Background/Introduction Healed coronary plaques, morphologically characterized by a layered pattern, are signatures of previous plaque disruption and healing. Recent optical coherence tomography (OCT) studies showed that layered plaque is associated with vascular vulnerability and rapid plaque progression. However, the diagnosis of layered plaque requires expertise in OCT image interpretation and is susceptible to interobserver variability. Purpose We aimed to develop a deep learning (DL) model for an accurate diagnosis of layered plaque. Methods We developed a Visual Transformer (ViT)-based DL model emulating the cardiologists who review consecutive OCT frames to make a diagnosis (Figure 1), and compared it to the standard convolutional neural network (CNN) model. We used 302,415 cross-sectional OCT images from 873 patients collected from 9 sites: 237,021 images from 581 patients for training and internal validation from 8 sites, and 65394 images from 292 patients collected from another site for external validation. Results Model performances were evaluated using the area under the receiver operating characteristics (AUC). In the five-fold cross validation, the ViT-based model showed better performance than the standard CNN-based model with AUC of 0.886 (95% confidence interval [CI], 0.882–0.891) compared with 0.797 (95% CI, 0.790–0.804). The ViT-based model also outperformed the standard CNN-based model in the external validation, with an AUC of 0.857 (95% CI, 0.849–0.864) compared to 0.806 (95% CI, 0.797–0.815) (Figure 2). Conclusion(s) The ViT-based DL model will help cardiologists to make an accurate diagnosis of layered plaque, which might help to stratify the risk of future adverse cardiac events. Funding Acknowledgement Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Mrs. Gillian Gray through the Allan Gray Fellowship Fund in Cardiology
Background Vascular inflammation has been recognized as one of the key factors in the pathogenesis of acute coronary syndromes (ACS). Peri-coronary adipose tissue (PCAT) attenuation by computed tomography angiography (CTA) has emerged as a marker specific for coronary artery inflammation. We examined the relationship between clinical presentation and coronary artery inflammation assessed by PCAT attenuation and coronary plaque characteristics. Methods Patients with ACS or stable angina pectoris (SAP) who underwent pre-intervention coronary CTA and optical coherence tomography (OCT) were enrolled. PCAT attenuation was measured around the culprit lesion and in the proximal 40 mm of all coronary arteries. PCAT attenuation and OCT findings were compared between patients with ACS versus SAP. Results Among 471 patients (ACS: 198, SAP: 273), PCAT attenuation was higher in ACS patients than in SAP patients both at the culprit plaque level (−67.5±9.6 Hounsfield unit [HU] vs. −71.5±11.0 HU, p<0.001) and the culprit vessel level (−68.3±7.7 HU vs. −71.1±7.9 HU, p<0.001). The mean PCAT attenuation of all 3 coronary arteries was also significantly higher in ACS patients than in SAP patients (−68.8±6.3 HU vs. −70.5±7.1 HU, p=0.007). After adjusting patient characteristics, not only thin-cap fibroatheroma (OR: 2.44; 95% CI: 1.63–3.65) and macrophages (OR: 2.07; 95% CI: 1.34–3.21) but also PCAT attenuation in the culprit plaque (OR: 1.04; 95% CI: 1.02–1.06) was associated with the clinical presentation of ACS. Conclusions PCAT attenuation at culprit plaque, culprit vessel, and pan-coronary levels was higher in ACS patients than in SAP patients. Vascular inflammation appears to play a crucial role in the development of ACS. Funding Acknowledgement Type of funding sources: None.
Background An incidence of cardiovascular events increases with age in women. The relationship between cardiovascular risk factors, and the underlying pathology and the prevalence of lipid plaques has not been systematically studied in different age groups in women presented with acute coronary syndromes (ACS). Purpose We investigated the underlying pathology and the prevalence of lipid plaques in culprit lesions by optical coherence tomography (OCT) in women with different risk factors. Methods A total of 382 women who underwent pre-intervention OCT imaging were included. The underlying pathology and the prevalence of lipid plaques in the culprit lesion was compared between women with and without cardiovascular risk factors (i.e. hypertension, smoking, hyperlipidemia, diabetes mellitus, family history and chronic kidney disease) in three different age groups (<60 yr, 60–70 yr, >70 yr). Results The relative prevalence of plaque erosion was higher in younger women (<60 yr) and decreased with age (from 51% to 28%, p<0.001). There was no significant difference in the prevalence of lipid plaques between women with and without risk factors, except a higher prevalence of lipid plaques in current smokers compared to non-smokers (79% vs. 63%, p=0.003). In women with hyperlipidemia, the prevalence of lipid plaques was modest in young ages (<60 yr), but increased steeply with age (p<0.001). The increasing age trend for lipid plaque was also observed in women with hypertension (p=0.03) and current smokers (p=0.01). In women with diabetes mellitus and family history, the prevalence of lipid plaques was high even in young ages (<60 yr) and did not increase with age. Conclusion The prevalence of plaque erosion was higher in younger women (<60 yr) and decreased with age. Current smokers had significantly higher prevalence of lipid plaque. Patients with diabetes and positive family history had a higher prevalence of lipid plaque at young age. The prevalence of lipid plaques increased with age particularly in women with hyperlipidemia and hypertension. Funding Acknowledgement Type of funding sources: Foundation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.