Journal of the Society for Cardiovascular Angiography & Interventions · 2025
Diagnostic Performance of a Novel AI-Guided Coronary Computed Tomography Algorithm for Predicting Myocardial Ischemia (AI-QCTISCHEMIA) Across Sex and Age Subgroups.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Kamila Putri Annisa, Hojjati Tara, Nurmohamed Nick S, Danad Ibrahim, Ding Yipu, Jukema Ruurt A, Raijmakers Pieter G, Driessen Roel S, Bom Michiel J, van Diemen Pepijn, Pontone Gianluca, Andreini Daniele, Chang Hyuk-Jae, Katz Richard J, Choi Andrew D, Knaapen Paul, Bax Jeroen J, van Rosendael Alexander, CREDENCE and PACIFIC-1 Investigators
Аннотация
BACKGROUND: AI-QCTISCHEMIA is a novel artificial intelligence algorithm that predicts myocardial ischemia using quantitative features from coronary computed tomography angiography, providing a noninvasive alternative to functional imaging. However, its diagnostic performance across key demographic subgroups, particularly by sex and age, remains underexplored. We aimed to evaluate the diagnostic performance of AI-QCTISCHEMIA for predicting myocardial ischemia across these subgroups.
METHODS: This post-hoc analysis included symptomatic patients with suspected coronary artery disease from the CREDENCE (Computed Tomographic Evaluation of Atherosclerotic Determinants of Myocardial Ischemia) (n = 305; 868 vessels) and PACIFIC-1 (Comparison of Coronary Computed Tomography Angiography, Single Photon Emission Computed Tomography [SPECT], Positron Emission Tomography [PET], and Hybrid Imaging for Diagnosis of Ischemic Heart Disease Determined by Fractional Flow Reserve) (n = 208; 612 vessels) studies. All patients underwent coronary computed tomography angiography, myocardial perfusion imaging (SPECT and/or PET), and invasive coronary angiography with 3-vessel fractional flow reserve as the reference standard. Diagnostic performance was evaluated at the vessel level using receiver operating characteristic analysis and under the curve (AUC), stratified by sex and age groups.
RESULTS: In computed tomographic evaluation of atherosclerotic determinants of myocardial ischemia, AI-QCTISCHEMIA demonstrated higher diagnostic performance than myocardial perfusion imaging, with AUCs of 0.87 vs 0.63 in men and 0.85 vs 0.71 in women (P < .001 for both). Similarly, in older (≥65 years) and younger (<65 years) patients, AUCs were 0.85 vs 0.67 and 0.87 vs 0.63 (P < .001 for both). In PACIFIC-1, AI-QCTISCHEMIA outperformed SPECT in men (AUC = 0.86 vs 0.67; P < .001) and women (0.81 vs 0.65; P < .001) while performing comparably with PET (0.86 vs 0.82; P = .140; 0.81 vs 0.72; P = .214). In older patients, AI-QCTISCHEMIA showed higher performance than SPECT (0.85 vs 0.73; P < .001) and was similar to PET (0.85 vs 0.86; P = .816). In younger patients, it also outperformed SPECT (0.87 vs 0.66; P < .001) with comparable performance with PET (0.87 vs 0.84; P = .338).
CONCLUSIONS: AI-QCTISCHEMIA demonstrated consistently high diagnostic performance to detect myocardial ischemia across sex and age groups, significantly outperforming SPECT and showing comparable performance with PET, supporting its role as a noninvasive alternative for ischemia assessment.
METHODS: This post-hoc analysis included symptomatic patients with suspected coronary artery disease from the CREDENCE (Computed Tomographic Evaluation of Atherosclerotic Determinants of Myocardial Ischemia) (n = 305; 868 vessels) and PACIFIC-1 (Comparison of Coronary Computed Tomography Angiography, Single Photon Emission Computed Tomography [SPECT], Positron Emission Tomography [PET], and Hybrid Imaging for Diagnosis of Ischemic Heart Disease Determined by Fractional Flow Reserve) (n = 208; 612 vessels) studies. All patients underwent coronary computed tomography angiography, myocardial perfusion imaging (SPECT and/or PET), and invasive coronary angiography with 3-vessel fractional flow reserve as the reference standard. Diagnostic performance was evaluated at the vessel level using receiver operating characteristic analysis and under the curve (AUC), stratified by sex and age groups.
RESULTS: In computed tomographic evaluation of atherosclerotic determinants of myocardial ischemia, AI-QCTISCHEMIA demonstrated higher diagnostic performance than myocardial perfusion imaging, with AUCs of 0.87 vs 0.63 in men and 0.85 vs 0.71 in women (P < .001 for both). Similarly, in older (≥65 years) and younger (<65 years) patients, AUCs were 0.85 vs 0.67 and 0.87 vs 0.63 (P < .001 for both). In PACIFIC-1, AI-QCTISCHEMIA outperformed SPECT in men (AUC = 0.86 vs 0.67; P < .001) and women (0.81 vs 0.65; P < .001) while performing comparably with PET (0.86 vs 0.82; P = .140; 0.81 vs 0.72; P = .214). In older patients, AI-QCTISCHEMIA showed higher performance than SPECT (0.85 vs 0.73; P < .001) and was similar to PET (0.85 vs 0.86; P = .816). In younger patients, it also outperformed SPECT (0.87 vs 0.66; P < .001) with comparable performance with PET (0.87 vs 0.84; P = .338).
CONCLUSIONS: AI-QCTISCHEMIA demonstrated consistently high diagnostic performance to detect myocardial ischemia across sex and age groups, significantly outperforming SPECT and showing comparable performance with PET, supporting its role as a noninvasive alternative for ischemia assessment.