The onset of mild cognitive impairment (MCI) is one of the earliest signs of incipient Alzheimer’s disease (AD). As the disease progresses, MCI is accompanied by subtle, difficult-to-detect alterations in the brain. Researchers from Kaunas University in Lithuania have published a paper in which they describe a newly created deep-learning computer system that can correctly identify and distinguish the phases of MCI from functional magnetic resonance imaging (fMRI) scans.
Early Symptoms Of Alzheimer’s Disease
MCI and its phases may be identified with more than 99 percent accuracy using the algorithm. MCI is a stage in the progression of usual age cognitive decline to dementia that occurs between the ages of 40 and 60. It may not always develop to Alzheimer’s disease, but it does so often, and early diagnosis of Alzheimer’s disease may enable those who have it to gain more from therapy.
Even while it is feasible to detect MCI in fMRI pictures, doing so manually takes a long time and needs a great deal of experience and understanding. As a result, it is an excellent candidate for automation via the use of deep learning. Deep learning is a kind of computer system that can teach how to identify patterns in data that are too tiny or obscure for people to detect readily. Deep learning is used to detect patterns in the data that are too small or complex for humans to detect easily. Researchers at KTU updated an existing algorithm known as ResNet 18 in collaboration with colleagues in artificial intelligence to make it more effective at identifying motor neuron disease (MCD).
Following the training procedure, the researchers put the system to the test by categorizing fMRI data from 138 different people. The scans revealed six phases of cognitive decline, beginning with healthy control and progressing through mild cognitive impairment to Alzheimer’s disease. When it came to distinguishing between early MCI and Alzheimer’s disease, the system was 99.99 percent correct. Additionally, it was 99.95 percent accurate in differentiating between late MCI and Alzheimer’s disease, as well as between MCI as well as early MCI.
Dr. Maskelinas was questioned by MNT about his hopes for the algorithm’s accuracy in the actual world, and he responded positively. According to his response, “I’d think a dependable 85+ percent would still be beneficial for a medical practitioner since it would reduce their burden associated with data processing.”
Dr. Sexton stated that it is too soon to determine the usefulness of the algorithm, stating that this is intriguing but tiny research (with just 25 individuals with Alzheimer’s disease). Therefore, we are unable to make any conclusions regarding the suggested novel diagnostic method at this time.
Dr. Sexton has the following to say about the algorithm: “Replication of these findings in bigger, more varied research groups is required to assess its potential,” he says. Dr. Maskelinas envisions the creation of algorithm-based software that physicians might use to detect mild cognitive impairment (MCI) in individuals at risk of Alzheimer’s disease. They would then be able to refer these people to appropriate therapy.
With over 15 years as a practicing journalist, Nikki Attkisson found herself at Powdersville Post now after working at several other publications. She is an award-winning journalist with an entrepreneurial spirit and worked as a journalist covering technology, innovation, environmental issues, politics, health etc. Nikki Attkisson has also worked on product development, content strategy, and editorial management for numerous media companies. She began her career at local news stations and worked as a reporter in national newspapers.