{"id":"W4412604178","doi":"10.1136/bjo-2025-327360","title":"Performance of DeepSeek-R1 in ophthalmology: an evaluation of clinical decision-making and cost-effectiveness","year":2025,"lang":"en","type":"article","venue":"British Journal of Ophthalmology","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":17,"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 Ottawa; University of Waterloo; University of Toronto","funders":"","keywords":"McNemar's test; Medicine; Subspecialty; Ophthalmology; Optometry; Machine learning; Computer science; Statistics; Family medicine; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02592577,0.001203459,0.001379394,0.001775702,0.0003198961,0.002180533,0.002055813,0.000874556,0.004464648],"category_scores_gemma":[0.05778411,0.000410013,0.002441106,0.001374751,0.0008827056,0.002145728,0.003247758,0.001136971,0.0009335246],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003235044,"about_ca_system_score_gemma":0.003737117,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0044373,"about_ca_topic_score_gemma":0.003879876,"domain_scores_codex":[0.9811194,0.01122898,0.001657664,0.002052742,0.003138227,0.0008030015],"domain_scores_gemma":[0.9198114,0.06055768,0.009634682,0.003068932,0.00380731,0.003120038],"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":[0.01485562,0.004624893,0.6321758,0.001717772,0.001370192,0.0001935072,0.001824144,0.04535481,0.000822502,0.001081614,0.004576355,0.2914028],"study_design_scores_gemma":[0.002571979,0.03202316,0.6059353,0.001100742,0.001531664,0.001159758,0.003217322,0.335802,0.005163212,0.003263994,0.007817902,0.0004129206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9877025,0.0008967243,0.004650782,0.0006818864,0.00003770804,0.0006326436,0.001375585,0.0003089246,0.003713301],"genre_scores_gemma":[0.9866751,0.0003647984,0.01062525,0.0001470621,0.00002420826,0.0003707311,0.00127007,0.00003742018,0.0004853389],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02592577,"threshold_uncertainty_score":0.1371102,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0785873835091986,"score_gpt":0.4703173014432211,"score_spread":0.3917299179340226,"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."}}