{"id":"W4391364829","doi":"10.2139/ssrn.4705376","title":"Protecting Privacy While Improving Choroid Layer Segmentation in OCT Images: A GAN-Based Image Synthesis Approach","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; Nova Scotia Health Authority","funders":"","keywords":"Segmentation; Layer (electronics); Image (mathematics); Image segmentation; Choroid; Computer science; Computer vision; Artificial intelligence; Materials science; Neuroscience; Psychology; Retina; Nanotechnology","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":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.004835674,0.0005476939,0.0008590802,0.001022971,0.0002165259,0.000435109,0.000370501,0.0002639202,0.00004059354],"category_scores_gemma":[0.0007952937,0.0004695686,0.0006495266,0.0005806554,0.00006956176,0.0001452431,0.0001630351,0.01170333,0.00003664268],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003499829,"about_ca_system_score_gemma":0.005383639,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006999976,"about_ca_topic_score_gemma":0.00005342916,"domain_scores_codex":[0.9949228,0.0003933815,0.0008929802,0.0008770254,0.0006924695,0.002221354],"domain_scores_gemma":[0.9984044,0.0001339061,0.00056493,0.0005158673,0.0002184162,0.0001625013],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00130664,0.001846264,0.01169564,0.01005882,0.004200706,0.0007856939,0.003170449,0.005146818,0.6250983,0.0002507193,0.0003117085,0.3361283],"study_design_scores_gemma":[0.01068621,0.002088878,0.003477923,0.02178907,0.01490892,0.009289693,0.03732306,0.3498178,0.4679521,0.07652369,0.0003133372,0.00582934],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8478684,0.01472589,0.121016,0.006854701,0.0003925409,0.002631795,0.00002736257,0.0004504398,0.006032838],"genre_scores_gemma":[0.9893513,0.00072023,0.00787588,0.00008504645,0.0006954629,0.0002396005,0.00004421312,0.0001412236,0.0008470059],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.344671,"threshold_uncertainty_score":0.9997756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01649891327570091,"score_gpt":0.2868919788594477,"score_spread":0.2703930655837468,"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."}}