Leveraging Image Complexity in Macro-Level Neural Network Design for Medical Image Segmentation

TL;DR

This paper explores how image complexity influences the design of neural networks for medical image segmentation, proposing guidelines for balancing performance and computational efficiency based on dataset complexity.

Contribution

It introduces a framework using image complexity measures to inform macro-level neural network design choices in medical segmentation tasks.

Findings

Median frequency best predicts suitable downsampling and network depth.

High-complexity images benefit from shallow networks on original images.

Low-complexity images perform better with deep networks on downsampled images.

Abstract

Recent progress in encoder-decoder neural network architecture design has led to significant performance improvements in a wide range of medical image segmentation tasks. However, state-of-the-art networks for a given task may be too computationally demanding to run on affordable hardware, and thus users often resort to practical workarounds by modifying various macro-level design aspects. Two common examples are downsampling of the input images and reducing the network depth to meet computer memory constraints. In this paper we investigate the effects of these changes on segmentation performance and show that image complexity can be used as a guideline in choosing what is best for a given dataset. We consider four statistical measures to quantify image complexity and evaluate their suitability on ten different public datasets. For the purpose of our experiments we also propose two new encoder-decoder architectures representing shallow and deep networks that are more memory efficient than currently popular networks. Our results suggest that median frequency is the best complexity measure in deciding about an acceptable input downsampling factor and network depth. For high-complexity datasets, a shallow network running on the original images may yield better segmentation results than a deep network running on downsampled images, whereas the opposite may be the case for low-complexity images.