As dementia becomes more widespread, Mayo Clinic researchers believe artificial intelligence could be key to enabling earlier and faster diagnoses.
By combining AI with EEG (electroencephalogram) testing, a team at the Mayo Clinic Neurology AI Program (NAIP) in Rochester, Minnesota, was able to identify certain types of dementia faster than human analysis.
Based on these findings, EEG may ultimately provide a more accessible, less expensive and less invasive way to assess brain health earlier, according to a press release from the hospital.
What is Artificial Intelligence?
The study was published last week in the journal Brain Communications.
What is EEG?
During an EEG, a technician places small metal electrodes on the patient’s scalp to measure electrical activity in the brain.
The test produces a recording of wavy lines that represent the brain’s electrical signals.
Electroencephalography (EEG) is a test that measures electrical activity in the brain using small metal disks (electrodes) placed on the scalp. This activity shows up as wavy lines on an EEG recording. (iStock)
While artificial intelligence is primarily used to diagnose epilepsy, it can also be used to identify other brain disorders, says Dr. David Jones, a practicing clinician specializing in behavioral neurology who directs the artificial intelligence program at the Mayo Clinic in Minnesota.
Mayo Clinic discovers new type of memory loss often mistaken for Alzheimer’s
The clinic performs thousands of EEG tests each year to evaluate patients with neurological disorders.
It has long been known that brainwave patterns change in people with dementia and cognitive impairment caused by Alzheimer’s disease and Lewy body disease, Jones told Fox News Digital in a phone interview.
“But extracting that information requires a lot of specialized analysis, expertise and manual work, so Alzheimer’s disease and dementia are not routinely assessed with EEG.”
Utilizing “hidden information”
In this study, the researchers used computer algorithms to try to find “hidden information” in patients’ brainwaves, rather than relying on manual methods, Jones said.
The AI tool was developed in-house at Mayo Clinic and trained on data from more than 11,000 patients who underwent EEG testing over a 10-year period.
EEG is primarily used to diagnose epilepsy, but can also be used to identify other brain conditions. (iStock)
In analyzing complex brainwaves, Jones said, the model identified six specific patterns seen in people with Alzheimer’s disease and Lewy body disease that weren’t seen in people without cognitive impairment.
This correlation was confirmed by looking at other indicators, including cognitive tests, blood biomarkers and brain PET scans.
Study finds that having Alzheimer’s in your maternal family may increase your risk of the disease
Overall, the AI tools reduced EEG reading times by 50% and improved reading accuracy “quite significantly,” Jones said.
“This shows us that there’s a lot of unused information in clinically acquired EEG that can be extracted automatically, and now we can start to build better tools, algorithms and methods,” Jones said.
A team from the Mayo Clinic Neurology AI Program (NAIP) in Rochester, Minnesota, was able to identify certain types of dementia faster than human analysis could. (iStock)
Neurologists say that carrying out this kind of analysis at scale without AI or machine learning techniques is “extremely difficult.”
“A major step forward”
Harvey Castro, a Dallas-based board-certified emergency physician and national speaker on artificial intelligence in medicine, who was not involved in the study, called the Mayo Clinic research “a major step forward.”
“The technology can quickly and accurately analyze brainwave patterns and identify early signs of dementia that are often invisible to the human eye,” he told Fox News Digital.
“There’s a lot of unused information in clinically acquired EEG that can be extracted automatically.”
Castro, the emergency physician, said he doesn’t typically use EEG tests because the results take time to interpret.
“But AI technology makes it possible to crunch large amounts of data, enabling faster and more informed decisions to be made about a patient’s cognitive health,” he said.
“So, I think this will be a new tool for me to use in the emergency room.”
The ultimate goal, the researchers said, is to combine brain scans, blood tests, cognitive tests and EEG into a “complete model of brain health.” (iStock)
Castro said AI-powered EEG analysis could be a “game changer” in rural and medically underserved areas.
“It offers a cost-effective, non-invasive way to screen for early cognitive problems when advanced diagnostic tools such as MRI and PET scans are limited.”
Next steps
The ultimate goal, Jones says, is to incorporate this AI-driven EEG analysis into a “multi-modal” approach to dementia testing.
“That means we can bring together brain scans, blood tests, cognitive tests and EEG into a complete model of brain health,” he told Fox News Digital.
UCLA study finds artificial intelligence detects cancer 25% more accurately than doctors
The next step is to introduce AI tools into everyday clinical practice.
“So if you have an EEG for epilepsy or a sleep study, it will also tell you something about your cognitive health at the same time and whether it finds anything that might indicate you need to see a behavioral neurologist,” Jones said.
The researchers noted that the technology still needs several years of research before it can be widely used. (iStock)
In the future, the neuroscientist predicts, EEG will become a “scalable, portable” technology, and people may even be able to undergo cognitive assessments remotely, “just like taking your blood pressure or heart rate at home.”
Click here to get the FOX News app
Jones noted that the technology still needs several years of research before it’s widely available.
Potential risks and limitations
Despite the benefits of this type of technology, Castro warned that there are challenges in integrating AI into clinical practice.
“AI can provide valuable insights, but the expertise and empathy of clinicians is irreplaceable.”
“This includes the need for healthcare workers to be adequately trained to use these tools effectively, as well as a potential over-reliance on AI at the expense of clinical judgment,” he told Fox News Digital.
Balancing the use of AI with the “human touch” is also important, Castro said.
Click here to sign up for our Health Newsletter
“While AI can provide valuable insights, clinician expertise and empathy remain essential to providing comprehensive patient care.”
Other considerations include ensuring the privacy of patient data, obtaining informed consent and working to prevent bias in AI algorithms, the doctor added.
By analysing complex EEG data, the model identified six specific patterns seen in people with Alzheimer’s disease and Lewy body disease that were not seen in people without cognitive impairment. (iStock)
Jones, the Mayo Clinic neurologist, acknowledged that there are risks in relying too heavily on algorithms but stressed that the technology is designed “with real-world data for real-world use.”
“Its value is measured by whether it helps with patient care. That’s our focus.”
For more health stories, visit www.foxnews/health
Jones told Fox News Digital that the team is “fully aware” of the potential issues and is taking steps to mitigate them.
“We follow good AI and machine learning practices as part of our software design ethos and Mayo Clinic values.”
