MDU-Net: Multi-scale Densely Connected U-Net for biomedical image segmentation

Abstract

Radiologist is "doctor's doctor", biomedical image segmentation plays a central role in quantitative analysis, clinical diagnosis, and medical intervention. In the light of the fully convolutional networks (FCN) and U-Net, deep convolutional networks (DNNs) have made significant contributions in biomedical image segmentation applications. In this paper, based on U-Net, we propose MDUnet, a multiscale densely connected U-net for biomedical image segmentation. we propose three different multi-scale dense connections for U shaped architectures encoder, decoder and across them. The highlights of our architecture is directly fuses the neighboring different scale feature maps from both higher layers and lower layers to strengthen feature propagation in current layer. Which can largely improves the information flow encoder, decoder and across them. Multi-scale dense connections, which means containing shorter connections between layers close to the input and output, also makes much deeper U-net possible. We adopt the optimal model based on the experiment and propose a novel Multi-scale Dense U-Net (MDU-Net) architecture with quantization. Which reduce overfitting in MDU-Net for better accuracy. We evaluate our purpose model on the MICCAI 2015 Gland Segmentation dataset (GlaS). The three multi-scale dense connections improve Unet performance by up to 1.8% on test A and 3.5% on test B in the MICCAI Gland dataset. Meanwhile the MDU-net with quantization achieves the superiority over U-Net performance by up to 3% on test A and 4.1% on test B.

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

“…We first presented UNet++ in our DLMIA 2018 paper [51]. UNet++ has since been quickly adopted by the research community, either as a strong baseline for comparison [52], [53], [54], [55], or as a source of inspiration for developing newer semantic segmentation architectures [56], [57], [58], [59], [60], [61]; it has also been utilized for multiple applications, such as segmenting objects in biomedical images [62], [63], natural images [64], and satellite images [65], [66]. Recently, Shenoy [67] has independently and systematically investigated UNet++ for the task of "contact prediction model PconsC4", demonstrating significant improvement over widely-used U-Net.…”

Learning to Segment Microscopy Images with Lazy Labels

“…These weak labels together with a part of the strong labels are used for training the multi-task U-net. In this experiment, we test the algorithm on different ratios of SL, and compare it with the baseline U-net (single task), PL approach (where PL are generated in the same way as the ones for the SEM dataset), and a fully supervised approach called Multi-scale Densely Connected U-Net (MDUnet) [45]. The results on two sets of test data are reported in Table 2.…”

A Novel Elastomeric UNet for Medical Image Segmentation

“…The improvement of UNet performance is apt to stagnate by the increase of depth. As a consequence, the diverse variants of UNet had been explored in the past 5 years for medical image segmentation Zhang et al, 2018;Liu et al, 2020). Inspired by the ResNet framework, a residual neural network with U-shape named U-ResNet was explored by Ibtehaz and Rahman (2020) and Drozdzal et al (2016).…”