Machine learning and computational fluid dynamics derived FFRCT demonstrate comparable diagnostic performance in patients with coronary artery disease; A Systematic Review and Meta-Analysis

• Published in: Journal of cardiovascular computed tomography Vol. 19; no. 2; pp. 232 - 246
• Main Authors: Narimani-Javid, Roozbeh, Moradi, Mehdi, Mahalleh, Mehrdad, Najafi-vosough, Roya, Arzhangzadeh, Alireza, Khalique, Omar, Mojibian, Hamid, Kuno, Toshiki, Mohsen, Amr, Alam, Mahboob, Shafiei, Sasan, Khansari, Nakisa, Shaghaghi, Zahra, Nozhat, Salma, Hosseini, Kaveh, Hosseini, Seyed Kianoosh
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.03.2025
• Subjects: Angiography; Computational fluid dynamics; Computed tomography; Computed Tomography Angiography; Coronary Angiography - methods; Coronary artery disease; Coronary Artery Disease - diagnostic imaging; Coronary Artery Disease - physiopathology; Coronary Stenosis - diagnostic imaging; Coronary Stenosis - physiopathology; Coronary Vessels - diagnostic imaging; Coronary Vessels - physiopathology; Fractional flow reserve; Fractional Flow Reserve, Myocardial; Humans; Hydrodynamics; Machine Learning; Models, Cardiovascular; Predictive Value of Tests; Radiographic Image Interpretation, Computer-Assisted; Reproducibility of Results; Severity of Illness Index; CFD; Computational fluid dynamics; Angiography; ICA; CAD; Fractional flow reserve; Coronary artery disease; CCTA; FFR; Computed tomography; Machine learning; PCI; FFRCT; ML
• ISSN: 1934-5925; 1876-861X; 1876-861X
• DOI: 10.1016/j.jcct.2025.02.004

As a new noninvasive diagnostic technique, computed tomography-derived fraction flow reserve (FFRCT) has been used to identify hemodynamically significant coronary artery stenosis. FFRCT can be calculated using computational fluid dynamics (CFD) or machine learning (ML) approaches. It was hypothesized that ML-based FFRCT (FFRCTML) has comparable diagnostic performance with CFD-based FFRCT (FFRCTCFD). We used invasive FFR as the reference test to evaluate the diagnostic performance of FFRCTML vs. FFRCTCFD. We searched PubMed, Cochrane Library, EMBASE, WOS, and Scopus for articles published until March 2024. We analyzed the synthesized sensitivity, specificity, and diagnostic odds ratio (DOR) of FFRCTML vs FFRCTCFD at both the patient and vessel levels. We generated summary receiver operating characteristic curves (SROC) and then calculated the area under the curve (AUC). This meta-analysis included 23 studies reporting FFRCTCFD diagnostic performance and 18 studies reporting FFRCTML diagnostic performance. In the FFRCTCFD group, 2501 patients and 3764 vessels or lesions were analyzed. In the FFRCTML group, 1323 patients and 4194 vessels or lesions were analyzed. Our results showed that at the per-patient level, FFRCTCFD and FFRCTML had comparable pooled specificity (Z ​= ​−0.59, P ​= ​0.55) and AUC (P ​= ​0.5). At the per-vessel level, FFRCTCFD and FFRCTML also showed comparable specificity (Z ​= ​0.94, P ​= ​0.34), DOR (Z ​= ​0.7, P ​= ​0.48), and AUC (P ​= ​0.74). However, the sensitivity of FFRCTML was significantly lower compared to FFRCTCFD at both patient (Z ​= ​−3.85, P ​= ​0.0001) and vessel (Z ​= ​−2.05, P ​= ​0.04) levels. The FFRCTML technique was comparable to standard CFD approaches in terms of AUC and specificity. However, it did not achieve the same level of sensitivity as FFRCTCFD. [Display omitted]