Clinical results have shown that older brains whose age predicted by artificial intelligence (AI) is higher than their actual age are 150% more likely to suffer from dementia.
On the 15th, the team led by Professor Chang-Hyung Hong, Sang-Jun Son, and Hyun-Woong Noh of the Department of Psychiatry at Ajou University Hospital conducted a joint study with Professor Howard Aizenstein (Professor Helmet Karim) of the University of Pittsburgh School of Medicine, Samsung Seoul Hospital, Inha University Hospital, Pusan National University Hospital, and Chonnam National University Hospital. It was revealed that dementia in the ‘old brain’ could be predicted.
The research team analyzed the brain age by analyzing the brain magnetic resonance imaging (MRI) data of 650 Koreans obtained through the Chronic Cerebrovascular Disease Biobank of the Korea Human Resources Bank Project with artificial intelligence.
As a result, the age predicted by artificial intelligence appears higher than the actual age, that is, in the case of the ‘old brain’, the probability of developing dementia after 2 years is about 150% higher than that of the brain without it.
This is a very high level compared to 22% of the cognitive decline prediction questionnaire, which is currently widely used for early detection of dementia, and 57% of the MRI temporal lobe visual measurement.
In addition, if the predicted age was higher, the probability of amyloid positivity, which is a pathological finding of Alzheimer’s dementia, which accounts for about 70% of the causes of dementia, was about 25% higher.
It was confirmed that if the difference between the artificial intelligence predicted brain age obtained through brain image analysis and the actual age is large, there is a high possibility of developing dementia in the future.
The artificial intelligence model used this time was developed through joint research by Prof. Sangjun Sang and Prof. Aizenstein Karim (Geriatric Psychiatry Neuroimage Lab). learning outcomes were used.
If the artificial intelligence model predicts older people than they actually are, even if they do not yet have dementia, if they already show the volume change pattern of a degenerative brain disease, it is highly likely that they will progress to dementia in the future.
Elderly dementia due to aging has emerged as a serious social problem worldwide, but the development of drugs or non-pharmacological methods that show clear effects is still far away.
Prior to the onset of dementia, risk factors such as physical disease, lifestyle correction, and early cognitive decline are focused on drug treatment.
However, it is pointed out that dementia screening tests have limitations in screening various socio-demographic factors related to dementia and complex factors that cause dementia.
Professor Sang-Joon Sang said, “It is significant that we confirmed the possibility of early prediction of dementia progression through brain image analysis through artificial intelligence. I hope it will help open the market,” he said.
The results of this study were published in the international journal ‘Molecular Psychiatry’ (IF=13.437) in August.
The research was carried out with the support of the Brain Disease Convergence Research Center (Center Director, Ajou University Professor Park Sang-myeon) and research-oriented hospital (Research Director, Ajou University Hospital Professor Hong Chang-hyung).