Colorectal Cancer Segmentation using Atrous Convolution and Residual Enhanced UNet

TL;DR

This paper introduces AtResUNet, a CNN-based model with atrous convolutions and residual connections, achieving effective colorectal cancer segmentation with reduced computation on high-resolution images.

Contribution

The paper proposes a novel CNN architecture, AtResUNet, combining atrous convolutions and residual connections for improved colorectal tumor segmentation.

Findings

Achieved a Dice Coefficient of 0.748 on the DigestPath 2019 dataset.

Utilized an efficient patch-based training and inference approach.

Demonstrated competitive performance in biomedical image segmentation.

Abstract

Colorectal cancer is a leading cause of death worldwide. However, early diagnosis dramatically increases the chances of survival, for which it is crucial to identify the tumor in the body. Since its imaging uses high-resolution techniques, annotating the tumor is time-consuming and requires particular expertise. Lately, methods built upon Convolutional Neural Networks(CNNs) have proven to be at par, if not better in many biomedical segmentation tasks. For the task at hand, we propose another CNN-based approach, which uses atrous convolutions and residual connections besides the conventional filters. The training and inference were made using an efficient patch-based approach, which significantly reduced unnecessary computations. The proposed AtResUNet was trained on the DigestPath 2019 Challenge dataset for colorectal cancer segmentation with results having a Dice Coefficient of 0.748.