A vast amount of information associated with spinal patient diagnosis, treatment and follow-up are generated every day. The application of multiple imaging tools (e.g. CT and MRI) has offered a great tool to diagnose and track diseases but at the same time poses significant challenges to human perception. One such challenge is the huge amount of information (a single Spinal Magnetic Resonance Imaging (MRI) is about 300M) associated with each scan. Often, a typical diagnosis and treatment of one spinal patient would involve multiple spinal imaging modalities and multiple follow-ups.

The ability to analyze changes in those data over time (longitudinal analysis) of all those data would provide much more accurate diagnosis and prediction. Despite the available data, there is a lack of automated tools to analyze the data to support physicians’ diagnosis, prognosis, and treatment. There is an urgent need of semi and fully-automated intelligent tools to help physicians with their quantitative measurement, longitudinal analysis, and population studies. To provide a comprehensive solution for the above challenges, we propose to develop a new generation of computer aided spinal data analytics system to improve the efficiency and accuracy of spinal patient care. This will be the first comprehensive cloud-based spinal diagnosis and prediction system, which will handle multiple spinal imaging modalities to facilitate detection, segmentation, direct historical diagnostic parameter extractions and medical records analytics. The automated data analytics will allow healthcare providers to manage the ever-growing amount of spinal data more effectively and positively contribute to immediate and long-term Canadian healthcare.

Industry Partner(s):Victoria Hospital Imaging Associates

Academic Institution:Western University

Academic Researcher: Shuo Li

Focus Areas: Digital Media, Health

Platforms: Cloud