{"id":"W4410157449","doi":"10.2196/67837","title":"Natural Language Processing and ICD-10 Coding for Detecting Bleeding Events in Discharge Summaries: Comparative Cross-Sectional Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung","keywords":"Preprint; ICD-10; Computer science; Coding (social sciences); Medicine; Natural language processing; World Wide Web; Statistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001322539,0.0001506822,0.0002582904,0.0002244948,0.00038961,0.0002289534,0.0004790543,0.0001093729,0.000009134064],"category_scores_gemma":[0.0007595435,0.0001290402,0.00003211245,0.0004790849,0.00006415408,0.0007916945,0.0004024045,0.0006727569,0.00000257073],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001143061,"about_ca_system_score_gemma":0.000206376,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004486386,"about_ca_topic_score_gemma":0.0001056998,"domain_scores_codex":[0.9981269,0.00007005366,0.0007285827,0.0001872856,0.0005433169,0.0003438455],"domain_scores_gemma":[0.9987648,0.0006400304,0.0002060259,0.0001546142,0.0001216508,0.0001128476],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004243044,0.0001027804,0.8880037,0.001210176,0.00002508524,0.000003540686,0.08285654,0.00007685406,0.000007191005,0.001315299,0.00005659371,0.0262998],"study_design_scores_gemma":[0.001046189,0.00008194263,0.2159653,0.0003086945,0.000002449279,0.000007711808,0.00754028,0.7746921,0.00001675251,0.00009770767,0.0001121137,0.0001288177],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9600788,0.0001503879,0.0379203,0.0001979772,0.0003474455,0.0008472693,0.000002923131,0.0001228199,0.0003320444],"genre_scores_gemma":[0.9939094,0.000001194941,0.005494514,0.0002612016,0.00006162471,0.0001318225,0.00000858126,0.000004810301,0.000126844],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7746152,"threshold_uncertainty_score":0.5262105,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03010248228020434,"score_gpt":0.408459988291995,"score_spread":0.3783575060117906,"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."}}