[ad_1]
- Mild cognitive impairment (MCI) often precedes the development of Alzheimer’s disease.
- Functional MRIs (fMRIs) can capture subtle signs of MCI, but they are difficult to interpret.
- Lithuanian researchers developed a deep learning algorithm that identified MCI in a small study.
One of the first indicators of the onset of Alzheimer’s disease (AD) is the development of MCI. Subtle and hard to detect changes in the brain accompany MCI as the disease progresses.
Today, a study by researchers at Kaunas University of Technology (KTU) in Lithuania presents a newly developed deep learning computer algorithm that can accurately detect and differentiate MCI stages from fMRI scans.
The algorithm can identify MCI and its stages with more than 99% accuracy.
MCI is a transitional state between normal age-related cognitive decline and dementia. It does not always progress to AD, but it often does, and early detection of AD may allow people with AD to benefit more from treatment.
“Healthcare professionals around the world are trying to raise awareness about early diagnosis of Alzheimer’s disease, which offers those affected a better chance of receiving treatment,” says the study’s chief investigator, the Dr Rytis Maskeliūnas.
Claire Sexton, DPhil, who is the director of science programs and outreach at the Alzheimer’s Association and was not involved in the research, said Medical News Today:
“An early and accurate diagnosis can have emotional, social and medical benefits, allowing individuals to develop legal, financial and care plans, explore treatment options and participate in clinical trials. “
The study, which KTU Ph.D. student Modupe Odusami led, appears in the journal MDPI.
Although it is possible to manually recognize the MCI in fMRI images, it is a tedious task that requires detailed knowledge. As such, it is an ideal candidate for automation using deep learning. Deep learning is a type of computer algorithm that can learn to detect patterns in data that may be too small or obscure for humans to easily recognize.
Working with collaborators in artificial intelligence, KTU researchers modified an existing well-known algorithm, ResNet 18, to refine it to detect MCI.
After the training process, the researchers tested the algorithm by classifying the fMRI scans of 138 people.
The scans represented six cognitive stages, starting with healthy control and moving from MCI to AD. Differentiating between early MCI and AD, the algorithm was 99.99% accurate. It was also 99.95% accurate in distinguishing between late MCI and AD, and between MCI and early MCI.
Dr Maskeliūnas notes:
“Although this is not the first attempt to diagnose early onset Alzheimer’s disease based on similar data, our main advance lies in the accuracy of the algorithm.”
“Obviously,” says Dr Maskeliūnas, “such high numbers are not indicators of real performance, but we are working with medical institutions to get more data. “
MNT asked Dr Maskeliūnas about his expectations for the algorithm’s accuracy in the real world. He replied, “I would say that a reliable rate of over 85% would still be beneficial for a healthcare professional, thus reducing [their] workload on data analysis.
“At this point,” he said, “we are working on fine-tuning algorithms, and despite some results on a controlled dataset put together by others, it is very likely that we will still have to rework it to. take into account variations. in real data.
Dr Sexton suggested it was too early to confirm the algorithm’s value, stating:
“This is an interesting but small study (25 participants with Alzheimer’s disease). As a result, we cannot yet draw any conclusions about the proposed new diagnostic technique. “
Says Dr Sexton of the algorithm: “Replication of these results in larger and more diverse study groups is needed to assess its potential.
Dr Maskeliūnas plans to develop an algorithm-based application that doctors could use to identify MCI in people at risk for AD. They could then refer these people for treatment.
He is also interested in the potential of integrating the team’s algorithm into a self-monitoring system that includes other early diagnostic methods currently under study. Examples of these other methods are eye movement tracking, face reading, and voice analysis.
According to Dr. Sexton, these new technologies “are still under study. Some are now being incorporated into trials, albeit on an exploratory basis, to collect additional data from larger studies. Conclusion: Although they are definitely progressing in terms of use, they are not yet in clinical use. “
In the KTU press releaseDr Maskeliūnas says: “We need to make the most of the data. This is why our research group is focusing on the European principle of open science, so that everyone can use our knowledge and develop it further. I believe that this principle contributes greatly to the advancement of society.
Dr Maskeliūnas concludes:
“Technology can make medicine more accessible and cheaper. While they will never – or at least not soon – truly replace the healthcare professional, technology can encourage the search for a diagnosis and timely help. “
[ad_2]
Source link