Abstract

The most recent research is showing the importance and suitability of neural networks for medical image processing tasks. Nonetheless, their efficiency in segmentation tasks is greatly dependent on the amount of training data. To overcome issues of using small datasets, various data augmentation techniques have been developed. We propose a convolutional neural network approach for the whole heart segmentation that is based on the 3D U-Net architecture and incorporates principle component analysis as an additional data augmentation technique. The network is trained end-to-end, so no pre-trained network is required. Evaluation of the proposed approach is performed on 20 3D CT images from MICCAI 2017 Multi-Modality Whole Heart Segmentation Challenge dataset, divided into 15 training and 5 validation images. Final segmentation results show a high Dice coefficient overlap to ground truth, indicating that the proposed approach is competitive to state-of-the-art. Additionally, we provide the experiments of the influence of different learning rates that shows the optimal learning rate of 0.005 to give best segmentation results.

Keywords

CT, data augmentation, medical image segmentation, neural networks, volumetric segmentation, whole heart segmentation