# Holistic Decomposition Convolution for Effective Semantic Segmentation of 3D MR Images

## TL;DR

This paper introduces Holistic Decomposition Convolution (HDC), a novel method that enhances 3D medical image segmentation by efficiently capturing larger context without excessive memory use, demonstrated on MRI datasets.

## Contribution

The paper presents HDC, a new convolution technique that reduces data size while preserving information, enabling better 3D segmentation with limited GPU memory.

## Key findings

- HDC improves segmentation accuracy on MRI datasets.
- HDC reduces computational load and memory usage.
- HDC effectively captures larger spatial context.

## Abstract

Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance in many different 2D medical image analysis tasks. In clinical practice, however, a large part of the medical imaging data available is in 3D. This has motivated the development of 3D CNNs for volumetric image segmentation in order to benefit from more spatial context. Due to GPU memory restrictions caused by moving to fully 3D, state-of-the-art methods depend on subvolume/patch processing and the size of the input patch is usually small, limiting the incorporation of larger context information for a better performance. In this paper, we propose a novel Holistic Decomposition Convolution (HDC), for an effective and efficient semantic segmentation of volumetric images. HDC consists of a periodic down-shuffling operation followed by a conventional 3D convolution. HDC has the advantage of significantly reducing the size of the data for sub-sequential processing while using all the information available in the input irrespective of the down-shuffling factors. Results obtained from comprehensive experiments conducted on hip T1 MR images and intervertebral disc T2 MR images demonstrate the efficacy of the present approach.