Scientists at Kaunas University of Technology (KTU) in Lithuania have developed a method that can predict the possible onset of Alzheimer's disease with almost one hundred percent accuracy.
Editor-in-Chief Ornament Health AG
According to the World Health Organization, up to 70 % of dementia cases are caused by Alzheimer's disease. Worldwide, about 24 million people suffer from the disease, and the number is expected to double every 20 years.
Medical professionals around the world strive to diagnose Alzheimer's disease as early as possible, which gives patients a better chance of successful treatment.
One of the first signs of Alzheimer's disease is mild cognitive impairment (MCI), which is a transitional stage between the expected cognitive decline of normal aging and dementia.
The earliest stages of MCI often have almost no clear symptoms, but in many cases can be detected by neuroimaging. However, although theoretically possible, manual analysis of MRI images to detect changes associated with Alzheimer's disease requires not only knowledge but also a lot of time.
The use of artificial intelligence can greatly speed up diagnosis. Modern signal processing makes it possible to delegate image processing of functional MRI brain scans to a machine, which can do it faster and more accurately.
However, researchers do not claim that the algorithm can be fully relied on. Rather, the machine is a robot capable of performing the most tedious task of sorting through data and searching for functional abnormalities.
"Once the computer algorithm selects cases of potential abnormalities, the technician can examine them more closely. And in the end, it will benefit everyone, as the patient will be diagnosed much faster and treatment will begin," said Rytis Maskeliunas, head of the development team and professor at KTU's Computer Science Department.
The model was developed through a collaboration of leading Lithuanian researchers in the field of artificial intelligence, using a modification of the well-known fine-tuned ResNet 18 (residual neural network) to classify functional MRI images from 138 patients. The images fell into six different categories, from healthy to trait with moderate cognitive impairment (MCI) to Alzheimer's disease. The model was able to effectively find MCI features in this dataset, achieving nearly 100 percent classification accuracy.
According to the researchers, the method can be turned into software that will analyze collected data from vulnerable groups (people over 65 with a history of brain injury, high blood pressure, and other risk factors) and notify medical personnel of abnormalities associated with early-onset Alzheimer's disease.
21 October 2021
You can discuss. Open this post in the Ornament app and add your opinion.