Kirti Sundar Sahu

My research aim is to study use of new generation big data (NexGen) for monitoring behavioural risk factors at the national, provincial, and household levels. In a pilot study we first collected data to assess the feasibility of smart thermostat data to measure average duration of sleep, physical activity, and sedentary behaviour in the Canadian population. In this study, we compared data from Fitbit wearables with ecobee smart thermostats and found a strong positive correlation among them. This idea won the Healthy Behaviour Data Challenge in 2018 organized by the Public Health Agency of Canada. In a follow up study, we did several statistical analyses, and applied machine learning and deep learning models to use smart thermostat data for monitoring different health indicators. In the third study for my PhD, we are focusing on comparing behaviour changes such as time spent indoors versus outdoors and sleep patterns at the population level (national, provincial, and city) due to the Covid-19 pandemic in Canada. This work will help to inform and reinforce existing public health surveillance systems in Canada.