University scientists in Kaunas (Lithuania) have developed a method that can predict the possible start of Alzheimer’s disease on brain images with an accuracy of more than 99 percent. The method was developed on the basis of the analysis of MRI functional images obtained from 138 patients. This is reported in the article published in the journal Diagnostics.
One of the possible first signs of Alzheimer’s disease is a slight cognitive impairment (MCI), which is an intermediate stage between the expected cognitive decline in normal aging and dementia. Functional magnetic resonance imaging (FMRT) can be used to identify the areas of the brain that may be associated with the beginning of Alzheimer’s disease.
The earliest MCI stages often almost do not have clear symptoms, but in many cases can be detected by neurovalization. The use of deep teaching and other methods of artificial intelligence (AI) can significantly speed up this process. In total, 51,443 and 27 310 images from the FMRT data set were selected for learning and checking. The model was able to effectively recognize the signs of early cognitive violations in this data set, reaching the best classification accuracy of 99.99 percent.
According to scientists, the algorithm can be transformed into software that will analyze the collected data from vulnerable groups (persons over 65 years old, having a history of brain injury and high blood pressure).