004195507,0.0001828658,0.9756732,5.117033e-7,0.01288881,0.0009062495,0.0001146475],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1659605,0.0001533105,0.8316741,0.0008950749,0.0006115875,0.000438164,0.000008195271,0.00009071118,0.0001683644],"genre_scores_gemma":[0.5646887,0.00006293703,0.4330227,0.002045895,0.00007133359,0.0000861816,0.000008428191,0.000008104718,0.000005642305],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.974,"threshold_uncertainty_score":0.5014772,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04610601280114218,"score_gpt":0.3593525320887804,"score_spread":0.3132465192876382,"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."}}