AI-based tool can measure plaque burden, stenosis severity on CCTA

USA: A recent study showed that in patients with stable chest pain, an artificial intelligence (AI)-based tool can rapidly and accurately measure plaque volume and stenosis severity from the images of coronary CT angiography (CCTA). Also, it has the potential for better prediction of future myocardial infarction (MI) as it agrees closely with expert readers and intravascular ultrasound. The study appears in the journal Lancet: Digital Health. 

Coronary CT angiography is the first-line test for the evaluation of coronary artery stenosis severity. AI algorithms are increasingly being applied to CCTAfor improving the efficiency and accuracy of image analysis, demonstrating high performance when compared with expert readers. Deep learning is a type of AI that uses artificial neural networks for generating automated predictions directly from image data. 

Quantification of Atherosclerotic plaque from CCTA enables accurate assessment of coronary artery disease burden and prognosis. Andrew Lin, Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, USA, and colleagues sought to develop and validate a deep learning system for CCTA-derived measures of plaque volume and stenosis severity in an international, multicentre study.

The study included nine cohorts of patients undergoing CCTA at 11 sites. They were assigned to training and test sets. Data were retrospectively collected on patients with a wide range of clinical presentations of coronary artery disease who underwent CCTA between Nov 18, 2010, and Jan 25, 2019. 

The researchers trained a novel deep learning convolutional neural network to segment coronary plaque in 921 patients (5045 lesions). Then, the deep learning network was applied to an independent test set that included an external validation cohort of 175 patients (1081 lesions) and 50 patients (84 lesions). This was assessed by intravascular ultrasound within 1 month of CCTA. 

The prognostic value of deep learning-based plaque measurements was evaluated for fatal or non-fatal myocardial infarction (primary outcome) in 1611 patients from the prospective SCOT-HEART trial.