Deep Learning-Based Quantification of Intraplaque Hemorrhage Reveals Impacts on Long-Term Carotid Plaque Progression

Motivation: Carotid atherosclerosis is a leading cause of stroke, and intraplaque hemorrhage (IPH) is a significant predictor of plaque progression and rupture. Understanding the role of IPH in asymptomatic patients over time is essential for effective management. Goal(s): Develop a precise, reproducible deep learning-based algorithm for IPH segmentation and apply it to analyze long-term plaque burden evolution. Approach: A segmentation algorithm was trained using 3D-SNAP images. The algorithm's validation included histology comparisons and reproducibility analysis. It was applied to long-term repeated scans to assess the relationship between IPH and plaque burden. Results: IPH presence and volume were significantly associated with greater plaque burden progression. Impact: This study underscores the significance of IPH in carotid plaque progression, offering a precise deep learning-based tool for monitoring. It enables more effective risk assessment and personalized management strategies, sparking new research into long-term IPH effects on asymptomatic patients.