High-grade brain cancer (glioblastoma) is a devastating disease that very few patients survive long-term. The average life expectancy is 15 months, and throughout therapy patients undergo serial MR imaging for monitoring tumor response. It is not well understood how heterogeneity at the cellular and molecular levels affects the macroscopic imaging characteristics of these tumors. The long-term goal of this project is to provide imaging tools and biomarker integration strategies for individualizing glioblastoma treatment. The overall objective is to combine radiographic imaging with histopathological samples (i.e., radio-pathomics) to create and validate predictive tools for accurately defining tumor margins and spatial molecular profiles. Our central hypothesis is that microscopic glioblastoma cytological features and spatially dependent molecular profiles are reliably detectable and quantifiable with macroscopic MR imaging.
Two specific aims will objectively test this hypothesis by first determining which microscopic tissue features contribute to distinct measurements with MR imaging, and second, determining the performance of machine learning algorithms for predictively mapping these heterogeneous histological features. !
This project will provide a detailed understanding of how brain tumors at the cellular level appear on macroscopic imaging by examining whole brain human samples and clinical MRI scans. We will develop computational algorithms trained with histology to recognize patterns in MRI scans (i.e. radiopathomics), which will negate the need for tissue. Patient care and clinical decision-making will improve dramatically as a complete picture of histological features underlying brain tumor imaging emerges from this proposal.
| Hurrell, Sarah L; McGarry, Sean D; Kaczmarowski, Amy et al. (2018) Optimized b-value selection for the discrimination of prostate cancer grades, including the cribriform pattern, using diffusion weighted imaging. J Med Imaging (Bellingham) 5:011004 |
| McGarry, Sean D; Hurrell, Sarah L; Iczkowski, Kenneth A et al. (2018) Radio-pathomic Maps of Epithelium and Lumen Density Predict the Location of High-Grade Prostate Cancer. Int J Radiat Oncol Biol Phys 101:1179-1187 |
