{"id":"W4413982530","doi":"10.1001/jamaophthalmol.2025.2918","title":"DeepSeek-R1 vs OpenAI o1 for Ophthalmic Diagnoses and Management Plans","year":2025,"lang":"en","type":"article","venue":"JAMA Ophthalmology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; St. Michael's Hospital; Hôpital Maisonneuve-Rosemont; Centre Hospitalier de l’Université de Montréal; McGill University; University of Waterloo; University of Toronto","funders":"","keywords":"McNemar's test; Medicine; Medical diagnosis; Clinical Practice; Medical physics; Family medicine; Pathology; Statistics","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":[0.0002680825,0.000143976,0.0002957619,0.0002073223,0.0001597104,0.00003228036,0.0001192264,0.0001982407,0.0001653921],"category_scores_gemma":[0.0002864729,0.0001341406,0.00005874454,0.0001853492,0.0001069786,0.00007943255,0.0000748556,0.0001502728,0.00004692742],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000765092,"about_ca_system_score_gemma":0.0001012653,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003358699,"about_ca_topic_score_gemma":0.00001477345,"domain_scores_codex":[0.9987925,0.00004955269,0.0003556416,0.0003768082,0.0000823943,0.0003431238],"domain_scores_gemma":[0.9988967,0.0005269999,0.00006187206,0.0002882766,0.0001173927,0.0001087601],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.002841136,0.0009975183,0.678472,0.003566688,0.000441058,0.001564028,0.001531245,0.00002225542,0.0004769136,0.03928816,0.05021318,0.2205858],"study_design_scores_gemma":[0.001566643,0.003747798,0.8006967,0.001144565,0.0006054015,0.007559805,0.006243668,0.001014784,0.0136898,0.05701307,0.1060034,0.0007143717],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9543092,0.000720803,0.0003895753,0.01926805,0.001084566,0.001367199,0.00001572477,0.00003701136,0.02280787],"genre_scores_gemma":[0.9898689,0.0003891431,0.001754393,0.001910144,0.0002331488,0.0005105847,0.00005796704,0.00001606067,0.005259608],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2198715,"threshold_uncertainty_score":0.5470096,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1100126829750908,"score_gpt":0.4478828120028748,"score_spread":0.337870129027784,"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."}}