Annotation-Free Cardiac Vessel Segmentation via Knowledge Transfer from Retinal Images
This addresses the challenge of time-consuming and infeasible manual annotation for medical imaging in cardiology, though it is incremental as it adapts existing transfer learning methods to a new domain.
The paper tackles the problem of segmenting coronary arteries without manual annotations by transferring knowledge from annotated retinal images, achieving superior accuracy on a dataset of 1092 angiography images.
Segmenting coronary arteries is challenging, as classic unsupervised methods fail to produce satisfactory results and modern supervised learning (deep learning) requires manual annotation which is often time-consuming and can some time be infeasible. To solve this problem, we propose a knowledge transfer based shape-consistent generative adversarial network (SC-GAN), which is an annotation-free approach that uses the knowledge from publicly available annotated fundus dataset to segment coronary arteries. The proposed network is trained in an end-to-end fashion, generating and segmenting synthetic images that maintain the background of coronary angiography and preserve the vascular structures of retinal vessels and coronary arteries. We train and evaluate the proposed model on a dataset of 1092 digital subtraction angiography images, and experiments demonstrate the supreme accuracy of the proposed method on coronary arteries segmentation.