AI tool could help detect early dementia, Scottish researchers say

By detecting erratic behaviour, the device could alert of a  medical emergency before it's too late

A Scottish innovation centre has developed an AI-powered device which could help detect early dementia.

Through Internet of Things technology, the tool can monitor the use of electric items at home, detecting unusual behaviour which might indicate a medical emergency.

Developed by CENSIS – Scotland’s innovation centre for sensing, imaging, and IoT technologies – the device is aimed at making independent living safer for older and vulnerable people who live alone.

Stephen Milne, director of strategic projects at CENSIS, said: “The system learns the typical activity of the individual living in the household and then spots any erratic behaviour, helping to identify when they may have issues. These could be one-off events, like a fall, and with further research, the system may be able to track changes over a longer time period that may indicate gradual, and more difficult-to-spot health issues, such as the onset of a condition such as dementia.”

The device is linked to a smart or conventional electric meter, so it can detect different high-power electrical items within a home, such as kettles or electric showers. It later uses a machine learning algorithm to analyse the power signatures coming from th house. Throughout the project Edinburgh University researchers created a library of these signatures, tagging each item to identify when these are turned on, allowing the system to detect any anomalies.

If an electrical appliance is not on at its usual time, the tool will first send an automated text message to the user. Then, if no response is received, an alert is sent to their nominated contact.

Milne continued: “While there are other technologies related to monitoring activity, this is the first full-service deployment that has been implemented through passively monitoring a property’s smart meter system. The device can also pick out each item being monitored, making it much more likely to spot any anomalies, and is barely noticeable for the householder.”

A hub processing algorithm manages the data, allowing information to be processed at the property rather than at a central point. Information is then kept in an industry-leading personal data store created by tech firm Mydex, which allows users to fully control who has access to the data.

The device was tested as part of Blackwood Homes and Care’s Peoplehood project, which aims to develop a future-proof model for independent living for its residents.

Almost 20 households across Glasgow, Dundee and Moray tried the tool, which proved lifesaving for Evie from Buckie.

In late November, Evie passed out on her bed after feeling unwell. As she didn’t use any kitchen appliance at dinner time, an SMS alert was sent. As she was unconscious no response was received, which triggered an automatic interactive voice recording phone call with her nominated responder. She then regained consciousness with her responder, and after a check up at hospital she was told she has suffered an angina attack.

Evie said, if it wasn’t for the alert system “she could’ve been there all night”, which could have led to a more serious outcome. She has since made a full recovery.

Carebuilders, a dutch firm that focuses on developing person-centred technology enabled care, also helped develop the device. 

CENSIS is now looking to commercialise the product and deploy it at “a much bigger scale”, potentially through long-term partnerships, Milne added.