In this seminar, our SmallData Associated Researchers, along with a guest speaker, will give 30 minute presentations of their work on small data and explore how it connects with other projects across the Collaborative Research Center. Each seminar will feature an open Q&A session.
Engineering immune cells for improved antitumor functionality
Cellular immunotherapy, including so-called engineered CAR T or CAR NK cell therapy, has shown remarkable clinical efficacy against advanced B cell malignancies. In this context, natural killer (NK) cells represent a possible “off-the-shelf” therapeutic concept, since these cells can be generated from allogeneic donors. Recent approaches have successfully enhanced the specificity and cytotoxicity of NK cells towards tumor-specific or associated target antigens by genetic engineering to express a chimeric antigen receptor (CAR). However, many patients do not benefit from this treatment either because they do not respond or subsequently relapse. It is therefore of paramount importance to understand the mechanisms of tumor resistance to specific immune cell subsets. Understanding what differentiates effective from ineffective NK cell – tumor cell interactions by high-resolution data will significantly contribute improving future clinical immune cell therapy.
Evelyn Ullrich (guest speaker)
Department for Pediatrics Experimental Immunology & Cell Therapy, Goethe University Frankfurt
Robust and Trustworthy AI for Cell Microscopy
Microscopy reveals the richness of cellular morphology and behavior, but manual interpretation is labor-intensive and often inconsistent. Artificial intelligence offers powerful tools to automate this process; however, data-driven methods must be robust to variability across devices, limited data from small studies, and distributional shifts between cohorts to be applicable in this domain. Our work on robust AI for cell microscopy enables the automatic detection of cell types and states as well as the analysis of dynamic single-cell behavior through tracking. Beyond automation, we employ clustering-based approaches to identify key subpopulations and explainable AI methods to discover novel digital biomarkers with clinical relevance.
Maria Kalweit (SmallData Associated Researcher)
Department of Computer Science, Faculty of Engineering, University of Freiburg
Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center –
University of Freiburg
