{"id":"W3090937968","doi":"10.1007/978-3-030-69541-5_41","title":"Utilizing Transfer Learning and a Customized Loss Function for Optic Disc Segmentation from Retinal Images","year":2021,"lang":"en","type":"preprint","venue":"Lecture notes in computer science","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Calgary","funders":"","keywords":"Optic disc; Artificial intelligence; Segmentation; Computer science; Retinal; Optic cup (embryology); Computer vision; Glaucoma; Fundus (uterus); Robustness (evolution); Sørensen–Dice coefficient; Image segmentation; Pattern recognition (psychology); Ophthalmology; Medicine","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":[],"consensus_categories":[],"category_scores_codex":[0.0005763312,0.0002208836,0.0004492285,0.0002539928,0.0001730407,0.0003684126,0.00013347,0.0001074421,0.00001250878],"category_scores_gemma":[0.0002382429,0.0001875488,0.000136027,0.0004000741,0.0002672273,0.0001338338,0.0001779733,0.0006324479,5.992264e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001018409,"about_ca_system_score_gemma":0.000173945,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001224942,"about_ca_topic_score_gemma":0.00001406239,"domain_scores_codex":[0.9981799,0.00008866587,0.0002562735,0.0008642965,0.0003477519,0.0002631858],"domain_scores_gemma":[0.9990976,0.000342997,0.0000747502,0.0002174311,0.0001824921,0.00008469444],"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":[0.000672002,0.000124819,0.06581684,0.0007018849,0.0001686295,0.0001347422,0.004890023,0.0921916,0.1465069,0.000005566154,0.000004573811,0.6887824],"study_design_scores_gemma":[0.001778272,0.0001790438,0.01095056,0.001538345,0.0004407655,0.00005438197,0.00006368674,0.9465142,0.03721203,0.0009233924,0.00001338588,0.0003319201],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4439501,0.000439727,0.5543514,0.0008450834,0.0002049462,0.0001756588,0.000001900075,0.00002794421,0.000003256069],"genre_scores_gemma":[0.7900945,0.00006256538,0.2093437,0.0001651617,0.0002204076,0.00002097295,0.00007696248,0.00001212825,0.000003601134],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8543226,"threshold_uncertainty_score":0.7648019,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01564782765400996,"score_gpt":0.2982146050625069,"score_spread":0.2825667774084969,"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."}}