StenUNet: Automatic Stenosis Detection from X-ray Coronary Angiography

TL;DR

This paper introduces StenUNet, an automated machine learning-based method for detecting coronary artery stenosis from X-ray angiography images, aiming to improve diagnostic reliability and assist clinical decision-making.

Contribution

The paper presents a novel deep learning architecture, StenUNet, specifically designed for stenosis detection, and demonstrates its competitive performance in the ARCADE challenge.

Findings

Achieved an F1 score of 0.5348 on the test set.

Placed 3rd in the ARCADE challenge.

Performed close to the second-best team with minimal score difference.

Abstract

Coronary angiography continues to serve as the primary method for diagnosing coronary artery disease (CAD), which is the leading global cause of mortality. The severity of CAD is quantified by the location, degree of narrowing (stenosis), and number of arteries involved. In current practice, this quantification is performed manually using visual inspection and thus suffers from poor inter- and intra-rater reliability. The MICCAI grand challenge: Automatic Region-based Coronary Artery Disease diagnostics using the X-ray angiography imagEs (ARCADE) curated a dataset with stenosis annotations, with the goal of creating an automated stenosis detection algorithm. Using a combination of machine learning and other computer vision techniques, we propose the architecture and algorithm StenUNet to accurately detect stenosis from X-ray Coronary Angiography. Our submission to the ARCADE challenge placed 3rd among all teams. We achieved an F1 score of 0.5348 on the test set, 0.0005 lower than the 2nd place.