David R. Cheriton School of Computer Science assistant professor Edith Law has been awarded a grant through the CIHR-NSERC Collaborative Health Research Project (CHRP) program of $145,000 per year for three years.
Law's research on human-computer interaction will be applied to create a framework for hybrid machine and human computation for the accurate and scalable analysis of human clinical EEG recordings. This research will significantly impact the approximate 250,000 Canadians that suffer from epilepsy, sleep disorders, and other neurological diseases.
The creation of a framework for hybrid machine-human analysis of these complex EEG recordings will help support EEG specialists in analyzing and clarifying ambiguous cases. The framework will also help reduce the amount of time needed to review each record allowing specialists to spend more time on complex recordings where a major issue may be present. In addition, this framework will increase accuracy, cost efficiency, and capacity of analysis.
Several key innovative aspects of this research will amplify ongoing attempts at increasing the efficiency of EEG analysis, including supporting the Bhutan Epilepsy Project which has developed hardware to allow for EEG analysis in remote and rural regions of the world. This work will tackle the software and algorithmic aspects needed for a truly scalable and portable solution to the problem of EEG interpretation in these regions.
Law, a world expert in human computation and her research team will be responsible for developing the techniques and algorithms, as well as interfaces for non-expert annotations and training. Her team will be collaborating with co-investigators Andrew Lim, a board certified neurologist and assistant professor at the University of Toronto, and Joelle Pineau, an expert in machine learning and associate professor at McGill University.