A new AI tool has been developed by researchers at the University of Sheffield to help doctors assess early signs of dementia and Alzheimer’s disease more quickly and efficiently.
Known as CognoSpeak, the system uses a virtual agent that appears on the screen to talk to patients. It asks memory-probing questions inspired by questions used in ambulatory care, and administers cognitive tests such as picture descriptions and verbal fluency tests.
The tool then uses artificial intelligence and voice technology to analyze language and speech patterns, looking for signs of dementia, Alzheimer’s disease, and other memory problems.
The researchers behind the technology believe that CognoSpeak could play a key role in reducing the burden on dementia assessment services once further testing is completed in GPs and secondary care memory practice clinics across the UK. says it has the potential.
The system is designed to work between primary and secondary care. This means that, once fully deployed, general practitioners can introduce people with memory impairments to the use of this technology. CognoSpeak will send the test results back to the GP, who will decide if the patient should be referred to a memory clinic for further evaluation.
CognoSpeak is accessible from a web browser, so patients can take the test in the comfort of their own home using a computer, laptop, or tablet, rather than waiting for a hospital appointment to take a paper-and-pencil test. It can cause undue stress and anxiety.
Early trials show the technology is as accurate in predicting Alzheimer’s disease as current pen-and-paper-based tests used to assess and screen for disorders of cognition, memory and thinking. . The researchers demonstrated that they could distinguish Alzheimer’s patients from cognitively healthy people with 90 percent accuracy.
The CognoSpeak system, developed by Dr Dan Blackburn and Professor Heidi Christensen from the University of Sheffield’s Department of Neuroscience and Computer Science, is still in the research stage, but has received a £1.4m grant from the National Institute for Health and Medical Research (NIHR). Developed with money. ), the technique has been tried more extensively. The researchers are recruiting his 700 participants from memory clinics across the UK to help further develop the system.
“Waiting for a dementia diagnosis can be a very anxious time for patients and their families,” said Dan Blackburn, Ph.D., Department of Neuroscience, University of Sheffield. “It could help us get started, reduce wait times, and help treat patients.” Certainty first.
“The CognoSpeak system has the potential to change the way dementia and other memory disorders are diagnosed by speeding up assessment. People who do will have access as soon as possible.”
“The way a person speaks can tell us a lot about their cognitive and mental health, and can show early signs of cognitive decline,” said Heidi Christensen, professor of computer science at the University of Sheffield. said. The system we developed in Sheffield uses voice technology to automatically extract these signs, which means automation can give everyone a consistent, accurate and fast assessment.
“CognoSpeak is advanced, high-tech, and based on world-leading research in the field. I am making use of it.”
The CognoSpeak tool was developed in collaboration with The Therapy Box, which specializes in spoken language technology, and the Device MedTech Cooperative for Dignity (D4D) at the National Institute of Health Research (NIHR). D4D is leading our efforts with patients and families. A minority community group to ensure AI is acceptable, trustworthy, and accessible to all future users.
The research team is also working to make this tool available to patients from minority communities who are less likely to attend dementia services and who may speak English as an additional language.
Lise Sproson, Director of Patient and Public Engagement at D4D, said: “We work closely with a wide range of community groups, including the Israeli Somali Community Association in Sheffield, to jointly develop the look and feel of the CognoSpeak system and ensure that it is acceptable.” , relevant and easy to use.
“We are training our AI to speak English as an additional language with different UK regional accents to maintain the accuracy previously demonstrated in tests with native English speakers across the population. “
Dr. Blackburn, Emeritus Consultant Neurologist at Sheffield Teaching Hospital NHS Foundation Trust and researcher at the NIHR Sheffield BRC, said: “There is a real clinical need for this kind of technology. There is a long line of memory clinics across the UK. There is a waiting list,” he added. But there are also inequalities in access to memory clinic services. CognoSpeak tools can help reduce these inequalities and make your service more efficient. “
About 900,000 people in the UK are currently living with dementia, and that number is expected to nearly double by 2040, according to the Alzheimer’s Association. Referrals for evaluation are increasing rapidly, and long wait times are often experienced at memory clinics.
Professor Mike Lewis, Director of the NIHR i4i Program, said, “Cognospeak is a game-changing example of how digital health technology can transform how we approach conditions such as dementia, ensuring that patients It will help us find ways to make patient discovery and assessment easier for us to access.” Appropriate treatment and support
Patients are being recruited to the CognoSpeak trial through memory clinics across the UK. To join the trial, please visit www.cognospeak.com.
