Project (Michael Altenbuchinger)

Bioinformatic Research in precise medicine

Federated Learning Methods for Lymphoma Pathology

Lymphoma is a heterogenous disease composed of multiple morphological and molecular subtypes. To determine the optimal treatment, these subtypes need to be distinguished/diagnosed. The state of the art in lymphoma diagnosis is to combine visual inspection of tissue with assessment of molecular data and merging these different types of data to identify the exact diagnosis. We want to bring both data sources closer together. However, a big obstacle on this route is the lack of computational approaches that allow us to automatize the joint modeling of these data and, moreover, state-of-the-art histology image analysis via deep learning requires huge amount of data. Corresponding data files are large and as health data underlie safety regulations complicating their exchange. In this project, the goal is to establish a federated learning platform that integrates high-throughput molecular and high-definition histology image data. Sensitive data will remain with the local pathologists and do not have to be send to other sites; only model parameters will be shared. On this route, methodology for distributed learning should be established that can deal with data heterogeneity among different data generating pathology labs.

Homepage Research Group

For more information see for instance:

  • S Schrod, A Schäfer, S Solbrig, R Lohmayer, W Gronwald, P J Oefner, T Beißbarth, R Spang, H U Zacharias, M Altenbuchinger, BITES: balanced individual treatment effect for survival data, Bioinformatics, Volume 38, Issue Supplement_1, July 2022, Pages i60–i67,