{"id":"W4415316174","doi":"10.1016/j.bspc.2025.108758","title":"Advanced artificial intelligence-based classification of corneal layers using residual self-attention and dilated convolutional networks: enhancing biomedical signal processing","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":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Residual; Convolutional neural network; Pattern recognition (psychology); Feature extraction; Keratoconus; Interpretability; Feature (linguistics); Visualization","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006726682,0.0005534202,0.0004213973,0.0005930651,0.000254969,0.0006863617,0.0006053688,0.0006035851,0.001260398],"category_scores_gemma":[0.00100792,0.0001952842,0.000582977,0.0004538452,0.0002970596,0.000640227,0.0005347767,0.0005973487,0.0003496567],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005290247,"about_ca_system_score_gemma":0.0007165188,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005541224,"about_ca_topic_score_gemma":0.007442456,"domain_scores_codex":[0.9998487,0.00002627524,0.00001016935,0.00003510394,0.00004138647,0.00003839416],"domain_scores_gemma":[0.999605,0.0001104251,0.00003523591,0.00004397089,0.0001785458,0.00002689543],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004932515,0.0002660833,0.003321993,0.0001581457,0.0001645813,0.0001690708,0.0001261251,0.154521,0.1396705,0.005171177,0.003618133,0.6923199],"study_design_scores_gemma":[0.000004452818,0.00003947159,0.001052146,0.000007041207,0.00002486355,0.00003042537,0.000008398218,0.9838625,0.01382956,0.0006895558,0.0004457826,0.000005897732],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1530678,0.001085706,0.8411459,0.0004127111,0.0001617803,0.00005771701,0.000133459,0.0009927064,0.002942176],"genre_scores_gemma":[0.7906384,0.0006033349,0.2014881,0.0001960214,0.00008821192,0.00003763425,0.0002796466,0.0001075406,0.006561057],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005541224,"threshold_uncertainty_score":0.01101798,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01739781934359907,"score_gpt":0.2918501864799043,"score_spread":0.2744523671363052,"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."}}