I am a PhD student at Empirical Inference Department, Max Planck Institute for Intelligent Systems and Werner Siemens Imaging Center, UKT, Tuebingen. My work focuses on the assessment of tumor heterogeneity using multi-modality (PET/MR) imaging for personalized cancer therapies. I am also very interested in applying statistical methods on the data originating from multi-modal and heterogeneous sources.
1. Assessment of tumor heterogeneity using PET-MRI and Machine Learning
Tumors exhibit widespread genetic and phenotypic heterogeneity. The complementary information provided by PET and MRI can be highly instrumental in probing the intra-tumoral diversity, potentially assisting radiologists in cancer therapy planning. For example, the quantitative analysis of [18F]-FDG PET can capture changes in tumor metabolism, and one may assess tumor cell proliferation based on [18F]-FLT uptake. Furthermore, non-invasive MRI biomarkers can be used as an indicator of cell density, perfusion, total blood volume, vascular density and average vessel diameter. Since tumor progression involves a complex interplay of several biological factors, combining these complementary modalities will provide additional insights for understanding the local tumor environment.
The examination of such intricate data, however, is beyond the capacity of trained radiologists and requires well defined statistical analyses. This project aims to model the complex relationship between different imaging parameters using machine learning algorithms. The workflow will not only automate the decision process, but will also provide a robust and repeatable quantification of the in-vivo imaging data.
Our goal is to understand the principles of Perception, Action and Learning in autonomous systems that successfully interact with complex environments and to use this understanding to design future systems