{"id":"W4407558201","doi":"10.1017/ice.2025.20","title":"Chatting new territory: large language models for infection surveillance from pilot to deployment","year":2025,"lang":"en","type":"article","venue":"Infection Control and Hospital Epidemiology","topic":"Bacterial Identification and Susceptibility Testing","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; University of Toronto","funders":"","keywords":"Software deployment; Identification (biology); Medicine; Health care; Medical emergency; Computer science; Computer security; Political science; Biology; Software engineering","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":[0.0007419652,0.0001392914,0.0002841151,0.00005881791,0.0001454115,0.00002587148,0.00005163952,0.0001384141,0.000007646608],"category_scores_gemma":[0.00171507,0.0001396479,0.00007988058,0.00005486432,0.00002547101,0.00001158294,0.0000534461,0.00007280975,0.000003905697],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002772491,"about_ca_system_score_gemma":0.0000419018,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004710177,"about_ca_topic_score_gemma":0.003266314,"domain_scores_codex":[0.9987343,0.0001965485,0.0003448587,0.0004498819,0.00002729861,0.0002471293],"domain_scores_gemma":[0.9992608,0.0002419835,0.0001172324,0.0002029175,0.00008084224,0.00009621363],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003857239,0.0002020345,0.7700495,0.00004674613,0.000194945,3.013167e-7,0.000149988,0.001039334,0.1827665,0.002633408,0.008713695,0.03381785],"study_design_scores_gemma":[0.003476131,0.001235281,0.924633,0.00004065716,0.00004811586,0.000001786894,0.00003277718,0.01334994,0.00202619,0.004927454,0.04984096,0.0003877094],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.67694,0.0003800192,0.3189073,0.002287382,0.0008702605,0.0004280825,0.00005006339,0.00004737852,0.00008959023],"genre_scores_gemma":[0.9957606,0.00007351042,0.0007185263,0.002210919,0.0006188686,0.0001125104,0.0001579074,0.00001174398,0.0003353705],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3188207,"threshold_uncertainty_score":0.712041,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02079052594779464,"score_gpt":0.3071956904737791,"score_spread":0.2864051645259845,"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."}}