{"id":"W4322766280","doi":"10.1016/j.compbiomed.2023.106595","title":"Segmentation-guided domain adaptation and data harmonization of multi-device retinal optical coherence tomography using cycle-consistent generative adversarial networks","year":2023,"lang":"en","type":"article","venue":"Computers in Biology and Medicine","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":14,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia; Memorial University of Newfoundland; Simon Fraser University","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; University of Nottingham; Wake Forest School of Medicine; Compute Canada","keywords":"Optical coherence tomography; Computer science; Adaptation (eye); Domain adaptation; Generative adversarial network; Adversarial system; Segmentation; Artificial intelligence; Coherence (philosophical gambling strategy); Generative grammar; Harmonization; Domain (mathematical analysis); Computer vision; Pattern recognition (psychology); Image (mathematics); Optics; Mathematics; 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":[0.0005015138,0.0001000587,0.0003033202,0.0002171682,0.00006100589,0.000004520522,0.00006094946,0.00007082903,0.000003866271],"category_scores_gemma":[0.0001173782,0.00007895823,0.00001712135,0.0004215783,0.0004285773,0.00004903778,0.00008308323,0.0001125809,2.566783e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001400067,"about_ca_system_score_gemma":0.00002602705,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001126915,"about_ca_topic_score_gemma":0.00001216052,"domain_scores_codex":[0.9990604,0.0001222225,0.0003096056,0.0003031698,0.00007646357,0.0001281173],"domain_scores_gemma":[0.9993986,0.0001934522,0.0001093729,0.0001514886,0.00007999199,0.00006702374],"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":[0.0007233587,0.0001866444,0.8686069,0.0002952045,0.0005793619,0.0001726967,0.004867851,0.01019081,0.06357247,0.001220445,0.0007975531,0.04878672],"study_design_scores_gemma":[0.00266839,0.0002024701,0.07447186,0.0003655989,0.0001952941,0.00006336152,0.001144412,0.9204859,0.0001375702,0.0001654597,0.00001845184,0.00008129554],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6797491,0.0007974649,0.317874,0.001169282,0.0001873949,0.0001775603,0.000005064446,0.00001787841,0.00002225128],"genre_scores_gemma":[0.9201289,0.0004581322,0.07845993,0.0003000178,0.0001547689,0.000002973297,0.0004839896,0.000005421984,0.000005816308],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.910295,"threshold_uncertainty_score":0.3219823,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09671338304553749,"score_gpt":0.3810224773843242,"score_spread":0.2843090943387866,"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."}}