{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01140582,0.00199406,0.001063315,0.001967462,0.0005384791,0.001905085,0.001364712,0.001722466,0.002162409],"category_scores_gemma":[0.04097172,0.000472669,0.001316095,0.001070808,0.0006504491,0.004598521,0.001961477,0.002353007,0.001386415],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001464554,"about_ca_system_score_gemma":0.001384512,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00614253,"about_ca_topic_score_gemma":0.005759567,"domain_scores_codex":[0.9921777,0.004744806,0.0008008208,0.00129283,0.0006861781,0.0002977255],"domain_scores_gemma":[0.9617385,0.03201904,0.000824238,0.002253614,0.00257058,0.0005941521],"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":[0.006827866,0.003211085,0.0377671,0.002459615,0.00160218,0.0005553272,0.001368646,0.1221573,0.01738767,0.001434339,0.01177955,0.7934493],"study_design_scores_gemma":[0.0004662287,0.005913389,0.02587784,0.0001973785,0.0009756897,0.0005839621,0.001601672,0.9284832,0.02792023,0.002761636,0.005048248,0.0001704897],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9258974,0.005638543,0.05472486,0.001073024,0.0004272975,0.000392111,0.002165634,0.005232766,0.004448329],"genre_scores_gemma":[0.9383199,0.001157377,0.05159983,0.0002689389,0.0001628161,0.0001771789,0.00648378,0.0002852434,0.001544981],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01140582,"threshold_uncertainty_score":0.0603205,"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."}}