BIOMEDICAL IMAGES ENCODE TRACES OF AN UNDERLYING PHENOTYPE, WE ARE LEARNING TO TRANSLATE THEM INTO COMPUTATIONAL BIOMARKERS.

Kyle Lafata, PhD

Assistant Professor of Radiology and Radiation Oncology

Email: kyle.lafata@duke.edu | Phone: 978-491-8730 | Office: Rm. 1531, 201 Trent Dr., Duke University, Durham, NC 27705

radiology.duke.edu | radonc.duke.edu | medschool.duke.edu

Research Interests

Our research interests are focused on the discovery of multiscale imaging biomarkers. We develop novel mathematical methods and computational tools to identify quantitative markers of disease otherwise dormant on biomedical images. By interrogating multiscale information from both radiological images (i.e., radiomics) and digital pathology images (i.e., pathomics), our research aims to capture the appearance and behavior of disease across different spatial, temporal, and functional domains. Methodologically, our team incorporates various computational and mathematical techniques into our work, including image processing, computer vision, statistical pattern recognition, and the applied analysis of stochastic differential equations.