{"id":"W4412392903","doi":"10.1016/j.compbiomed.2025.110613","title":"Tiny-objective segmentation for spot signs on multi-phase CT angiography via contrastive learning with dynamic-updated positive–negative memory banks","year":2025,"lang":"en","type":"article","venue":"Computers in Biology and Medicine","topic":"Intracerebral and Subarachnoid Hemorrhage Research","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Manitoba; University of Calgary","funders":"","keywords":"Segmentation; Computer science; Artificial intelligence; Sign (mathematics); Sensitivity (control systems); Pattern recognition (psychology); Hematoma; Radiology; Medicine; Mathematics","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":[0.0007182416,0.0007098669,0.0004940326,0.0008416124,0.0003554618,0.0007920427,0.0009741051,0.0008519791,0.002914179],"category_scores_gemma":[0.001544056,0.0004269703,0.0003809796,0.0004450059,0.0003178935,0.0009824341,0.001227296,0.0006147523,0.0006023719],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003202314,"about_ca_system_score_gemma":0.0008044476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001596787,"about_ca_topic_score_gemma":0.003606559,"domain_scores_codex":[0.9998298,0.00002110454,0.00001822646,0.00004586091,0.00005701857,0.0000280018],"domain_scores_gemma":[0.9995982,0.0001246549,0.00004458923,0.00005819238,0.0001367364,0.00003768547],"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":[0.000710358,0.0001804606,0.004266969,0.0003060954,0.0001242038,0.0003037594,0.0001493121,0.03755808,0.09519697,0.004058671,0.00550393,0.8516411],"study_design_scores_gemma":[0.00003332418,0.0001746958,0.002422074,0.00002889699,0.00009090506,0.0005262975,0.0000574887,0.9285303,0.06136359,0.003665945,0.00306818,0.00003845339],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06155025,0.0005309906,0.9335594,0.0002041287,0.00008726043,0.0001175274,0.0001712767,0.002017793,0.001761325],"genre_scores_gemma":[0.3957874,0.0004757886,0.5981944,0.0003180026,0.00008649955,0.0001880853,0.0004046668,0.0004425462,0.004102565],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002914179,"threshold_uncertainty_score":0.009748936,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01177512920670647,"score_gpt":0.3534600484211777,"score_spread":0.3416849192144712,"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."}}