{"id":"W4412049030","doi":"10.1016/j.xops.2025.100877","title":"Deep Learning with Disc Photos or OCT Scans in Glaucoma Detection","year":2025,"lang":"en","type":"article","venue":"Ophthalmology Science","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Eye Institute; Massachusetts Lions Eye Research Fund; National Institutes of Health; Glaucoma Foundation","keywords":"Optical coherence tomography; Glaucoma; Optometry; Tomography; Coherence (philosophical gambling strategy); Optic disc; Artificial intelligence; Ophthalmology; Optics; Computer science; Computer vision; Medicine; Physics","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":[0.0004200211,0.00008727542,0.0001898863,0.0004226271,0.0001939677,0.00002709791,0.0001422352,0.00003623815,0.00008175866],"category_scores_gemma":[0.0003097224,0.00005714787,0.00002904191,0.00208515,0.0006119451,0.0001119767,0.00004184783,0.0002441877,0.00001439449],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008493072,"about_ca_system_score_gemma":0.0001787313,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003345377,"about_ca_topic_score_gemma":0.00005738994,"domain_scores_codex":[0.9990191,0.00004797927,0.0001312782,0.0003499734,0.0001751659,0.0002765221],"domain_scores_gemma":[0.9995829,0.00006174598,0.00004324311,0.0001803513,0.00006909943,0.00006270186],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001961544,0.00005277193,0.959961,0.0000178749,0.00001103899,0.0006704797,0.0001202558,0.000113638,0.0335612,0.0000240836,0.000001705156,0.005269831],"study_design_scores_gemma":[0.0006481886,0.00054823,0.9426234,0.0001563441,0.00005941376,0.003783935,0.0006960529,0.03526365,0.01594462,0.0000809379,0.00008354292,0.0001116575],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9825543,0.00004941582,0.001516891,0.0005016319,0.0000628967,0.000091364,8.922403e-8,0.0000246685,0.01519871],"genre_scores_gemma":[0.9943129,0.000006406632,0.0007256902,0.00008303403,0.00001221372,0.00001292072,7.300408e-7,0.000004142034,0.004841937],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03515001,"threshold_uncertainty_score":0.2330423,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01075375575070845,"score_gpt":0.3153943311391375,"score_spread":0.304640575388429,"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."}}