{"id":"W4404667492","doi":"10.2196/60095","title":"Developing an ICD-10 Coding Assistant: Pilot Study Using RoBERTa and GPT-4 for Term Extraction and Description-Based Code Selection","year":2024,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Coding (social sciences); Selection (genetic algorithm); Code (set theory); Term (time); Computer science; Statistics; Mathematics; Artificial intelligence; World Wide Web; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007671332,0.0009101961,0.000587051,0.001387642,0.0005707584,0.000988553,0.001642342,0.0008745989,0.006516016],"category_scores_gemma":[0.02151906,0.0003241536,0.0007568752,0.001168583,0.0004760891,0.001309554,0.001957598,0.001781369,0.003475549],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001447261,"about_ca_system_score_gemma":0.00260359,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01298049,"about_ca_topic_score_gemma":0.01422604,"domain_scores_codex":[0.9970837,0.001379258,0.0002445301,0.000747991,0.0003800748,0.0001643511],"domain_scores_gemma":[0.9859119,0.008563716,0.0003871259,0.001795058,0.002696585,0.0006455979],"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":[0.002317895,0.003316548,0.03687867,0.001580762,0.0002117378,0.002305513,0.005655663,0.02158242,0.03173775,0.002055396,0.085862,0.8064957],"study_design_scores_gemma":[0.003795424,0.005514432,0.1008268,0.0005421935,0.0005416414,0.003033734,0.009341121,0.5419566,0.1269991,0.006212653,0.2007477,0.0004885145],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7812924,0.000723536,0.1514527,0.001797176,0.0006095283,0.004699996,0.02414674,0.02901691,0.006261101],"genre_scores_gemma":[0.5358311,0.0004381171,0.3722708,0.0008691289,0.000159343,0.003598842,0.07664649,0.001472951,0.008713236],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01298049,"threshold_uncertainty_score":0.04057032,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1830089616067131,"score_gpt":0.4688349562210808,"score_spread":0.2858259946143677,"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."}}