Combining Pyramid Pooling and Attention Mechanism for Pelvic MR Image Semantic Segmentaion

TL;DR

This paper introduces APNet, a novel neural network that combines pyramid pooling and attention mechanisms to improve automatic segmentation of pelvic MR images, addressing challenges of organ size variability, similar appearances, and limited data.

Contribution

The paper presents a new attention-pyramid network (APNet) and a data augmentation technique tailored for pelvic MR image segmentation, outperforming existing methods.

Findings

APNet achieves superior segmentation accuracy on a new pelvic MR dataset.

The data augmentation method significantly improves model performance.

APNet effectively captures both local and global contexts for accurate segmentation.

Abstract

One of the time-consuming routine work for a radiologist is to discern anatomical structures from tomographic images. For assisting radiologists, this paper develops an automatic segmentation method for pelvic magnetic resonance (MR) images. The task has three major challenges 1) A pelvic organ can have various sizes and shapes depending on the axial image, which requires local contexts to segment correctly. 2) Different organs often have quite similar appearance in MR images, which requires global context to segment. 3) The number of available annotated images are very small to use the latest segmentation algorithms. To address the challenges, we propose a novel convolutional neural network called Attention-Pyramid network (APNet) that effectively exploits both local and global contexts, in addition to a data-augmentation technique that is particularly effective for MR images. In order to evaluate our method, we construct fine-grained (50 pelvic organs) MR image segmentation dataset, and experimentally confirm the superior performance of our techniques over the state-of-the-art image segmentation methods.