BSCN: bidirectional symmetric cascade network for retinal vessel segmentation

Abstract

Background: Retinal blood vessel segmentation has an important guiding significance for the analysis and diagnosis of cardiovascular diseases such as hypertension and diabetes. But the traditional manual method of retinal blood vessel segmentation is not only time-consuming and laborious but also cannot guarantee the accuracy and efficiency of diagnosis. Therefore, it is especially significant to create a computer-aided method of automatic and accurate retinal vessel segmentation.

Methods: In order to extract the blood vessels' contours of different diameters to realize fine segmentation of retinal vessels, we propose a Bidirectional Symmetric Cascade Network (BSCN) where each layer is supervised by vessel contour labels of specific diameter scale instead of using one general ground truth to train different network layers. In addition, to increase the multi-scale feature representation of retinal blood vessels, we propose the Dense Dilated Convolution Module (DDCM), which extracts retinal vessel features of different diameters by adjusting the dilation rate in the dilated convolution branches and generates two blood vessel contour prediction results by two directions respectively. All dense dilated convolution module outputs are fused to obtain the final vessel segmentation results.

Results: We experimented the three datasets of DRIVE, STARE, HRF and CHASE_DB1, and the proposed method reaches accuracy of 0.9846/0.9872/0.9856/0.9889 and AUC of 0.9874/0.9941/0.9882/0.9874 on DRIVE, STARE, HRF and CHASE_DB1.

Conclusions: The experimental results show that compared with the state-of-art methods, the proposed method has strong robustness, it not only avoids the adverse interference of the lesion background but also detects the tiny blood vessels at the intersection accurately.

Keywords: Bidirectional symmetric cascade network; Dense dilated convolution; Retinal vessel segmentation; Scale detection; Specific diameter scale.

Fig. 1

Comparison of two manual segmentation results of the retinal. a original image b 1st manual label c 2nd manual label

Fig. 2

The overall architecture of the bidirectional symmetric cascade network

Fig. 3

The detailed architecture of the bidirectional symmetric cascade network

Fig. 4

The detailed architecture of the dense dilation convolution module

Fig. 5

Examples of blood vessel contour detected by different dense dilated convolution module (DDCM). Each DDCM generates two blood vessel contour predictions, P^l2h and P^h2l, respectively

Fig. 6

Image patches on DRIVE. a patches of the original image b ground truth patches corresponding to the original image

Fig. 7

ROC curve of different methods. a ROC curve on DRIVE b ROC curve on STARE c ROC curve on CHASE_DB1 d ROC curve on HRF

Fig. 8

Qualitative results comparison of different methods on DRIVE dataset

Fig. 9

Local detail results comparison of different methods on DRIVE dataset

Fig. 10

Accuracy and loss results comparison of different method on DRIVE dataset. a training set results on DRIVE b validation set results on DRIVE

Fig. 11

Qualitative results comparison of different methods on STARE dataset

Fig. 12

Local detail results comparison of different methods on STARE dataset

Fig. 13

ACC and LOSS results comparison of different method on STARE dataset. a training set results on STARE b validation set results on STARE

Fig. 14

Qualitative results on HRF dataset

Fig. 15

Qualitative results comparion of different methods on CHASE_DB1 dataset

Fig. 16

ACC and LOSS results comparison of different method on CHASE_DB1 dataset. a training set results on CHASE_DB1 b validation set results on CHASE_DB1

Fig. 17

Experimental results with VE-MSC [45] and proposed method on DRIVE and STARE datasets. a results on DRIVE b results on SATRE