Professor Keiichi Onoda of Otemon Gakuin University, Professor Atsushi Nagai of the Department of Neurology, Shimane University School of Medicine, and a joint research team of ERISA Co., Ltd., who specialize in cognitive neuroscience, said from MRI structural images of the brain, "When and with what probability Alzheimer I succeeded in predicting whether I will develop the disease.
The research team of Professor Onoda et al. Analyzed the data of 2142 cases including MRI images of the brains of healthy people and Alzheimer's disease patients and the subsequent onset status by deep learning.We succeeded in predicting the probability of onset over time.So far, there have been studies on whether Alzheimer's disease is likely to occur or not, but in this study, the probability of developing Alzheimer's disease after 1 year, 2 years, and every year is determined. Established a method that can be predicted at the level.The accuracy of its onset estimation reached 83.5%.
In addition, when we investigated which brain region was important for predicting the onset of Alzheimer's disease, not only the default mode network, which has been suggested to be related to Alzheimer's disease, but also the prominent network such as the anterior cingulate gyrus and the insular cortex is important information. It was shown to be the source.
This study is positioned as a basic study to predict the onset time of Alzheimer's disease, which is one of the main causes of dementia.Regarding this achievement, Professor Onoda said, "It has become possible to evaluate the risk of developing Alzheimer's disease at the individual level according to the number of years elapsed in the future. In the future, the research results may be applied to medical treatment, etc."
Paper information:[Brain Communications] Prediction of conversion to Alzheimer's disease using deep survival analysis of MRI images
