MDA-Net: Multi-Dimensional Attention-Based Neural Network for 3D Image Segmentation

TL;DR

MDA-Net introduces a multi-dimensional attention mechanism within a U-Net framework to enhance 3D image segmentation accuracy while maintaining low computational costs, validated on medical imaging datasets.

Contribution

The paper presents a novel multi-dimensional attention network that effectively combines slice-wise, spatial, and channel-wise attention for efficient 3D image segmentation.

Findings

Achieves higher segmentation accuracy than existing methods.

Reduces computational complexity and memory usage.

Demonstrates consistent improvements on MICCAI iSeg and IBSR datasets.

Abstract

Segmenting an entire 3D image often has high computational complexity and requires large memory consumption; by contrast, performing volumetric segmentation in a slice-by-slice manner is efficient but does not fully leverage the 3D data. To address this challenge, we propose a multi-dimensional attention network (MDA-Net) to efficiently integrate slice-wise, spatial, and channel-wise attention into a U-Net based network, which results in high segmentation accuracy with a low computational cost. We evaluate our model on the MICCAI iSeg and IBSR datasets, and the experimental results demonstrate consistent improvements over existing methods.