{"id":"W3099733385","doi":"10.1016/j.artmed.2019.101758","title":"Ophthalmic diagnosis using deep learning with fundus images – A critical review","year":2019,"lang":"en","type":"review","venue":"Artificial Intelligence in Medicine","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":200,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Deep learning; Fundus (uterus); Artificial intelligence; Computer science; Diabetic retinopathy; Glaucoma; Macular degeneration; Segmentation; Optic disc; Ophthalmology; Optic cup (embryology); Optometry; Medicine","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001416948,0.0009899244,0.001849073,0.002961779,0.0002537279,0.001406296,0.001115688,0.001601316,0.002726036],"category_scores_gemma":[0.00410839,0.0004589659,0.001256786,0.001748105,0.000704265,0.002127912,0.0009038821,0.001691325,0.001091554],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007095433,"about_ca_system_score_gemma":0.001952645,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001564115,"about_ca_topic_score_gemma":0.003551724,"domain_scores_codex":[0.9995669,0.00009311442,0.000111975,0.00009237482,0.0001113203,0.00002438191],"domain_scores_gemma":[0.9975318,0.00163784,0.0002797953,0.00005378478,0.0004248452,0.00007183484],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001118354,0.00007357091,0.0006859958,0.03719487,0.0005437435,0.0003456287,0.0000454172,0.0003732228,0.0006466262,0.001434928,0.03415655,0.9243875],"study_design_scores_gemma":[0.0001515396,0.0004713745,0.005502007,0.06953882,0.004867546,0.007150355,0.0002435755,0.001487285,0.001791042,0.00670814,0.9018822,0.0002060809],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00007335349,0.9990748,0.0001541103,0.0003118705,0.0001348705,0.000004518314,0.00001331994,0.000005167482,0.0002279297],"genre_scores_gemma":[0.001229165,0.9969373,0.0004384842,0.0006929126,0.0005134181,0.00000894418,0.0000339249,0.000002706322,0.0001432214],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.002961779,"threshold_uncertainty_score":0.009119511,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2746009303429246,"score_gpt":0.4895945420956003,"score_spread":0.2149936117526757,"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."}}