{"id":"W4400606045","doi":"10.1038/s41598-024-65645-6","title":"Identifying symptom etiologies using syntactic patterns and large language models","year":2024,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University Health Centre","funders":"H2020 European Research Council; Israel Science Foundation; European Commission","keywords":"Computer science; Etiology; Bootstrapping (finance); Natural language processing; Medical diagnosis; Artificial intelligence; Generative grammar; Reliability (semiconductor); Pipeline (software); Machine learning; Pathology; Medicine; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003801674,0.001399105,0.0006622532,0.00886129,0.001003395,0.003175167,0.00155195,0.001109525,0.002429821],"category_scores_gemma":[0.01585221,0.0007764975,0.003231174,0.004416529,0.001182299,0.004878361,0.003046347,0.002330283,0.001768101],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001164636,"about_ca_system_score_gemma":0.002576439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003979402,"about_ca_topic_score_gemma":0.008396247,"domain_scores_codex":[0.9967456,0.001088337,0.0004462355,0.0007116815,0.0008940952,0.0001139976],"domain_scores_gemma":[0.9850165,0.01135463,0.001645798,0.001007557,0.0008029197,0.0001726647],"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":[0.0004911348,0.000552193,0.08193129,0.003129865,0.001248091,0.006732504,0.005752748,0.05613432,0.02357616,0.1138801,0.02347923,0.6830924],"study_design_scores_gemma":[0.00009186152,0.000154824,0.01759283,0.0006503756,0.0008654342,0.003395038,0.001862353,0.5843114,0.01202467,0.3265805,0.05231299,0.0001577664],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04349556,0.00163077,0.9337175,0.002887439,0.00008990303,0.0004676751,0.008786474,0.005292324,0.003632321],"genre_scores_gemma":[0.3450043,0.001728437,0.6306818,0.0006029609,0.0001765672,0.0005306161,0.01940993,0.0005415439,0.001323804],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00886129,"threshold_uncertainty_score":0.02010542,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03139023252115055,"score_gpt":0.3188436165696493,"score_spread":0.2874533840484988,"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."}}