In new ways in the field of medical image analysis supported by artificial intelligence
For more than a decade data mining, including machine learning has been an active topic of research and teaching at the Mathematical Institute at the Eötvös Loránd University with the objective of keeping up a good balance of mathematical and engineering approach. In recent years it was a good base to immerse ourselves in the world of Deep Learning, the latest breakthrough tool for Artificial Intelligence. A promising direction of research is to bridge the gap between mathematical theory and machine learning practice by exploiting newly discovered deep connections between fundamental results related to the study of large networks and the more applied domain of machine learning. As our competencies include state-of-the-art image retrieval techniques and recent network models: residual networks, variational autoencoders, and generative adversarial networks (GAN), we started to use them for Medical Image Analysis. Our approach here is twofold: first, to develop solutions to pressing problems within the field such as inconsistent inter-rater reliability and the declining amount of practicing radiologists by introducing deep learning backed automation in the diagnostic pipeline, secondly to improve upon existing state-of-the-art methods by studying the application of GANs to medical imaging data. Currently, we are researching structure-correcting adversarial networks on X-ray segmentation tasks, as well as super-resolution methods on computed tomography scans.
Chest X-ray is the most common test among medical imaging modalities. It is applied for detection and differentiation of, among others, lung cancer, tuberculosis, and pneumonia, the last with importance due to the COVID-19 disease. In order to identify lung nodules, lung segmentation of chest X-rays is essential, and this step is vital in other diagnostic pipelines as well, such as calculating the cardiothoracic ratio, which is the primary indicator of cardiomegaly. For this reason, a robust algorithm to perform this otherwise arduous segmentation task is much desired in the field of medical imaging. Semantic segmentation aims to solve the challenging problem of assigning a pre-defined class to each pixel of the image. This task requires a high level of visual understanding, in which state-of-the-art performance is attained by methods utilizing Fully Convolutional Networks (FCN). Our novel deep learning approach for lung segmentation uses attention gated FCNs in conjunction with an adversarial critic model, where we use Focal Tversky Loss for teaching the network. A further important element of the developed model was the preprocessing of the images. X-rays are grayscale images with typically low contrast, which makes their analysis a difficult task. This obstacle might be overcome by using some sort of histogram equalization technique. Our choice was to use Contrast Limited Adaptive Histogram Equalization (CLAHE). Applying CLAHE to an X-ray image has visually appealing results. As our experiments displayed, it does not merely help human vision, but also neural networks.
For training- and validation data, we used the JSRT dataset, as well as the Montgomery- and Shenzhen dataset, all of which are public datasets of chest X-rays with available organ segmentation masks reviewed by expert radiologists.
More details about this work can be read in the paper Gaál, G., Maga, B., & Lukács, A.: Attention U-Net based adversarial architectures for chest X-ray lung segmentation, https://arxiv.org/abs/2003.10304.