Medical imaging

computer vision, deep learning, computed tomography, X-Ray

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and advances in convolutional neural networks. Digitalization in medicine and introduction of electronic health records significantly contributed to the availability of the large amounts of medical imaging data. While medical imaging datasets have been growing in size, a challenge for supervised machine learning algorithms that  is frequently mentioned is the lack of annotated data. As a result, unsupervised or self-supervised machine learning methods has to be used to solve learning problems on unlabeled data. We focus on the proposal of the new unsupervised and semi-supervised methods and algorithms for medical images, based on deep neural networks and transfer learning.

Related publiactions:

M. Gazda, J. Plavka, J. Gazda and P. Drotár, "Self-Supervised Deep Convolutional Neural Network for Chest X-Ray Classification," in IEEE Access, vol. 9, pp. 151972-151982, 2021, doi: 10.1109/ACCESS.2021.3125324.

Gazda, M., Bugata, P., Gazda, J., Hubacek, D., Hresko, D.J., Drotar, P. (2022). Mixup Augmentation for Kidney and Kidney Tumor Segmentation. In: Heller, N., Isensee, F., Trofimova, D., Tejpaul, R., Papanikolopoulos, N., Weight, C. (eds) Kidney and Kidney Tumor Segmentation. KiTS 2021. Lecture Notes in Computer Science, vol 13168. Springer, Cham.