{"id":"W4386246031","doi":"10.59275/j.melba.2023-553a","title":"Weakly Supervised Intracranial Hemorrhage Segmentation using Head-Wise Gradient-Infused Self-Attention Maps from a Swin Transformer in Categorical Learning","year":2023,"lang":"en","type":"article","venue":"The Journal of Machine Learning for Biomedical Imaging","topic":"Intracerebral and Subarachnoid Hemorrhage Research","field":"Medicine","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Fonds de recherche du Québec – Nature et technologies","keywords":"Segmentation; Artificial intelligence; Computer science; Deep learning; Categorical variable; Pattern recognition (psychology); Binary classification; Code (set theory); Machine learning","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":[0.003365948,0.0002671612,0.0005696935,0.0007572509,0.0004050339,0.00007264091,0.0002534284,0.0001292616,0.0001396531],"category_scores_gemma":[0.0006370316,0.000183618,0.0002929658,0.001020336,0.0001456428,0.0003134005,0.00004895523,0.002076301,0.00002865811],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003449901,"about_ca_system_score_gemma":0.0002112654,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006130896,"about_ca_topic_score_gemma":0.00001795802,"domain_scores_codex":[0.9965562,0.0005197962,0.000963989,0.0002547004,0.0009845458,0.0007207347],"domain_scores_gemma":[0.998487,0.0005811292,0.0002877454,0.0001316262,0.0002021372,0.0003103947],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003395756,0.0006438046,0.08184686,0.0005683351,0.0002229474,0.001898861,0.005410959,0.001550071,0.5336555,0.00001843849,0.000140133,0.3706483],"study_design_scores_gemma":[0.006572607,0.0005501282,0.008714723,0.0001995495,0.0003001126,0.001131903,0.002043114,0.9789099,0.0008720905,0.0003056114,0.0001986935,0.0002016102],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9519249,0.0003791604,0.03917371,0.007580855,0.000306521,0.000488962,0.000009184798,0.0001076167,0.00002914929],"genre_scores_gemma":[0.9869846,0.0001494775,0.01159792,0.0001467999,0.0006719668,0.000009236535,0.0002486079,0.00006457198,0.0001268264],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9773598,"threshold_uncertainty_score":0.9020609,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0198977324395852,"score_gpt":0.3105897827932063,"score_spread":0.2906920503536211,"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."}}