One Network to Segment Them All: A General, Lightweight System for Accurate 3D Medical Image Segmentation
This provides a more accessible and efficient solution for medical professionals and researchers by reducing the complexity and overfitting risks in 3D medical image segmentation.
The authors tackled the problem of specialized, task-specific medical image segmentation systems by proposing a general, lightweight framework that eliminates the need for model selection and task-specific modifications, achieving performance comparable to or better than specialized methods across multiple tasks, including ranking 5th and 6th in the 2018 Medical Segmentation Decathlon.
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.