U.K.University of CambridgeScientists have developed an artificial intelligence tool that can accurately predict the progression of early dementia patients with more than 80%.Alzheimer's diseaseThe new approach could reduce the need for invasive, expensive diagnostic tests while improving treatment outcomes at earlier stages of the disease, increasing the chances that interventions, such as lifestyle changes or new drugs, will work, the team said.
More than 55 million people worldwide suffer from dementia, of which 60-80% is the type of Alzheimer's disease, causing economic losses of $820 billion each year. The number of dementia cases is expected to nearly triple in the next 50 years.
Early detection is crucial for Alzheimer's disease, when treatment is most effectiveHowever, the diagnosis and prognosis of early dementia can be inaccurate without invasive or expensive tests such as positron emission tomography (PET) or lumbar puncture, which are not available in all memory disorder clinics. As a result, up to a third of patients may be misdiagnosed, and others may be diagnosed after treatment has expired.
A team led by the University of Cambridge's Department of Psychology has developed a machine learning model that can predict whether and how quickly individuals with mild memory and thinking problems will develop Alzheimer's disease. The team showed in a study published in the journal eClinical Medicine that the model is more accurate than current clinical diagnostic tools.
Based on cognitive test and MRI scan data from 400 patients with gray matter atrophy collected by a US research team, the research team used machine learning algorithms to build an AI prediction model and tested the model using real-world data from multiple clinics in the UK, Singapore and other countries.
The test results show thatThe model was able to accurately identify people who would develop Alzheimer's disease within three years with an accuracy of 82%, and people who would not develop Alzheimer's disease within three years with an accuracy of 81%.
The algorithm was about three times more accurate in predicting Alzheimer's disease progression than the current standard of care (i.e., standard clinical markers such as gray matter atrophy or cognitive scores) or clinical diagnosis, suggesting that the model could significantly reduce misdiagnoses.
In the future, the research team hopes to expand the model to predict other types of dementia, such as vascular dementia and frontotemporal dementia, and use different types of data, such as markers in blood tests.