Diffusion-Based User-Guided Data Augmentation for Coronary Stenosis Detection
This work addresses the challenge of data scarcity for clinicians and researchers in medical imaging, specifically for coronary stenosis assessment, though it is incremental as it builds on existing diffusion models for data augmentation.
The authors tackled the problem of limited labeled data and class imbalance in automated coronary stenosis detection by proposing a diffusion-based data augmentation method that generates realistic lesions with user-guided severity control, achieving superior performance in lesion detection and severity classification across various dataset sizes, including on a large-scale in-house dataset and a public dataset.
Coronary stenosis is a major risk factor for ischemic heart events leading to increased mortality, and medical treatments for this condition require meticulous, labor-intensive analysis. Coronary angiography provides critical visual cues for assessing stenosis, supporting clinicians in making informed decisions for diagnosis and treatment. Recent advances in deep learning have shown great potential for automated localization and severity measurement of stenosis. In real-world scenarios, however, the success of these competent approaches is often hindered by challenges such as limited labeled data and class imbalance. In this study, we propose a novel data augmentation approach that uses an inpainting method based on a diffusion model to generate realistic lesions, allowing user-guided control of severity. Extensive evaluation on lesion detection and severity classification across various synthetic dataset sizes shows superior performance of our method on both a large-scale in-house dataset and a public coronary angiography dataset. Furthermore, our approach maintains high detection and classification performance even when trained with limited data, highlighting its clinical importance in improving the assessment of severity of stenosis and optimizing data utilization for more reliable decision support.