{"id":"W4406213437","doi":"10.1016/j.cageo.2025.105853","title":"Addressing class imbalance in micro-CT image segmentation: A modified U-Net model with pixel-level class weighting","year":2025,"lang":"en","type":"article","venue":"Computers & Geosciences","topic":"Medical Image Segmentation Techniques","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Class (philosophy); Weighting; Pixel; Segmentation; Artificial intelligence; Image (mathematics); Computer science; Net (polyhedron); Image segmentation; Pattern recognition (psychology); Computer vision; Mathematics; Medicine; Radiology","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"],"consensus_categories":[],"category_scores_codex":[0.0007871455,0.0002977485,0.0003444246,0.0004969964,0.0003531133,0.0008238571,0.001989856,0.00005383185,0.000005131192],"category_scores_gemma":[0.00005332586,0.0002533698,0.00005972201,0.001785668,0.0005363803,0.001706502,0.0005482535,0.0003053223,0.000006996157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001859289,"about_ca_system_score_gemma":0.0004463102,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009539212,"about_ca_topic_score_gemma":0.00004556213,"domain_scores_codex":[0.9969499,0.000163657,0.0005614845,0.0009868401,0.000712837,0.0006253537],"domain_scores_gemma":[0.998648,0.0002370743,0.0002556218,0.0005646241,0.0001427843,0.0001518344],"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":[0.0001040575,0.0009958174,0.0116581,0.0005456605,0.0001048419,0.0004969938,0.008017703,0.03395863,0.3468151,0.02577566,0.02046943,0.5510581],"study_design_scores_gemma":[0.0007333125,0.00006619994,0.001862166,0.0004023067,0.000006572573,0.00002178874,0.0001710157,0.9577952,0.03622191,0.00230738,0.00008275687,0.0003293758],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03522173,0.0001072241,0.9596938,0.002038723,0.0004010152,0.0004057016,0.000005952515,0.0003421052,0.001783716],"genre_scores_gemma":[0.2693322,0.00002088889,0.7276942,0.002503689,0.00002821632,0.00005872836,0.000006808877,0.000008458119,0.0003467764],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9238366,"threshold_uncertainty_score":0.9999918,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04641570247423604,"score_gpt":0.3167057665127214,"score_spread":0.2702900640384853,"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."}}