Scatter Reduction By Exploiting Behaviour of Convolutional Neural Networks in Frequency Domain

Abstract

In X-ray imaging, scattered radiation can produce a number of artifacts that greatly undermine the image quality. There are hardware solutions, such as anti-scatter grids. However, they are costly. A software-based solution is a better option because it is cheaper and can achieve a higher scatter reduction. Most of the current software-based approaches are model-based. The main issues with them are the lack of flexibility, expressivity, and the requirement of a model. In consideration of this, we decided to apply Convolutional Neural Networks (CNNs), since they do not have any of the previously mentioned issues. In our approach we split the image into three frequency bands: low, high low and high high and process each of them separately with a CNN. Then, we downsample the low frequency band and upsample the high frequency band, so that the frequency is increased and decreased respectively. Finally, we train three CNNs with each of the components and put them back together to have the reconstruction of the image. We demonstrate theoretically that doing this leads to better results, and provide comprehensive empirical evidence of the capability of our algorithm for doing scatter correction.