Abstract | ||
---|---|---|
Deep neural network models used for medical image segmentation are large because they are trained with high-resolution three-dimensional (3D) images. Graphics processing units (GPUs) are widely used to accelerate training. However, the memory on a GPU is not large enough to train the models. A popular approach to tackling this problem is the patch-based method, which divides a large image into small patches and trains the models with these small patches. However, this method degrades the segmentation quality if a target object spans multiple patches. In this paper, we propose a novel approach for 3D medical image segmentation that utilizes the data-swapping method, which swaps out intermediate data from GPU memory to CPU memory to enlarge the effective GPU memory size for training high-resolution 3D medical images without patching. We enhanced the existing data-swapping method by introducing swapping inside forward propagation and selective swapping of analysis path in order to train 3D U-Net effectively. We applied this approach to train 3D U-Net with full-size images of 192 x 192 x 192 voxels for a brain tumor dataset. Compared with the patch-based method for patches of 128 x 128 x 128 voxels, our approach improved the mean Dice score by 3.9 percentage points and 4.1 percentage points when detecting a whole tumor sub-region and a tumor core sub-region, respectively. The total training time was reduced from 164 h to 47 h, resulting in an acceleration of 3.53 times. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1007/978-3-030-32248-9_27 | Lecture Notes in Computer Science |
Keywords | DocType | Volume |
Deep learning,Image segmentation,3D U-Net,Data-swapping method | Conference | 11766 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
haruki imai | 1 | 0 | 1.35 |
Samuel Matzek | 2 | 0 | 0.34 |
Tung D. Le | 3 | 2 | 2.08 |
Yasushi Negishi | 4 | 36 | 6.36 |
Kawachiya, K. | 5 | 145 | 16.81 |