Researchers at the University of Sheffield have been developing CognoSpeak, a tool that can help to identify early signs of dementia and Alzheimer’s Disease. The system uses a virtual agent displayed on a screen to engage a patient in a conversation.
As part of the dialogue, the tool asks memory-probing questions inspired by those used in outpatient consultations and conducts cognitive tests, such as picture descriptions and verbal fluency tests. It then uses AI and speech technology to analyse language and speech patterns to look for signs of memory disorders. The tool could be revolutionary in the global healthcare sector, helping with early detection of brain diseases.
AI in healthcare: Breaking barriers with algorithms
AI has already proven to be instrumental for global healthcare. In May 2023, scientists announced that AI was used to discover a new antibiotic, abaucin, that was evidence to be effective against superbugs that the World Health Organisation had deemed a critical threat to humanity.
Other breakthrough discoveries have been Google DeepMind’s AlphaFold which can predict protein structures and has therefore changed the way that this type of science operates, allowing for quicker results and more efficient tackling of diseases.
CognoSpeak has already been hailed as having great potential for assessing people with memory problems faster and in a more effective way. With great importance placed on early intervention, efficiency is key. With the number of referrals having drastically increased over the past decade. It could result in more people being seen faster, thereby reducing delays.
Scientists at the Sheffield Institute for Translational Neuroscience (SITraN) are continually working to find ways to detect dementia earlier, as well as investigating potential treatments to slow the progress of the disease that has no current cure.
CognoSpeak’s system presents as a talking head 'digital doctor' on a computer screen. The department’s website states that to make a diagnosis, speech technology and machine learning is used to extract and classify distinct patterns of the person’s speech indicative of neurological dementia or functional memory disorder.
Because of this, the tool is a low-cost, repeatable, non-invasive and less stressful alternative to current cognitive assessment methods. It can even be conducted in the comfort of a patient’s home. This also offers possibility that the tool can be used across the world in the future.
It may also allow some people with diseases like Alzheimer’s to be reassured more quickly and be provided with appropriate advice and support. Taking a hybrid approach with clinicians and technology in this way has the potential to positively revolutionise healthcare in the future.
With AI being discussed as part of providing much-needed mental health solutions, the application of such a technology could speed up wait times and improve patient outcomes.
According to the university, early testing of the AI showed a 90% accuracy in predicting clinical diagnosis of functional memory disorder or neurological dementia. It has the potential to change screenings for early signs of dementia and Alzheimer’s by changing how brain scans are interpreted as well as using different memory tests.
Dr Dan Blackburn from the University of Sheffield's Department of Neuroscience said: “The CognoSpeak system could transform how dementia and other memory disorders are diagnosed by speeding up assessments.
“This would also free-up clinicians' valuable time and mean that those who need specialist care get access to it as quickly as possible."