{"id":"W4410381965","doi":"10.2196/64986","title":"Enhancing Diagnostic Accuracy of Ophthalmological Conditions with Complex Prompts in GPT-4: A Comparative Analysis of Global and LMIC-Specific Pathologies (Preprint)","year":2024,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Computer science; Computational biology; Biology; World Wide Web","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":[0.007711602,0.0003697761,0.0005080146,0.001810701,0.0003332909,0.001205048,0.0005844045,0.0009259683,0.003338433],"category_scores_gemma":[0.05297553,0.0001805669,0.0008950369,0.001254046,0.0006609704,0.001223822,0.001918395,0.0007078282,0.0006862133],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001129995,"about_ca_system_score_gemma":0.0009015203,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001819113,"about_ca_topic_score_gemma":0.002276806,"domain_scores_codex":[0.9953323,0.002664489,0.0004575516,0.0004967353,0.0007315789,0.0003173148],"domain_scores_gemma":[0.9713264,0.01907303,0.004649959,0.0009160212,0.002879266,0.00115533],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.005076728,0.00062832,0.9056075,0.0009910705,0.0002434734,0.0009133833,0.002968405,0.001072102,0.001333968,0.0002256254,0.003346668,0.07759283],"study_design_scores_gemma":[0.0002563266,0.004618919,0.9767636,0.0005062117,0.0003225329,0.001783222,0.003898921,0.005019307,0.002101989,0.0004365031,0.004228041,0.0000643045],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9963472,0.0004196963,0.0004892864,0.000392561,0.00002749087,0.0001710735,0.0006029722,0.0000492545,0.001500444],"genre_scores_gemma":[0.9970357,0.0002205217,0.001847698,0.0001422653,0.00003282003,0.000104421,0.0004316215,0.000009340053,0.0001757051],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007711602,"threshold_uncertainty_score":0.04078335,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3706237983337047,"score_gpt":0.5714735000853204,"score_spread":0.2008497017516157,"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."}}