Deep Self-knowledge Distillation: A hierarchical supervised learning for coronary artery segmentation

2025-09-03

AI Summary

Coronary artery X-ray angiography segmentation faces challenges including inefficient manual annotation, low accuracy and poor generalizability of existing automated methods—whether rule-based or deep learning–based—and insufficient exploitation of hierarchical knowledge within models, particularly in knowledge distillation strategies. To address these issues, we propose a **hierarchical deep self-knowledge distillation framework**, the first to introduce self-distillation to this task. Our method jointly optimizes a **deep distribution loss**, modeling global probability distributions, and a **pixel-wise distillation loss**, enforcing fine-grained structural supervision, thereby establishing a dual-regularized hierarchical knowledge transfer mechanism. Evaluated on the XCAD and DCA1 benchmarks, our approach consistently outperforms state-of-the-art methods, achieving superior Dice score, IoU, sensitivity, and accuracy. Results demonstrate that hierarchical supervision significantly enhances model expressiveness and robustness.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Unsupervised & Self-Supervised Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Coronary artery disease is a leading cause of mortality, underscoring the critical importance of precise diagnosis through X-ray angiography. Manual coronary artery segmentation from these images is time-consuming and inefficient, prompting the development of automated models. However, existing methods, whether rule-based or deep learning models, struggle with issues like poor performance and limited generalizability. Moreover, current knowledge distillation methods applied in this field have not fully exploited the hierarchical knowledge of the model, leading to certain information waste and insufficient enhancement of the model's performance capabilities for segmentation tasks. To address these issues, this paper introduces Deep Self-knowledge Distillation, a novel approach for coronary artery segmentation that leverages hierarchical outputs for supervision. By combining Deep Distribution Loss and Pixel-wise Self-knowledge Distillation Loss, our method enhances the student model's segmentation performance through a hierarchical learning strategy, effectively transferring knowledge from the teacher model. Our method combines a loosely constrained probabilistic distribution vector with tightly constrained pixel-wise supervision, providing dual regularization for the segmentation model while also enhancing its generalization and robustness. Extensive experiments on XCAD and DCA1 datasets demonstrate that our approach outperforms the dice coefficient, accuracy, sensitivity and IoU compared to other models in comparative evaluations.

Problem

• Automating coronary artery segmentation from X-ray angiography images
• Overcoming poor performance and limited generalizability in existing methods
• Addressing hierarchical knowledge underutilization in current distillation techniques

Innovation

• Hierarchical outputs for supervision
• Deep Distribution Loss combination
• Pixel-wise Self-knowledge Distillation