One Network to Segment Them All: A General, Lightweight System for Accurate 3D Medical Image Segmentation

TL;DR

This paper introduces a universal, lightweight 3D medical image segmentation system that performs competitively across various tasks without task-specific tuning, simplifying deployment and reducing overfitting.

Contribution

The authors present a fixed-topology, hyperparameter-agnostic deep learning framework that generalizes well across multiple medical segmentation tasks without customization.

Findings

Outperforms or matches specialized methods on three segmentation tasks.

Achieved top rankings in the 2018 Medical Segmentation Decathlon.

Utilizes multi-planar data augmentation with a 2D U-Net architecture.

Abstract

Many recent medical segmentation systems rely on powerful deep learning models to solve highly specific tasks. To maximize performance, it is standard practice to evaluate numerous pipelines with varying model topologies, optimization parameters, pre- & postprocessing steps, and even model cascades. It is often not clear how the resulting pipeline transfers to different tasks. We propose a simple and thoroughly evaluated deep learning framework for segmentation of arbitrary medical image volumes. The system requires no task-specific information, no human interaction and is based on a fixed model topology and a fixed hyperparameter set, eliminating the process of model selection and its inherent tendency to cause method-level over-fitting. The system is available in open source and does not require deep learning expertise to use. Without task-specific modifications, the system performed better than or similar to highly specialized deep learning methods across 3 separate segmentation tasks. In addition, it ranked 5-th and 6-th in the first and second round of the 2018 Medical Segmentation Decathlon comprising another 10 tasks. The system relies on multi-planar data augmentation which facilitates the application of a single 2D architecture based on the familiar U-Net. Multi-planar training combines the parameter efficiency of a 2D fully convolutional neural network with a systematic train- and test-time augmentation scheme, which allows the 2D model to learn a representation of the 3D image volume that fosters generalization.