{"id":"W4386826466","doi":"10.1016/j.media.2023.102938","title":"GAMMA challenge: Glaucoma grAding from Multi-Modality imAges","year":2023,"lang":"en","type":"article","venue":"Medical Image Analysis","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":102,"is_retracted":false,"has_abstract":false,"ca_institutions":"DiagnoCure (Canada); École de Technologie Supérieure","funders":"","keywords":"Glaucoma; Optical coherence tomography; Fundus photography; Grading (engineering); Medicine; Optometry; Fundus (uterus); Ophthalmology; Optic disc; Modality (human–computer interaction); Modalities; Artificial intelligence; Blindness; Computer science; Retinal; Fluorescein angiography","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001203433,0.0003221816,0.001157537,0.0009301575,0.0001897172,0.00009885024,0.0003554042,0.0001883186,0.003187837],"category_scores_gemma":[0.001912712,0.0002583994,0.001171641,0.003321148,0.0003461221,0.0001696768,0.0001846792,0.0005835181,0.001083751],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007196612,"about_ca_system_score_gemma":0.00009079603,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002783703,"about_ca_topic_score_gemma":0.0001535018,"domain_scores_codex":[0.9961387,0.0002295853,0.0006792493,0.0008193525,0.001506631,0.0006264611],"domain_scores_gemma":[0.9976083,0.0003544904,0.000151456,0.0008585713,0.0002227584,0.0008044177],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001548356,0.00148342,0.7921714,0.0002548855,0.01969479,0.01153457,0.001368696,0.00001856371,0.01706553,0.0000403518,0.02706745,0.1291456],"study_design_scores_gemma":[0.00244044,0.0000848342,0.5145689,0.0001979908,0.01566839,0.00002676595,0.0008727188,0.4581045,0.00300611,0.0004875109,0.003903877,0.0006379038],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8263194,0.001380266,0.07419523,0.09313556,0.0001775589,0.0002291154,0.0000938736,0.001207254,0.003261772],"genre_scores_gemma":[0.9888029,0.0009949997,0.003881134,0.000831745,0.0004895465,0.00002619802,0.0007483506,0.00004490411,0.004180193],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.458086,"threshold_uncertainty_score":0.9999868,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02778191829118971,"score_gpt":0.3494678826587598,"score_spread":0.3216859643675701,"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."}}