A Cross-Modality Learning Approach for Vessel Segmentation in Retinal Images

Abstract

This paper presents a new supervised method for vessel segmentation in retinal images. This method remolds the task of segmentation as a problem of cross-modality data transformation from retinal image to vessel map. A wide and deep neural network with strong induction ability is proposed to model the transformation, and an efficient training strategy is presented. Instead of a single label of the center pixel, the network can output the label map of all pixels for a given image patch. Our approach outperforms reported state-of-the-art methods in terms of sensitivity, specificity and accuracy. The result of cross-training evaluation indicates its robustness to the training set. The approach needs no artificially designed feature and no preprocessing step, reducing the impact of subjective factors. The proposed method has the potential for application in image diagnosis of ophthalmologic diseases, and it may provide a new, general, high-performance computing framework for image segmentation.

Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search

“…One of the most established realizations of deep learning is the convolutional neural network (CNN) [38], which automatically learns a hierarchy of increasingly complex features and a related classifier directly from training data sets. Recent works have extended the CNN framework to complex medical image analysis, such as retinal blood vessels segmentation [39,40], retinal hemorrhage detection [41], brain tumor segmentation [42], and cerebral microbleeds detection [43]. Very recently, a CNN based method was proposed [44] as an alternative to classic machine learning methods [5] for classification of normal and pathologic OCT images.…”

Retinal Vessel Segmentation from Simple to Difficult

“…Compared to supervised methods, the proposed method still achieves better sensitivity. The performance of the supervised methods [8] and [10] significantly outperforms the proposed method on DRIVE when the models are trained on the training images from DRIVE. However when the models are trained on STARE, their accuracies decrease to 0.9456 and 0.9486 respectively while ours is 0.9451.…”

“…In [9], gray-level and moment invariants based features are extracted to learn a neural network scheme for vessel segmentation. In [10], a deep neural network is trained to learn a cross-modality data transform from retinal image to vessel map. Supervised methods are more invariable to vessel deformations and brightness since they combine different priors about the vessels.…”

Performance Analysis of Adaptive NEBF Filter Using Cellular Automata for Retinal Vessel Segmentation

“…The actual retinal problematic vein morphology distinguishes a accelerating stages of development of varied sight debilitating illnesses and thus clears a strategy to help characterize it has the seriousness. Qiaoliang et al [13] has presented a broad and strong nerve organs system with formidable induction potential is actually consist of design of the particular transformation. Chengzhang et al [14] has presented a method for segmentation of retina eye image with the help of a technique known as extreme learning machine which uses 3-D element view of every pixel of an eye image differencing with the help of neighboring pixels and functions implied on those pixels…”