Machine learning and Artificial Intelligence
Machine learning is a field with tremendous progress in the last decade, especially with deep neural networks trained on large datasets. Research in machine learning at CEC encompasses both the development of new algorithms, and a wide range of mainly medical applications.
Our research on new algorithms is focused on neural networks and deep learning, including methods for imputation of missing data, survival analysis, and generative models. When applying machine learning, our main focus is on biological and medical data, ranging from model building in order to understand biological processes, to support for specific medical decision. We conduct all applications of machine learning in close collaboration with various partners, primarily in the field of medicine. For example, this includes the analysis of proteomic and genomic data for early detection of cancer and autoimmune diseases, methods for identifying risk factors, and optimal matching between donor and recipient in heart transplantation.
Analysis of images from various imaging techniques is a broad subfield with many applications, from satellite images to high-resolution pathology images. We work, among other things, with hyperspectral images. The spectroscopic information adds an important dimension when images are segmented or otherwise transformed into numbers and predictions.