{"id":"W4417438495","doi":"10.1109/tai.2025.3644335","title":"Fully Perturbed Self-Ensemble Framework Using Cascaded Parallel CNN-Transformer for Semisupervised Medical Image Segmentation","year":2025,"lang":"","type":"article","venue":"IEEE Transactions on Artificial Intelligence","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Natural Science Foundation of China","keywords":"Segmentation; Exploit; Transformer; Image segmentation; Deep learning; Labeled data; Medical imaging; Pattern recognition (psychology)","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","sts"],"consensus_categories":[],"category_scores_codex":[0.0007078703,0.0008443571,0.0007465053,0.000542193,0.002057405,0.000561879,0.001702883,0.0008280391,0.000463995],"category_scores_gemma":[0.0001092559,0.0009334008,0.0006395668,0.002947497,0.0004744439,0.001126598,0.00001284042,0.001496121,0.0002299813],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005844005,"about_ca_system_score_gemma":0.001015761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007294757,"about_ca_topic_score_gemma":0.0001500249,"domain_scores_codex":[0.9935185,0.0002984547,0.001912174,0.001847462,0.001107351,0.001316069],"domain_scores_gemma":[0.9951291,0.002044443,0.0003157554,0.001296909,0.0006632851,0.0005504931],"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":[0.000766204,0.002116143,0.0000014647,0.0003023127,0.0003571458,0.00002394479,0.003523853,0.2329149,0.06702163,0.02870992,0.0001963516,0.6640661],"study_design_scores_gemma":[0.0001871178,0.0002010567,0.000001437082,0.0003719351,0.0002093112,0.00001907424,0.0003737327,0.6416487,0.3235993,0.03245132,0.000364814,0.0005722729],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003438803,0.0002469981,0.9819435,0.006918294,0.003483459,0.003204347,0.00007122663,0.0004980509,0.0001952696],"genre_scores_gemma":[0.6033237,0.0008586749,0.3922352,0.002098938,0.0002954798,0.0007498533,0.00000952201,0.00007653226,0.000352019],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6634939,"threshold_uncertainty_score":0.9993117,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0519471167204318,"score_gpt":0.3542854030578109,"score_spread":0.3023382863373791,"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."}}