Department of Oncology & Department of Physics
University of Alberta, Edmonton, Canada
Using Mathematical Measures of Network Complexity and Image Analysis for Cancer Diagnostics and Therapy Design
In this talk I will describe my group’s research into two aspects of cancer diagnostics and therapeutics that can benefit from a quantitative analysis of two sets of empirical data: protein-protein interaction networks and histopathology slides from cancer patients. In both cases, the objective is to find appropriate quantification measures that can aid in the automated analysis of the data.
Each cancer type has its own molecular signaling network and we have investigated algebraic and topological indices for network complexity for protein-protein interaction networks of human cancers. Then, the automatic analysis of histopathology tissue slides will be discussed. The main avenue of cancer diagnosis is through manual examination of an expert pathologist via a process, which is often subjective and error-ridden. Testing the fractal dimension as a viable image feature can lead to a high level of confidence in the resultant classification. Using machine learning, specifically the support vector machine (SVM) method, the F1 score for classification accuracy of the 40X slides was found to be 0.979. This method has been also applied to a smaller set of brain cancer slides (Glioblastoma Multiforme) and near perfect classification has been obtained so it is possible to discriminates not only between benign and malignant samples, but also between grade II and grade III gliomas.
Jack Tuszynski is a Fellow of the National Institute for Nanotechnology of Canada. He is Full Professor in Experimental Oncology in the Department of Oncology at the University of Alberta’s Cross Cancer Institute and in the Department of Physics. He received the M.Sc. in Physics from the University of Poznan (Poland) in 1980 and the PhD in Condensed Matter Physics from the University of Calgary in 1983. He had visiting professorships in Germany, Denmark, France, Belgium, Israel and China. He has published over 400 peer-reviewed papers. In 2005 he was appointed to the prestigious Allard Research Chair in Oncology at the University of Alberta.
Jack Tuszynski heads a multi-disciplinary team creating “designer drugs” for cancer chemotherapy using computational biophysics methods. The goal of Tuszynski’s computational biophysics work is to create optimized drugs that would target cancerous cells with minimal side-effects to the healthy cells.