A team of researchers from the School of Computing are working closely with partners from MiiCare Ltd, East Kent Hospitals University Foundation Trust, and Bristol City Council, in order to investigate the feasibility of using smart digital assistants that can be installed in peoples homes or in care homes, to sense residents’ footsteps and to infer their potential risk of falling accidents.
The Innovate UK project, entitled ADAPTIVE, is led by Dr. Christos Efstratiou and Dr. Palaniappan Ramaswamy from the School of Computing.
Older individuals tend to have a higher risk of physical accidents or falls as part of their daily lives. The overall physical decline associated with aging can make such accidents the cause of serious complications with long term effects on an individual’s health and wellbeing. People living with dementia are more prone to falls than cognitive-intact people, falling twice as often and suffering fractures/muscle tone loss that result in high rates of morbidity, mortality and hospitalisation. Gait characteristics of an individual can be linked to potential risk of accidental falls. This project aims to investigate the feasibility of detecting changes in gait pattern that can be linked to higher risk of falling accidents, non intrusively through the use of sound.
The project relies on the use of an AI-based digital health companion that has been developed by MiiCare Ltd, named MONICA, capable of responding to voice commands and assisting older adults in their daily lives. The objective of the project is to utilise the acoustic sensing capabilities of MiiCare’s devices, in order to investigate and develop machine learning techniques that can (i) extract footstep sounds of older people, and people with dementia, (ii) identify gait pattern features through the acoustic signals, and (iii) estimate the level of risk of a falling accident based on their gait characteristics.
In order to achieve this objective, the project involves the deployment of more than 60 such devices within care homes in the UK. The project team has established partnerships with The Graham Care Group, and Avante Care & Support, who will be hosting the deployment. An important aspect of this effort is that the development of the acoustic gait analysis algorithms will be based on real-world data collected through the living environments of older adults and people with dementia.
The research team at the University of Kent is very excited with this opportunity, as this is one of the first attempts to utilise acoustic gait analysis for predicting risk of falling accidents, through deployment of such technology in the wild.
The successful outcome of this project has the potential to lead to the design of novel assistive living solutions that can assess the falling risk of older adults non intrusively, and raise alerts when an individual requires assistance.
Further information about the project can be found on MiiCare Ltd’s website, here: