AIAI Seminar - 14 February 2022 - Talk by Longfei Chen Talk by Longfei Chen Title: Helping elderly singles live longer in their own homes by monitoring daily behaviours Abstract: As the population ages fast, the proportion of elderly people living alone is increasing globally. According to the Office for National Statistics in 2019, more than 3 million people over the age of 70 lived alone in the UK. Lonely elderly are expected to have higher healthcare needs, however, more than 90% of elderlies express a strong desire to remain independent. They mostly prefer to live in their own home rather than stay in a nursing home. This could be a great potential when it comes to home vision AI care. Home vision AI systems can provide 24/7 autonomous anonymous monitoring, provide faster and more accurate analysis in a user-centralized and cost-effective manner, provide alerts when dangerous situations are detected, save lives, and save labour. We present applications of real-time anonymous visual monitoring of aging people in their homes. Specific events in two daily scenarios are detected: a) unexpected long-term room occupation, b) lack of motion in a certain home place. These events can relate to dangerous medical situations of the elderly, such as a long bathroom entry time caused by falls; or lack of movement when staying on a favourite couch, caused by physical capacity deterioration. We are also collecting data observing people eating. A light-weighted camera system is set in the home to monitor the elderly's daily behaviours, containing an RGB-D camera (Intel Realsense) and a small computer processor (Jetson Nano), providing a long-term and anonymous way of the data capture. In order to ensure the anonymity of the subject, the video data is discarded after real-time processing, no personal information is stored. Human detection and motion detection methods mainly based on depth, colour, and morphology are designed to recognize human and activity events. The results show that the system has good accuracy, sensitivity, and robustness in different environmental settings, enabling long-term anonymous monitoring of the elderlies in later life to support them, and reveal individual characteristics such as motion habits and health conditions. The project is undertaken in the Advanced Care Research Centre (ACRC) at the University of Edinburgh, a cross-university centre focussing on the technological, medical, and sociological issues needed to support the day-to-day living of aging people. The vision is high‐quality data‐driven, personalized, and affordable care that supports the independence, dignity, and quality‐of‐life of people in later life living in their own homes and in supported care environments. This project will be part of the ACRC's New Technologies of Care research program. This project potentially benefits the elderly and their support communities by reducing the amount of human monitoring, and by detecting situations of interest. Feb 14 2022 14.00 - 15.00 AIAI Seminar - 14 February 2022 - Talk by Longfei Chen AIAI Seminar talk hosted by Longfei Chen Online
AIAI Seminar - 14 February 2022 - Talk by Longfei Chen Talk by Longfei Chen Title: Helping elderly singles live longer in their own homes by monitoring daily behaviours Abstract: As the population ages fast, the proportion of elderly people living alone is increasing globally. According to the Office for National Statistics in 2019, more than 3 million people over the age of 70 lived alone in the UK. Lonely elderly are expected to have higher healthcare needs, however, more than 90% of elderlies express a strong desire to remain independent. They mostly prefer to live in their own home rather than stay in a nursing home. This could be a great potential when it comes to home vision AI care. Home vision AI systems can provide 24/7 autonomous anonymous monitoring, provide faster and more accurate analysis in a user-centralized and cost-effective manner, provide alerts when dangerous situations are detected, save lives, and save labour. We present applications of real-time anonymous visual monitoring of aging people in their homes. Specific events in two daily scenarios are detected: a) unexpected long-term room occupation, b) lack of motion in a certain home place. These events can relate to dangerous medical situations of the elderly, such as a long bathroom entry time caused by falls; or lack of movement when staying on a favourite couch, caused by physical capacity deterioration. We are also collecting data observing people eating. A light-weighted camera system is set in the home to monitor the elderly's daily behaviours, containing an RGB-D camera (Intel Realsense) and a small computer processor (Jetson Nano), providing a long-term and anonymous way of the data capture. In order to ensure the anonymity of the subject, the video data is discarded after real-time processing, no personal information is stored. Human detection and motion detection methods mainly based on depth, colour, and morphology are designed to recognize human and activity events. The results show that the system has good accuracy, sensitivity, and robustness in different environmental settings, enabling long-term anonymous monitoring of the elderlies in later life to support them, and reveal individual characteristics such as motion habits and health conditions. The project is undertaken in the Advanced Care Research Centre (ACRC) at the University of Edinburgh, a cross-university centre focussing on the technological, medical, and sociological issues needed to support the day-to-day living of aging people. The vision is high‐quality data‐driven, personalized, and affordable care that supports the independence, dignity, and quality‐of‐life of people in later life living in their own homes and in supported care environments. This project will be part of the ACRC's New Technologies of Care research program. This project potentially benefits the elderly and their support communities by reducing the amount of human monitoring, and by detecting situations of interest. Feb 14 2022 14.00 - 15.00 AIAI Seminar - 14 February 2022 - Talk by Longfei Chen AIAI Seminar talk hosted by Longfei Chen Online
Feb 14 2022 14.00 - 15.00 AIAI Seminar - 14 February 2022 - Talk by Longfei Chen AIAI Seminar talk hosted by Longfei Chen