LEARNING-BASED ACTIVITY RECOGNITION IN SENSOR NETWORKS FOR OLDER ADULTS 

AUTHOR(S) & CREDENTIALS: Hui Chen, Ph.D. Computer Science, Sherbrooke University and Cassy Hemphill, Communications and Engagement Coordinator  

AFFILIATED INSTITUTION(S): University of Sherbrooke and AGE-WELL National Innovation Hub: APPTA, AGE-WELL NCE (Network of Centres of Excellence). ACKNOWLEDGEMENTS: This research was supported by AGE-WELL NCE Inc, MEDTEQ (INNOVATION FOR HEALTH) and Réseau Québécois de Recherche sur le Vieillissement (RQRV) 

Hui Chen, who is currently pursuing a Ph.D. in Computer Science at the University of Sherbrooke, found her inspiration for her research in the struggles faced by her grandmother and other older adults in performing basic daily activities (ADLs). Witnessing their challenges ignited a passion within Chen to explore ways to enhance the safety and independence of older adults living alone. With this inspiration, Chen worked to create a system capable of recognizing and understanding the daily activities of older adults. Her aim was not just to monitor but to adapt the system to the unique needs and routines of each individual, in hopes of fostering a safer living environment while preserving their autonomy. 

Her research centered on utilizing ambient sensors to detect and provide insights into changes in environmental conditions within senior living communities. Using the sensors, Chen attempted to recognize activities and routines of the older adults and provide cognitive assistance and telemonitoring for those with frailties or cognitive deficits.  

This led Chen to identify a critical gap. Recognizing daily activities in a real-world setting posed challenges due to the lack of ground truth data (i.e., real world data). To overcome this hurdle, Chen and her colleagues developed the LOADA annotation application and inserted them in real-world monitoring and assistive systems within senior communities, allowing older adults to record their daily activities.   

While performing her research, Chen recognized the importance of ensuring autonomy within the home and lives of older adults. Chen emphasized the need for updated legal frameworks to address emerging challenges in artificial intelligence (AI) ethics, data privacy, and security, while also ensuring equitable access to intelligent monitoring systems for all older adults. She also advocated for a balanced approach, ensuring policies consider privacy, accessibility, and caregiver support. 

She believed that research in this area was critical, due to the global population aging rapidly, leading to an increasing need to support older adults in living independently and safely. Chen stated, “This technology facilitates the early detection of potential health issues which in turn can help enable timely interventions, and provide personalized support, thereby extending the period in which they can live safely at home. Furthermore, this research could offer caregivers and healthcare professionals valuable insights into the daily routines and potential needs of older adults. Therefore, the development of more accurate and efficient algorithms for activity recognition could lead to an increase in the number of older adults able to live independently and safely within their own communities.”