{"id":"W3015711121","doi":"10.1167/tvst.9.2.20","title":"Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography","year":2020,"lang":"en","type":"article","venue":"Translational Vision Science & Technology","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":131,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Simon Fraser University","funders":"","keywords":"Artificial intelligence; Computer science; Ensemble learning; Optical coherence tomography; Pattern recognition (psychology); Deep learning; Diabetic retinopathy; Majority rule; Classifier (UML); Machine learning; Medicine; Ophthalmology","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":[0.00135474,0.000679184,0.0006719041,0.00108233,0.0002405192,0.0005235034,0.0006386863,0.0005291725,0.0004853857],"category_scores_gemma":[0.002313201,0.0002217948,0.0005994122,0.0005658681,0.0001671479,0.0006768213,0.0006112771,0.0007520678,0.0001602688],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006346971,"about_ca_system_score_gemma":0.0004823301,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005381593,"about_ca_topic_score_gemma":0.007635769,"domain_scores_codex":[0.9996687,0.00007731723,0.00001578499,0.00009252266,0.00008651756,0.00005919023],"domain_scores_gemma":[0.9992452,0.0002732288,0.000113133,0.0000977025,0.0002188799,0.00005194635],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000570282,0.0003749507,0.04804695,0.00008587428,0.0004782099,0.0001713173,0.00009091201,0.3085331,0.01654331,0.0008108109,0.003173497,0.6211208],"study_design_scores_gemma":[0.000006813398,0.00008785494,0.004962442,0.00001285749,0.00005519282,0.00005254039,0.0000140685,0.9899621,0.003944762,0.0006181779,0.0002740535,0.000009108765],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6681547,0.002243057,0.3238827,0.0004730345,0.00009574962,0.00008065756,0.000562329,0.002083206,0.002424664],"genre_scores_gemma":[0.9610623,0.0002735619,0.03736141,0.00009569045,0.00003655985,0.00003175481,0.0004155799,0.00002827153,0.0006948098],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005381593,"threshold_uncertainty_score":0.01070052,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02208978926930519,"score_gpt":0.3151593437827645,"score_spread":0.2930695545134593,"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."}}