Dr, Shehroz Khan holds a PhD degree from the University of Waterloo, Canada in Computer Science with specialization in Machine Learning. His PhD research is one of the first works to formulate fall detection as an anomaly detection problem. Prior to joining academia, Dr. Khan worked for around 10 years in India and Ireland in various scientific and research roles in the government and corporate sectors. His research program is funded through NSERC, CIHR, AGEWELL, AMS, SSHRC, CABHI and UAE University. He has published 40 research papers in top international journals and conferences. He is the founder and organizer of the peer-reviewed International Workshop on Artificial Intelligence for ARIAL held in conjunction with top conferences in the field from 2017-19. He is an Associate Editor of the Journal of Rehabilitation and Assistive Technologies Engineering. He is a reviewer of more than 30 international journals and actively acts as a program committee member of major AI conferences, including Canadian AI, AAAI, IJCAI and ICMI. He has been a keynote speaker at the Technology Investment Conference, Bangkok, Thailand, 2019, 1st Shanghai International Geriatric Rehabilitation Forum, Shanghai, China, 2020 and International Conference on Gerontechnology, Hong Kong, 2020.
Dr. Khan's research focus is on the development of deep learning and machine learning algorithms within the realms of Aging, Rehabilitation and Intelligent Assisted Living (ARIAL). Dr. Khan's team works with different types of multimodal sensors, including wearable devices, computer vision, ambient/environmental sensors. His studies are designed to work both in the hospital/long-term care and in community settings.
Some of the key projects the team is working on are: detecting agitation in people with dementia, fall detection, assessing social isolation, detecting the effects of cannabinoid oil in people with dementia, and providing virtual cardiac rehab. The team is also building a novel software platform to collect data from multimodal sensors and store them in a private cloud to facilitate federated learning and building personalized models in the cloud.