{"id":"W2990422126","doi":"10.1016/j.jbi.2020.103489","title":"Towards unstructured mortality prediction with free-text clinical notes","year":2020,"lang":"en","type":"preprint","venue":"Journal of Biomedical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Montréal","funders":"","keywords":"Unstructured data; Computer science; Preprocessor; Raw data; Artificial intelligence; Data pre-processing; Machine learning; Data mining; Big data","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":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.002516927,0.0003608718,0.001039326,0.0002973076,0.00009617411,0.0003361779,0.003183783,0.0007648011,0.00003345995],"category_scores_gemma":[0.002853259,0.0002435108,0.0003679898,0.0004827315,0.000356712,0.0005943131,0.002071008,0.004668498,0.0000108326],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001500278,"about_ca_system_score_gemma":0.002226177,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005593108,"about_ca_topic_score_gemma":0.000004847119,"domain_scores_codex":[0.9931912,0.0003214183,0.003540816,0.0002365604,0.002346993,0.0003629621],"domain_scores_gemma":[0.9933729,0.0004043931,0.003493292,0.001106644,0.000691208,0.0009315162],"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":[0.0004543102,0.0007882099,0.09669755,0.01004981,0.003036103,0.0009328663,0.02272544,0.007994944,0.00001281891,0.007385905,0.06680857,0.7831135],"study_design_scores_gemma":[0.002786592,0.003562005,0.1244918,0.001767724,0.0002991704,0.001081179,0.0002640417,0.8176144,0.00003716533,0.02284624,0.02452458,0.0007251736],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0476688,0.0001210944,0.9304779,0.01611307,0.00473763,0.0002933969,0.00009056774,0.0001467147,0.000350866],"genre_scores_gemma":[0.4435523,0.0002654404,0.5514051,0.002392755,0.00229321,0.000004513142,0.00005591709,0.00002573403,0.00000495357],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8096194,"threshold_uncertainty_score":0.9976278,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06376128820391279,"score_gpt":0.3713350145335206,"score_spread":0.3075737263296078,"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."}}