A-DenseUNet: Adaptive Densely Connected UNet for Polyp Segmentation in Colonoscopy Images with Atrous Convolution.

Sirojbek Safarov,Taeg Keun Whangbo

doi:10.3390/s21041441

Abstract

Colon carcinoma is one of the leading causes of cancer-related death in both men and women. Automatic colorectal polyp segmentation and detection in colonoscopy videos help endoscopists to identify colorectal disease more easily, making it a promising method to prevent colon cancer. In this study, we developed a fully automated pixel-wise polyp segmentation model named A-DenseUNet. The proposed architecture adapts different datasets, adjusting for the unknown depth of the network by sharing multiscale encoding information to the different levels of the decoder side. We also used multiple dilated convolutions with various atrous rates to observe a large field of view without increasing the computational cost and prevent loss of spatial information, which would cause dimensionality reduction. We utilized an attention mechanism to remove noise and inappropriate information, leading to the comprehensive re-establishment of contextual features. Our experiments demonstrated that the proposed architecture achieved significant segmentation results on public datasets. A-DenseUNet achieved a 90% Dice coefficient score on the Kvasir-SEG dataset and a 91% Dice coefficient score on the CVC-612 dataset, both of which were higher than the scores of other deep learning models such as UNet++, ResUNet, U-Net, PraNet, and ResUNet++ for segmenting polyps in colonoscopy images.

Highlights

We evaluated our model on the Kvasir-SEG and CVC-612 datasets, and the experimental results show that it achieved the highest intersection over union (IoU) and
The results show that the predicted segmentation mask in A-DenseUNet is closer to the ground truth mask than that in other state-ofthe-art architectures
We have presented an end-to-end biomedical image segmentation architecture, A-DenseUNet, to achieve more accurate segmentation results

Summary

Introduction

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. The third most common form of cancer worldwide for both men and women is colorectal cancer, and its prevalence is increasing every year [1]. The primary cause of colorectal cancer is the growth of glandular tissue in the colonic mucosa. Precise and earlier determination of polyps from virtual colonoscopy screenings is of great significance for the avoidance and timely treatment of colon cancer [2]. Manual detection depends on proficient endoscopists, and it takes a long time. Recent surveys have shown that more than 25% of polyps in patients undergoing colonoscopy are not detected [3]. The late diagnosis of missed polyps can lead to a low survival rate for colon cancer patients [4]

Methods

Results

Discussion

Conclusion