Principal investigators: Dr. Gregory Czarnota, Dr. Ali Sadeghi-Naini, Dr. William Tran and Dr. Alex Kiss
This project is designed to tackle some of the most difficult problems in cure-oriented cancer research today. A team of world-leading experts with complementary skills are collaborating on projects that feed into and will be enriched by this core technology, in order to open up new frontiers and transform the state of knowledge regarding the diagnosis, treatment, and fundamental understanding of cancer.
The biostatistical and computational core of this project provides support for therapy and imaging studies. The data being generated in the four projects, which will flow into the core for analysis, are multi-modal and multivariate in nature, with complex quantitative ultrasound and MRI-CEST methods being used to characterize cancer, predict tumour responses a priori, and monitor therapy in order to deliver the best possible care. Modern methods in statistics and machine learning represent the most robust techniques to navigate this data, characterize the best parameters to be used for each project, and identify synergies between the projects.
The biomedical computational core consists of personnel with computational, machine learning, statistical, and mathematical modelling expertise. Team members are experts in fields ranging from biomarker discovery and validation, to clinical biostatistics, personalized medicine, cancer imaging, biomarkers, and/or cancer genomics, in addition to computational modeling and machine learning. The core team will collaborate with program project investigators to enhance the quality of the research.
We are collaborating with investigators through all stages of the research program. Our collaboration spans the experimental design and data collection stages, to data analysis and interpretation—including statistical design, database development, quality control and computational support. All project teams work directly with one or more core computer science experts and biostatisticians, chosen such that the expertise, experience and interests of the statisticians match the needs of the projects. Data analyses conducted within the core unify therapy preparation and analytical methods, integrate animal models of various human cancers and imaging characterization, and centralize data processing protocols. Machine learning analyses identify synergies among the different types of project data and support the overall scientific goals of the program. Experimentally validated mathematical models identified through the core aid in the development and optimization of mechanism-based therapies to enhance tumour response, with the eventual goal of improving outcomes for cancer patients.