{"id":"W4408188883","doi":"10.1016/j.bspc.2025.107747","title":"Enhanced retinal arteries and veins segmentation through deep learning with conditional random fields","year":2025,"lang":"en","type":"article","venue":"Biomedical Signal Processing and Control","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McMaster University","funders":"Precursory Research for Embryonic Science and Technology; Japan Science and Technology Agency; Japan Science and Technology Corporation","keywords":"Segmentation; Retinal; Artificial intelligence; Computer science; Conditional random field; Deep learning; Pattern recognition (psychology); 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000152831,0.0001157187,0.0002716285,0.00006661301,0.0002519884,0.00007991515,0.00002438379,0.00006641896,0.00004457817],"category_scores_gemma":[0.00005265142,0.00007943875,0.00003233941,0.0001611195,0.0003732151,0.00009474337,0.000009084636,0.0001996034,0.000001047695],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001252298,"about_ca_system_score_gemma":0.00007172825,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001497556,"about_ca_topic_score_gemma":0.000001406907,"domain_scores_codex":[0.9991544,0.00004881018,0.0001861147,0.0002267214,0.0002279879,0.0001559709],"domain_scores_gemma":[0.999608,0.0001121695,0.00006833936,0.00003633874,0.00008636629,0.00008873346],"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":[0.005798236,0.0003623233,0.01164448,0.0009883718,0.0005090376,0.0001024463,0.001821428,0.00002897905,0.09678146,0.0003319449,0.0001727467,0.8814585],"study_design_scores_gemma":[0.2038942,0.01683346,0.0977619,0.0136771,0.01116133,0.001554166,0.02306801,0.5156315,0.05136818,0.02262309,0.0393778,0.003049287],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1553484,0.001621571,0.8372225,0.004900619,0.00001316753,0.0001114233,0.00000190731,0.00004972197,0.0007306389],"genre_scores_gemma":[0.997274,0.0000655437,0.0009108137,0.001092808,0.00009654063,0.00002402466,0.00004342954,0.000005562119,0.0004873171],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8784093,"threshold_uncertainty_score":0.3239418,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.004827450009120422,"score_gpt":0.2600727030189702,"score_spread":0.2552452530098497,"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."}}