{"id":"W2917393555","doi":"10.1016/j.media.2019.02.011","title":"Deep-learning based multiclass retinal fluid segmentation and detection in optical coherence tomography images using a fully convolutional neural network","year":2019,"lang":"en","type":"article","venue":"Medical Image Analysis","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":160,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia; Simon Fraser University","funders":"Fondation pour la Recherche sur Alzheimer; Genome British Columbia; Simon Fraser University; Michael Smith Health Research BC; Canadian Institutes of Health Research; Alzheimer Society; Natural Sciences and Engineering Research Council of Canada; Fondation Brain Canada","keywords":"Optical coherence tomography; Artificial intelligence; Computer science; Convolutional neural network; Segmentation; Retinal; Pattern recognition (psychology); Computer vision; Cut; Retina; Pixel; Image segmentation; Ophthalmology; Medicine; Optics; Physics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009642023,0.0002160341,0.0006000248,0.0006131512,0.0001255082,0.00008451418,0.0000891763,0.0001296647,0.0008414329],"category_scores_gemma":[0.0005528582,0.000191783,0.0003318312,0.002192515,0.0003004459,0.0001864798,0.00005094709,0.0005624504,0.00001854552],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009180622,"about_ca_system_score_gemma":0.00006782568,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003047552,"about_ca_topic_score_gemma":0.00005928088,"domain_scores_codex":[0.9973518,0.0003000829,0.0005156581,0.000529384,0.0008828038,0.000420236],"domain_scores_gemma":[0.9988341,0.0002947666,0.0001415257,0.0001989844,0.0002060814,0.0003245745],"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":[0.0003066237,0.0001547691,0.9010406,0.00009854491,0.0006218084,0.0002557713,0.00007793517,0.02009266,0.06319029,0.000002973347,0.00001692424,0.01414106],"study_design_scores_gemma":[0.001162061,0.0001225484,0.1859899,0.00009548943,0.001639282,0.00004377592,0.0002051358,0.8090285,0.001537837,0.00000815834,0.0000111934,0.000156055],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8229982,0.0002966448,0.1757198,0.0006279142,0.00003784549,0.0001512361,0.000001109593,0.00004976581,0.0001174964],"genre_scores_gemma":[0.9840202,0.00002770639,0.01529902,0.0003475684,0.0001231237,0.00001195578,0.00007435455,0.00001670442,0.00007940182],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7889358,"threshold_uncertainty_score":0.92131,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007756989084423179,"score_gpt":0.2748887174620818,"score_spread":0.2671317283776586,"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."}}