Medical images such as 3D computerized tomography (CT) scans and pathology images, have hundreds of millions or billions of voxels/pixels. It is infeasible to train CNN models directly on such high resolution images, because neural activations of a single image do not fit in the memory of a single GPU/TPU, and naive data and model parallelism approaches do not work. Existing image analysis approaches alleviate this problem by cropping or down-sampling input images, which leads to complicated implementation and sub-optimal performance due to information loss. In this paper, we implement spatial partitioning, which internally distributes the input and output of convolutional layers across GPUs/TPUs. Our implementation is based on the Mesh-TensorFlow framework and the computation distribution is transparent to end users. With this technique, we train a 3D Unet on up to 512 by 512 by 512 resolution data. To the best of our knowledge, this is the first work for handling such high resolution images end-to-end.
@article{arxiv.1909.03108,
title = {High Resolution Medical Image Analysis with Spatial Partitioning},
author = {Le Hou and Youlong Cheng and Noam Shazeer and Niki Parmar and Yeqing Li and Panagiotis Korfiatis and Travis M. Drucker and Daniel J. Blezek and Xiaodan Song},
journal= {arXiv preprint arXiv:1909.03108},
year = {2019}
}