{"id":"W4409351696","doi":"10.1016/j.compmedimag.2025.102550","title":"CASCADE-FSL: Few-shot learning for collateral evaluation in ischemic stroke","year":2025,"lang":"en","type":"article","venue":"Computerized Medical Imaging and Graphics","topic":"Acute Ischemic Stroke Management","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Cascade; Collateral; Collateral damage; Shot (pellet); Stroke (engine); Ischemic stroke; Collateral circulation; Medicine; Computer science; Artificial intelligence; Cardiology; Ischemia; Psychology; Physics; Business; Engineering; Chemistry","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001196389,0.001163177,0.001086092,0.001178008,0.0004934972,0.0005508224,0.001178492,0.001134111,0.002229706],"category_scores_gemma":[0.002885819,0.000381928,0.0007699061,0.0005003497,0.0004940703,0.00112114,0.001077214,0.001040013,0.0006688875],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007150702,"about_ca_system_score_gemma":0.001096442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008477777,"about_ca_topic_score_gemma":0.01125385,"domain_scores_codex":[0.9994423,0.00009760144,0.0000370394,0.0001822316,0.0001522239,0.00008851645],"domain_scores_gemma":[0.9993698,0.0002468435,0.00006347172,0.00008170585,0.0001651717,0.00007302561],"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":[0.0007544448,0.0004559612,0.01042157,0.0001786588,0.000154556,0.0004146861,0.0001525466,0.1236132,0.02415705,0.001669916,0.01048318,0.8275442],"study_design_scores_gemma":[0.00001889085,0.0001465568,0.001815468,0.00001532422,0.00002408123,0.0001157522,0.00001915298,0.9877393,0.007801523,0.001617914,0.000667948,0.00001810361],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2103032,0.00191509,0.7750487,0.0005533632,0.0002632787,0.000404512,0.001048538,0.006980958,0.003482338],"genre_scores_gemma":[0.8420978,0.0004548938,0.1491912,0.0003685473,0.0001145973,0.0001917593,0.00196388,0.0001642266,0.00545312],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008477777,"threshold_uncertainty_score":0.01685685,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01926121004152707,"score_gpt":0.3237557017531286,"score_spread":0.3044944917116015,"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."}}