A Deep Learning Model for Coronary Artery Segmentation and Quantitative Stenosis Detection in Angiographic Images

TL;DR

This work addresses automated coronary artery segmentation and quantitative stenosis detection from angiographic images to reduce observer variability in CAD assessment. It introduces SAM-VMNet, a dual-branch architecture that fuses MedSAM's local feature extraction with VM-UNet's long-range context, paired with a dynamic queue-based stenosis detection along vessel centerlines. On a hybrid dataset and the ARCADE subset, the method achieves IoU around $0.63$, high specificity near $1.00$, and strong stenosis metrics ($TPR\approx0.59$, $PPV\approx0.59$, $ARMSE\approx1.73$, $RRMSE\approx0.167$), indicating robust segmentation and accurate local stenosis quantification. The approach offers a practical, end-to-end tool for automated CAD assessment with potential clinical impact, with code available at the provided GitHub repository.

Abstract

Coronary artery disease (CAD) is a leading cause of cardiovascular-related mortality, and accurate stenosis detection is crucial for effective clinical decision-making. Coronary angiography remains the gold standard for diagnosing CAD, but manual analysis of angiograms is prone to errors and subjectivity. This study aims to develop a deep learning-based approach for the automatic segmentation of coronary arteries from angiographic images and the quantitative detection of stenosis, thereby improving the accuracy and efficiency of CAD diagnosis. We propose a novel deep learning-based method for the automatic segmentation of coronary arteries in angiographic images, coupled with a dynamic cohort method for stenosis detection. The segmentation model combines the MedSAM and VM-UNet architectures to achieve high-performance results. After segmentation, the vascular centerline is extracted, vessel diameter is computed, and the degree of stenosis is measured with high precision, enabling accurate identification of arterial stenosis. On the mixed dataset (including the ARCADE, DCA1, and GH datasets), the model achieved an average IoU of 0.6308, with sensitivity and specificity of 0.9772 and 0.9903, respectively. On the ARCADE dataset, the average IoU was 0.6303, with sensitivity of 0.9832 and specificity of 0.9933. Additionally, the stenosis detection algorithm achieved a true positive rate (TPR) of 0.5867 and a positive predictive value (PPV) of 0.5911, demonstrating the effectiveness of our model in analyzing coronary angiography images. SAM-VMNet offers a promising tool for the automated segmentation and detection of coronary artery stenosis. The model's high accuracy and robustness provide significant clinical value for the early diagnosis and treatment planning of CAD. The code and examples are available at https://github.com/qimingfan10/SAM-VMNet.

Figures (10)

• Figure 1: Workflow of our method. (a) Artery segmentation using a deep learning-based approach; (b) Stenosis detection, including arterial centerline extraction, arterial diameter calculation, and stenosis degree determination.
• Figure 2: Diagram of datasets for SAM-VMNet training and evaluation.
• Figure 3: Structure of the vssblock.
• Figure 4: SAM-VMNet architecture diagram. SAM-VMNet contains two parallel branches.
• Figure 5: Stenosis detection flow chart. (a): segmentation results, (b): extraction of center line and segmentation, (c): measurement of diameter and detection of stenosis points (shown as one section), where blue points are on the center line, and red points are stenosis points,(d): final stenosis detection results,where red color is labeled as severe stenosis, green color is moderate stenosis, and blue color is mild stenosis.