deepvessel

Back to Homepage

DeepVessel: Vessel Segmentation via Deep Learning

Introduction:

Retinal vessel segmentation is a fundamental step for various ocular imaging applications. In this paper, we formulate the retinal vessel segmentation problem as a boundary detection task and solve it using a novel deep learning architecture. Our method is based on two key ideas: (1) applying a multi-scale and multi-level Convolutional Neural Network (CNN) with a side-output layer to learn a rich hierarchical representation, and (2) utilizing a Conditional Random Field (CRF) to model the long-range interactions between pixels. We combine the CNN and CRF layers into an integrated deep network called DeepVessel. Our experiments show that the DeepVessel system achieves state-of-the-art retinal vessel segmentation performance on the DRIVE, STARE, and CHASE_DB1 datasets with an efficient running time.

Figure: (A) Fundus images from the DRIVE and STARE datasets, (B) Ground truth, (C) Fusion results of side-output layers, (D) Our DeepVessel results, (E) Thresholded DeepVessel results.

Paper:

Huazhu Fu, Yanwu Xu, Stephen Lin, Damon Wing Kee Wong, Jiang Liu, "DeepVessel: Retinal Vessel Segmentation via Deep Learning and Conditional Random Field", in International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2016, pp. 132-139.

Results on DRIVE, STARE, and CHASE_DB1 datasets:
[on OneDrive] [on BaiduYun](~44MB)

Related Links:

The vessel segmentation datasets:
1. DRIVE dataset: Link
2. STARE dataset: Link
3. CHASE_DB1 dataset: Link