{"id":"W4409229190","doi":"10.1016/j.bspc.2025.107824","title":"Attention-based hybrid deep learning models and its scientific validation for cardiovascular disease risk stratification","year":2025,"lang":"en","type":"article","venue":"Biomedical Signal Processing and Control","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"Queen's University; University of Toronto","funders":"","keywords":"Risk stratification; Stratification (seeds); Computer science; Artificial intelligence; Disease; Machine learning; Deep learning; Medicine; Cardiology; Internal medicine","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.002032691,0.0001419205,0.0002532997,0.000162545,0.003437627,0.0001403117,0.00009362206,0.0001313552,0.00001583875],"category_scores_gemma":[0.0006774107,0.0001249487,0.0001069339,0.0002434163,0.0002589548,0.000192786,0.00002104299,0.000357666,0.000006074348],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006782744,"about_ca_system_score_gemma":0.000720172,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009878688,"about_ca_topic_score_gemma":0.000008833063,"domain_scores_codex":[0.9976905,0.0005099864,0.0005259477,0.0005359635,0.0003682607,0.0003693496],"domain_scores_gemma":[0.9981819,0.0006628308,0.0001967166,0.0001401794,0.000537562,0.0002807759],"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":[0.0006507966,0.0001567572,0.01664359,0.003774156,0.0001986146,0.000003757996,0.0007130816,0.007127034,0.002899081,0.003110033,0.0001058501,0.9646173],"study_design_scores_gemma":[0.0007350569,0.00005022584,0.0007563112,0.0004558912,0.0002859704,1.652111e-7,0.0004545412,0.9831233,0.0001171585,0.01205863,0.001835446,0.0001272901],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.162054,0.009970902,0.8247076,0.001527713,0.0002392795,0.001309376,0.00005458311,0.0001067338,0.00002980881],"genre_scores_gemma":[0.9986006,0.00004163028,0.0002319876,0.0001536245,0.0001744484,0.000488347,0.0001241715,0.00001355858,0.0001716352],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9759963,"threshold_uncertainty_score":0.9978598,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07233119977492414,"score_gpt":0.3812535670607506,"score_spread":0.3089223672858265,"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."}}