Construction of Retinal Vessel Segmentation Models Based on Convolutional Neural Network
- PDF / 2,578,307 Bytes
- 18 Pages / 439.37 x 666.142 pts Page_size
- 72 Downloads / 202 Views
Construction of Retinal Vessel Segmentation Models Based on Convolutional Neural Network Qiangguo Jin1 · Qi Chen1,2 · Zhaopeng Meng1,3 · Bing Wang4 · Ran Su1
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract Segmentation of retinal vessels in fundus images plays a very important role in diagnosing relevant diseases. In this paper, we have constructed automated segmentation models for the retinal vessel segmentation task based on convolutional neural networks. Since some typical deep convolutional neural networks need to be fed by high-resolution patches, small retinal patches should be interpolated to the specific resolution. The interpolated patches sometimes would introduce additional noises. Thus, we modify some typical deep architectures by inserting a set of convolutional layers. In this way, our models have the ability to adapt to different resolutions. Overall, five models are analyzed and compared in our studies including LeNet, M-AlexNet (modified AlexNet), M-ZF-Net (Modified ZF-Net), M-VGG (Modified VGG) and Deformable-ConvNet. Deformable-ConvNet captures the vascular structure and is used to do the retinal vessel segmentation task for the first time. We train the models from scratch and compare their ability to discriminate vessels/non-vessel pixels on two retinal fundus image datasets, DRIVE and STARE. Results are analyzed and compared in our studies. We obtain the highest accuracy of 0.9628/0.9690, lowest loss of 0.1045/0.0968, and highest AUC of 0.9764/0.9844 on DRIVE/STARE respectively. We also compare the CNN models with other segmentation methods. The results demonstrate the high effectiveness of the CNN-based approaches. Keywords Retinal blood vessel · Segmentation · Convolutional neural networks (CNN) · Modified CNN
1 Introduction The morphological and topographical changes of retinal vessels may indicate some pathological diseases, such as cardiovascular and ophthalmologic diseases [1], including diabetes, hypertension, and diabetic retinopathy (DR). DR is a complication of diabetes, accompanying with the swelling of the retinal vessels, caused by elevated blood sugar levels [2].
Qiangguo Jin and Qi Chen have contributed equally to this study.
B
Ran Su [email protected]
Extended author information available on the last page of the article
123
Q. Jin et al.
Detection and segmentation of retinal vessels can benefit the screening programs for diabetic retinopathy [3]. This boosts the proposal of more accurate retinal blood vessel segmentation methods to facilitate the early diagnosis of pathological diseases. Due to the complicated vascular tree such as length, width, branching, and angles [1], manual segmentation of retinal vessel is a tedious, time-consuming, and in high-demand of skilled technical staff method [4]. Therefore, it is commonly accepted by the medical community that the automatic detection of retinal vessels is a vital and challenging step in the development of a computer-assisted diagnostic system [5]. Deep learning, especially convolutional n
Data Loading...
