M2U-Net: Effective and Efficient Retinal Vessel Segmentation for Real-World Applications

Abstract

In this paper, we present a novel neural network architecture for retinal vessel segmentation that improves over the state of the art on two benchmark datasets, is the first to run in real time on high resolution images, and its small memory and processing requirements make it deployable in mobile and embedded systems.The M2U-Net has a new encoder-decoder architecture that is inspired by the U-Net. It adds pretrained components of MobileNetV2 in the encoder part and novel contractive bottleneck blocks in the decoder part that, combined with bilinear upsampling, drastically reduce the parameter count to 0.55M compared to 31.03M in the original U-Net.We have evaluated its performance against a wide body of previously published results on three public datasets. On two of them, the M2U-Net achieves new state-of-the-art performance by a considerable margin. When implemented on a GPU, our method is the first to achieve real-time inference speeds on high-resolution fundus images. We also implemented our proposed network on an ARM-based embedded system where it segments images in between 0.6 and 15 sec, depending on the resolution. Thus, the M2U-Net enables a number of applications of retinal vessel structure extraction, such as early diagnosis of eye diseases, retinal biometric authentication systems, and robot assisted microsurgery.

BSCN: bidirectional symmetric cascade network for retinal vessel segmentation

“…Jin et al [28] proposed DUNet for retinal vessel segmentation in an end to end manner and experimented on DRIVE [25], STARE [29] and CHASE_DB1 [30] dataset. Laibacher et al [31] improved the traditional U-Net and proposed M2U-net, added the pre-training component of MobileNetV2 in the encoder part, added the new bottleneck block in the decoder part, and integrate with bilinear sampling, reduced the number of parameters greatly. Inspired by the success of ResNet [32] and R2U-Net [33], Zhuang et al [34] proposed LadderNet for retinal blood vessel.…”

“…Table 9 shows the average computation time comparison of different models for retinal image vessel segmentation. The DUNet [28], M2U-Net [31] and R2U-Net [33] all use the encoder and the decoder structure to achieve vessel segmentation, while the downsampling and upsampling process produce lots of computational redundancy. Our method produces two blood vessel contour prediction maps from the high-level to the low-level and low-level to high-level paths in each scale detection block.…”

Automated Segmentation of Retinal Fluid Volumes From Structural and Angiographic Optical Coherence Tomography Using Deep Learning

“…Contemporary deep-learning-based methods have shown a great advantage for image segmentation tasks. [11][12][13][14][15][16][17][18] In ophthalmology, researchers have proposed a number of deep neural networks to solve specific problems, such as retinal layer segmentation in OCT, [19][20][21] retinal vessel segmentation in fundus photography, 22,23 choroidal neovascularization segmentation, 24 high-resolution reconstruction on OCT angiograms, 25 and retinal nonperfusion area segmentation in OCTA. [26][27][28] Deep-learning-based retinal fluid segmentation on cross-sectional OCT has also been reported by many scholars.…”

Chaining a U-Net With a Residual U-Net for Retinal Blood Vessels Segmentation

“…One example is the hemorrhage, a disorder of the eye in which bleeding occurs in the light-sensitive tissue on the back wall of the eye [6]. It can be related to diabetic retinopathy, which cause the formation of small fragile blood vessels, which can be easily damaged by high blood sugar level, potentially causing the growth or creation of new blood vessels [7]. Another related disease is hypertensive retinopathy, which consists of damage to the retina from high blood pressure which may result in deformation of the retinal blood vessels.…”