Skip to main content

Research Repository

Advanced Search

Fall Prediction in Social Housing

People Involved

Project Description

The cost of falls to the NHS is over £2bn/yr and 4 million bed days, and is set to rise. Albyn Housing Society, with Carbon Dynamic and NHS Highland has developed a technology-enabled house that will:-
• enable residents to remain in their homes for longer
• prevent hospital admission and enable early discharge
• provide a solution for people not wanting to enter a care home
• predict health related events and enable mitigation to enhance wellness.
Falls within the home are one of the main contributors to hospital admission and represents a frequent reason for people going into a care home. It is imperative that solutions are sought to prevent and detect falls, so an immediate response can be mounted.
We aim to use data derived from sensors within specially designed ‘Fit Homes’ and state-of-the-art data science to help predict falls. We will set up a system of sensors targeting at specific activities identified as pre-cursors to individuals being unsteady on their feet: Dehydration; changes in dietary patterns and taking prescribed medication; activity levels (sleep/movement around the house); and social interactions which may be an indirect measure of mental health.
This project will develop a case-based system based on patterns of activity and behaviours learned from the sensor data. This system will be delivered as a prototype application that takes inputs from unobtrusive sensors installed in the homes and analyses the data received in order to identify the timing, frequency and duration of the residents’ normal activities (getting up, preparing a meal, watching TV, etc). Changes in a ‘Fit Home’ resident’s own activity patterns over time can be used to detect deterioration in health; comparisons with other residents can highlight risk of falls. It is envisaged that this project is the precursor to more research and development in this sector.

In partnership with Albyn Housing Association

Type of Project Project
Project Acronym FITsense 1
Status Project Complete
Funder(s) The Data Lab
Value £97,133.00
Project Dates Sep 1, 2017 - Nov 30, 2018

You might also like

A decision support system for self management of low back pain Jan 1, 2016 - Mar 31, 2021
Low back pain (LBP) is the most significant contributor to disability in Europe. Most patients have non-specific LBP i.e., pain that cannot reliably be attributed to a specific disease/ pathology. LBP is the fourth most common diagnosis seen in prima... Read More