Our laboratory focuses on research in the novel field of computational pathology. We develop and apply quantitative methods for the analysis of digital microscopy slides and relate the resulting statistical descriptors to patient outcomes. Specifically, we use supervised and unsupervised machine-learning algorithms and methods from computer vision and biostatistics to extract information from terabytes of data generated by modern pathology departments to gain insight into research questions such as how genetic mutations influence tissue morphology or which combinations of immunohistochemical markers predict patient survival. In addition to these research questions, we aim to improve clinical practice in pathology by developing intelligent decision-support systems that not only automate cumbersome or repetitive tasks but also lead to more-objective and reproducible results. The overarching goal is to lead the way in transforming pathology from a qualitative to a quantitative science.
Thomas J. Fuchs, PhD
Research FocusData scientist Thomas Fuchs develops and applies advanced machine learning and computer vision techniques for tackling large-scale computational pathology challenges in cancer research and clinical practice.
- Fuchs, Thomas J., and Joachim M. Buhmann. “Computational pathology: Challenges and promises for tissue analysis.” Computerized Medical Imaging and Graphics 35.7 (2011): 515-530.
- Fuchs, Thomas J., Peter Wild, Holger Moch, and Joachim Buhmann. “Computational pathology analysis of tissue microarrays predicts survival of renal clear cell carcinoma patients.” Medical Image Computing and Computer-Assisted Intervention–MICCAI 2008 (2008): 1-8.