MDAN-UNet: Multi-Scale and Dual Attention Enhanced Nested U-Net Architecture for Segmentation of Optical Coherence Tomography Images

Abstract

Optical coherence tomography (OCT) is an optical high-resolution imaging technique for ophthalmic diagnosis. In this paper, we take advantages of multi-scale input, multi-scale side output and dual attention mechanism and present an enhanced nested U-Net architecture (MDAN-UNet), a new powerful fully convolutional network for automatic end-to-end segmentation of OCT images. We have evaluated two versions of MDAN-UNet (MDAN-UNet-16 and MDAN-UNet-32) on two publicly available benchmark datasets which are the Duke Diabetic Macular Edema (DME) dataset and the RETOUCH dataset, in comparison with other state-of-the-art segmentation methods. Our experiment demonstrates that MDAN-UNet-32 achieved the best performance, followed by MDAN-UNet-16 with smaller parameter, for multi-layer segmentation and multi-fluid segmentation respectively.

Representation of Traffic Congestion Data for Urban Road Traffic Networks Based on Pooling Operations

“…The nearest neighbor value (NNV), as defined by Equation 2, is based on a common image resampling algorithm [51]. In our experiment, this operation returns the value in the upper left corner of a grid [52];…”

TMD-Unet: Triple-Unet with Multi-Scale Input Features and Dense Skip Connection for Medical Image Segmentation

“…The approaches of these studies were to change the internal structure of the nodes in the encoder and decoder blocks [ 29 , 30 , 31 , 32 ] or change the connection between the blocks [ 33 , 34 ]. Other approaches were to change the skip connection of the conventional Unet architecture [ 9 , 35 , 36 , 37 ]. Some studies used a cascade structure [ 10 , 38 , 39 ], or used the hybrid methods [ 8 , 40 , 41 , 42 ].…”

“…The skip connection path, which consists of 3 × 3 filters and 1 × 1 filters accompany the residual connections replaced the traditional skip connection. Liu et al [ 37 ] integrated the multi-scale input, multi-scale side output, and attention mechanism into the Unet++ for optical coherence tomography image segmentation.…”

A Multiple Layer U-Net, Un-Net, for Liver and Liver Tumor Segmentation in CT

“…They used deep supervision to learn the full-scale aggregated feature maps. Liu et al [37] have enhanced the Unet++ [35] by integrating the multi-scale input, multi-scale side output, and an attention mechanism segmentation on optical coherence tomography images.…”