Data augmentation using learned transformations for one-shot medical image segmentation

TL;DR

This paper introduces an automated data augmentation technique that learns complex transformations from unlabeled medical images to synthesize additional labeled data, significantly improving one-shot MRI brain scan segmentation accuracy.

Contribution

The authors propose a semi-supervised data augmentation method that learns transformation models from unlabeled images to generate realistic labeled examples for improved segmentation.

Findings

Enhanced segmentation accuracy with synthesized data

Effective use of unlabeled scans for transformation learning

Significant improvement over existing one-shot segmentation methods

Abstract

Image segmentation is an important task in many medical applications. Methods based on convolutional neural networks attain state-of-the-art accuracy; however, they typically rely on supervised training with large labeled datasets. Labeling medical images requires significant expertise and time, and typical hand-tuned approaches for data augmentation fail to capture the complex variations in such images. We present an automated data augmentation method for synthesizing labeled medical images. We demonstrate our method on the task of segmenting magnetic resonance imaging (MRI) brain scans. Our method requires only a single segmented scan, and leverages other unlabeled scans in a semi-supervised approach. We learn a model of transformations from the images, and use the model along with the labeled example to synthesize additional labeled examples. Each transformation is comprised of a spatial deformation field and an intensity change, enabling the synthesis of complex effects such as variations in anatomy and image acquisition procedures. We show that training a supervised segmenter with these new examples provides significant improvements over state-of-the-art methods for one-shot biomedical image segmentation. Our code is available at https://github.com/xamyzhao/brainstorm.