{"id":"W4391644728","doi":"10.2196/53216","title":"Investigating the Impact of Prompt Engineering on the Performance of Large Language Models for Standardizing Obstetric Diagnosis Text: Comparative Study","year":2024,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":29,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Guangxi Key Research and Development Program","keywords":"Medical diagnosis; Computer science; Artificial intelligence; Standardization; Consistency (knowledge bases); Medical record; Natural language processing; Data science; Machine learning; Medicine; Pathology","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":[0.004294361,0.0001565127,0.0002680236,0.0003941918,0.000382352,0.0001619895,0.001021585,0.00003710169,0.000006634915],"category_scores_gemma":[0.0005408472,0.0000798403,0.0001185735,0.001943111,0.00009678089,0.0005818242,0.0004156453,0.0008974477,0.000004825886],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000221842,"about_ca_system_score_gemma":0.0002515497,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001434341,"about_ca_topic_score_gemma":0.000005903028,"domain_scores_codex":[0.9973384,0.0005830775,0.0003920232,0.0002448474,0.000954297,0.0004873369],"domain_scores_gemma":[0.9928294,0.006032853,0.0001205586,0.0005223523,0.0004290542,0.00006572266],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003888633,0.0004006585,0.01349104,0.001958176,0.000293578,0.000003657733,0.7758337,0.1540073,0.0004444034,0.04146068,0.001205405,0.01086246],"study_design_scores_gemma":[0.0001921682,0.001846533,0.01501199,0.0004939227,0.000002507803,0.000001237014,0.005818752,0.9739821,0.002338851,0.000201741,0.00002851335,0.00008167944],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9905699,0.0005186144,0.005965017,0.0003559747,0.00006637332,0.002169867,0.00006062454,0.00005981313,0.000233793],"genre_scores_gemma":[0.998555,0.00001009915,0.0004767066,0.000006874738,0.00003204004,0.0008894043,0.000002397864,0.00001424055,0.00001320885],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8199748,"threshold_uncertainty_score":0.3899013,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1009751027643847,"score_gpt":0.4614439547127733,"score_spread":0.3604688519483886,"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."}}