{"id":"W4414951670","doi":"10.1038/s41598-025-19121-4","title":"A hybrid approach for enhancing pseudo-labeling in medical images through pseudo-label refinement","year":2025,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Medical Image Segmentation Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Segmentation; Robustness (evolution); Deep learning; Image segmentation; Pruning; Medical imaging; Pattern recognition (psychology)","routes":{"ca_aff":true,"ca_fund":true,"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":[0.006009179,0.0002374467,0.0003684403,0.0004275715,0.000349611,0.0006202661,0.0009805067,0.000102956,0.00006460446],"category_scores_gemma":[0.001740718,0.0002171484,0.00009997918,0.001201596,0.0002976084,0.0007677472,0.0005814599,0.0002758619,0.000005803082],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001808962,"about_ca_system_score_gemma":0.000736892,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006225998,"about_ca_topic_score_gemma":0.00001813244,"domain_scores_codex":[0.9950196,0.0001130666,0.001291172,0.001552342,0.0013925,0.000631355],"domain_scores_gemma":[0.9976294,0.000184454,0.0003380335,0.001378747,0.0002920316,0.0001773099],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00002651497,0.001951997,0.0007626749,0.0009506588,0.00009987124,0.001720416,0.002129768,0.00009486513,0.318403,0.01058519,0.3379897,0.3252853],"study_design_scores_gemma":[0.001003251,0.00006444196,0.000042499,0.000493019,0.00002098316,0.0002429555,0.0001489228,0.06529324,0.8760933,0.05173551,0.004397206,0.0004647269],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004025298,0.000230747,0.9863125,0.001226509,0.00341779,0.0009334307,0.000001637133,0.0003763739,0.003475699],"genre_scores_gemma":[0.04150933,0.00002232309,0.9531946,0.001219391,0.00006400383,0.0005016471,0.00005288356,0.00001622345,0.003419612],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5576903,"threshold_uncertainty_score":0.8855053,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02196825798798245,"score_gpt":0.3229647532610594,"score_spread":0.3009964952730769,"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."}}