{"id":"W4401506867","doi":"10.1007/s00464-024-11155-5","title":"Clinical artificial intelligence: teaching a large language model to generate recommendations that align with guidelines for the surgical management of GERD","year":2024,"lang":"en","type":"article","venue":"Surgical Endoscopy","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":14,"is_retracted":false,"has_abstract":false,"ca_institutions":"McMaster University","funders":"","keywords":"GERD; Personalization; Medicine; Disease; Computer science; Reflux; Pathology; World Wide Web","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":[0.002373709,0.0001859605,0.0003563396,0.0001151127,0.0002214372,0.00006217054,0.0001673889,0.0001153019,0.00021862],"category_scores_gemma":[0.0001676164,0.0001117387,0.0002177245,0.0003000921,0.0000837273,0.0000783408,0.00006247299,0.0003593248,0.00005273125],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005871123,"about_ca_system_score_gemma":0.000189435,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001090885,"about_ca_topic_score_gemma":0.00003737706,"domain_scores_codex":[0.9977448,0.0001177654,0.000954389,0.0004468946,0.0003382003,0.0003979801],"domain_scores_gemma":[0.9979259,0.001142582,0.00008383367,0.0003786388,0.0002548347,0.0002141595],"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":[0.002848782,0.0007872527,0.0007628892,0.0006298124,0.0004986691,0.0003422711,0.004062851,0.003773422,0.00006622727,0.1638013,0.01885061,0.8035759],"study_design_scores_gemma":[0.0007247825,0.001628535,0.00004275652,0.002190124,0.0008326419,0.0002228317,0.02015622,0.6865331,0.01680255,0.008801154,0.2614218,0.0006435557],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.49043,0.00239038,0.4190884,0.0801371,0.002250607,0.002904234,0.0001574116,0.0002492849,0.002392584],"genre_scores_gemma":[0.9539925,0.0009256804,0.04082521,0.0009077668,0.001253305,0.0003278467,0.0001049894,0.00004217008,0.001620561],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8029323,"threshold_uncertainty_score":0.4556573,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4052754540355933,"score_gpt":0.5630672812007421,"score_spread":0.1577918271651488,"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."}}