Partial Vessels Annotation-based Coronary Artery Segmentation with Self-training and Prototype Learning
This work addresses the need for label-efficient algorithms in medical imaging to reduce the labor-intensive annotation process for coronary artery segmentation, representing an incremental improvement in domain-specific methods.
The paper tackles the problem of coronary artery segmentation from CCTA images with limited annotations by proposing a partial vessels annotation (PVA) approach and a progressive weakly supervised learning framework, achieving a 24.29% vessel annotation rate and comparable trunk continuity to fully annotated baselines.
Coronary artery segmentation on coronary-computed tomography angiography (CCTA) images is crucial for clinical use. Due to the expertise-required and labor-intensive annotation process, there is a growing demand for the relevant label-efficient learning algorithms. To this end, we propose partial vessels annotation (PVA) based on the challenges of coronary artery segmentation and clinical diagnostic characteristics. Further, we propose a progressive weakly supervised learning framework to achieve accurate segmentation under PVA. First, our proposed framework learns the local features of vessels to propagate the knowledge to unlabeled regions. Subsequently, it learns the global structure by utilizing the propagated knowledge, and corrects the errors introduced in the propagation process. Finally, it leverages the similarity between feature embeddings and the feature prototype to enhance testing outputs. Experiments on clinical data reveals that our proposed framework outperforms the competing methods under PVA (24.29% vessels), and achieves comparable performance in trunk continuity with the baseline model using full annotation (100% vessels).