{"id":"W3198627688","doi":"10.1016/j.compbiomed.2021.104849","title":"Improvement of automatic ischemic stroke lesion segmentation in CT perfusion maps using a learned deep neural network","year":2021,"lang":"en","type":"article","venue":"Computers in Biology and Medicine","topic":"Acute Ischemic Stroke Management","field":"Medicine","cited_by":54,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"University of Alberta","keywords":"Segmentation; Jaccard index; Artificial intelligence; Thresholding; Deep learning; Computer science; Stroke (engine); Pattern recognition (psychology); Similarity (geometry); Medicine; Sørensen–Dice coefficient; Feature (linguistics); Dice; Artificial neural network; Image segmentation; Image (mathematics); Mathematics; Statistics","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":[0.0005617397,0.0009897754,0.0007877594,0.0009740471,0.0003405024,0.0007914761,0.0009489016,0.001060381,0.001871278],"category_scores_gemma":[0.00141388,0.0004877027,0.0006973665,0.0005995987,0.0002405055,0.0006750106,0.00075887,0.0008892099,0.0005882807],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006803148,"about_ca_system_score_gemma":0.001503916,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01467108,"about_ca_topic_score_gemma":0.01563656,"domain_scores_codex":[0.9997554,0.00002784461,0.00001527978,0.0000834108,0.0000718806,0.0000462263],"domain_scores_gemma":[0.9996025,0.0001206201,0.0000367426,0.00004543909,0.0001675515,0.0000272617],"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":[0.0006470055,0.0003293137,0.003204438,0.0001698154,0.0001860345,0.0002041977,0.00006917343,0.2048586,0.04244046,0.001317305,0.006390356,0.7401833],"study_design_scores_gemma":[0.0000093378,0.00003525612,0.0006581596,0.000006628542,0.00002095472,0.00004012434,0.000005048696,0.992343,0.00621013,0.0003793059,0.000285148,0.000006889591],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1636381,0.001169295,0.8257912,0.0005388992,0.0002544622,0.0001439311,0.0006123198,0.005115313,0.002736514],"genre_scores_gemma":[0.7283429,0.0004821153,0.2644892,0.0002771881,0.00009924892,0.00009086043,0.001114107,0.0002914812,0.004812936],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01467108,"threshold_uncertainty_score":0.02917141,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02545256565107613,"score_gpt":0.3195878562236634,"score_spread":0.2941352905725873,"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."}}