Become a member of BioTechniques (it's free!) and receive the latest news in the life sciences, member-exclusives and 10% off BioTechniques article processing fees

Nothing’s perfect, but predicting Alzheimer’s onset nearly is

Written by Ryan Gilroy (Commissioning Editor)

MRI images of the brain for predicting Alzheimer's onset

Deep learningbased prediction of Alzheimer’s onset from MRI scans boasts 99% accuracy.

Kaunas University of Technology and Vytautas Magnus University (both Kaunas, Lithuania) researchers have developed a deep learning-based method that can predict the possible onset of Alzheimer’s from MRI brain images with over 99% accuracy, performing better than previously developed methods in terms of accuracy, sensitivity and specificity. 

“Medical professionals all over the world attempt to raise awareness of an early Alzheimer’s diagnosis, which provides the affected with a better chance of benefiting from treatment,” commented author Rytis Maskeliūnas (Kaunas University of Technology). 

One of the possible indicators of predicting Alzheimer’s disease (AD) is mild cognitive impairment (MCI), which is the stage between the expected cognitive decline of normal aging and dementia. Based on previous research, functional magnetic resonance imaging (fMRI) can be used to identify regions in the brain that can be associated with onset of AD. The earliest stages of MCI often have almost no clear symptoms, but in many cases can be detected by neuroimaging. 

Manual analysis of fMRI images in an attempt to identify predictive AD-associated changes not only requires specific knowledge but is also time-consuming. Applying deep learning can speed this up by a significant time margin. Finding MCI features does not necessarily imply the presence of illness, but it is a useful indicator during evaluation by a medical professional. 

“Modern signal processing allows delegating the image processing to the machine, which can complete it faster and accurately enough. Of course, we don’t dare to suggest that a medical professional should ever rely on any algorithm one hundred percent. Think of a machine as a robot capable of doing the most tedious task of sorting the data and searching for features. In this scenario, after the computer algorithm selects potentially affected cases, the specialist can look into them more closely, and at the end, everybody benefits as the diagnosis and the treatment reaches the patient much faster,” described Maskeliūnas. 

group of elderly people exercising with weights

Regular exercise could help prevent onset of Alzheimer’s disease

Regular physical exercise has been shown to help modulate iron metabolism in the brain and in muscles in a new study.

The deep learning-based model was developed using a modification of finetuned ResNet 18 (residual neural network) to classify fMRI images obtained from 138 subjects. The images fell into six different categories: normal control, MCI, early MCI, late MCI, significant memory concern, and AD. In total, 51,443 and 27,310 images were selected, for training and validation respectively, from The Alzheimer’s Disease Neuroimaging Initiative fMRI dataset . 

The model was able to effectively find MCI features in the given dataset, achieving the best classification accuracy of 99.99%, 99.95% and 99.95% for early MCI vs. AD, late MCI vs. AD, and MCI vs. early MCI, respectively. 

“Although this was not the first attempt to diagnose the early onset of Alzheimer’s from similar data, our main breakthrough is the accuracy of the algorithm. Obviously, such high numbers are not indicators of true real-life performance, but we’re working with medical institutions to get more data,” added Maskeliūnas. 

Researchers hope the algorithm could be developed into software, which would analyze the collected data from vulnerable groups (over 65s, history of brain injury, high blood pressure) and notify medical personnel about the anomalies which could predict the early onset of AD. 

Additionally, the model can be integrated into a more complex system, analyzing multiple parameters, such as eye movement tracking, face reading and voice analyzing to enable improved prediction of Alzheimer’s. Such technology could then be used for self-checking and alerting patients to seek professional advice.