Abstract
Automatic segmentation and centerline extraction of blood vessels from retinal fundus images is an essential step to measure the state of retinal blood vessels and achieve the goal of auxiliary diagnosis. Combining the information of blood vessel segments and centerline can help improve the continuity of results and performance. However, previous studies have usually treated these two tasks as separate research topics. Therefore, we propose a novel multitask learning network (MSC-Net) for retinal vessel segmentation and centerline extraction. The network uses a multibranch design to combine information between two tasks. Channel and atrous spatial fusion block (CAS-FB) is designed to fuse and correct the features of different branches and different scales. The clDice loss function is also used to constrain the topological continuity of blood vessel segments and centerline. Experimental results on different fundus blood vessel datasets (DRIVE, STARE, and CHASE) show that our method can obtain better segmentation and centerline extraction results at different scales and has better topological continuity than state-of-the-art methods.
Highlights
The morphological transformation of the retinal vascular system is closely related to various ophthalmological and cardiovascular diseases, such as glaucoma, hypertension, diabetes, and arteriosclerosis [1]
The supervised methods for retinal blood vessel segmentation are based on traditional machine learning methods and some deep learning methods
This paper proposes a multitask learning network to combine the two tasks, blood vessel segmentation and centerline extraction
Summary
The morphological transformation of the retinal vascular system is closely related to various ophthalmological and cardiovascular diseases, such as glaucoma, hypertension, diabetes, and arteriosclerosis [1]. The development of computer vision has led to the emergence of many methods for retinal blood vessel segmentation. These methods can usually be organized into traditional image processing methods and machine learning methods. Many machine learning methods, especially deep learning methods, have been used for retinal vascular segmentation and achieved good results. Machine learning methods can be further divided into supervised and unsupervised methods Unsupervised methods, such as the Gaussian mixture model [11], fuzzy techniques [12], and k-means clustering [13], do not require labeled data for blood vessel segmentation. The supervised methods for retinal blood vessel segmentation are based on traditional machine learning methods and some deep learning methods. Orlando et al [14] used support vector machines for vessel segmentation, and Fraz et al [15] designed an integrated system based on decision tree to achieve the segmentation of blood vessels from retinal images
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
