As dementia becomes increasingly prevalent, researchers at the Mayo Clinic believe that artificial intelligence is essential for facilitating earlier and quicker diagnoses.
By integrating AI with EEG (electroencephalogram) tests, the Mayo Clinic Neurology AI Program (NAIP) in Rochester, Minnesota, successfully identified particular types of dementia sooner than what human analysis could achieve.
According to a hospital press release, these findings suggest that EEGs could ultimately offer a more accessible, cost-effective, and less invasive method for early brain health assessments.
WHAT IS ARTIFICIAL INTELLIGENCE?
The research was published in the journal Brain Communications last week.
What is an EEG?
An EEG involves a technician placing small metal electrodes on the patient’s scalp to measure electrical activity in the brain.
This test produces a graphic representation of brain electrical impulses manifested as wavy lines.
An electroencephalogram (EEG) is a test that measures electrical activity in the brain using small metal discs (electrodes) attached to the scalp. This activity is displayed as wavy lines on an EEG recording. (iStock)
This technique is mainly used for diagnosing epilepsy but can also identify other brain conditions, as noted by Dr. David Jones, a clinician who specializes in behavioral neurology and leads the artificial intelligence program at Mayo Clinic in Minnesota.
MAYO CLINIC IDENTIFIES NEW TYPE OF MEMORY LOSS MIMICKING ALZHEIMER’S
The clinic conducts thousands of EEGs annually for patients with neurological issues.
It’s been established that brainwave patterns alter in individuals with dementia or cognitive disorders related to Alzheimer’s or Lewy body disease, Jones shared in a phone interview with Fox News Digital.
“However, achieving this information requires extensive expert analysis, expertise, and manual effort; thus, EEGs are not routinely utilized to assess Alzheimer’s and dementia,” he stated.
Uncovering ‘hidden data’
In this study, the researchers aimed to uncover “hidden data” within patients’ brain waves using computer algorithms, eliminating the need for manual labor, according to Jones.
The AI model was developed internally at Mayo Clinic, trained on data from over 11,000 patients undergoing EEGs over a span of ten years.
EEGs are primarily used to diagnose epilepsy, but can also help identify other brain conditions. (iStock)
Through the examination of complex brain waves, the model discovered six distinct patterns present in patients with Alzheimer’s or Lewy body disease that were absent in healthy individuals, as stated by Jones.
This correlation was verified by analyzing additional metrics such as cognitive assessments, blood biomarkers, and brain PET scans.
INCREASED RISK OF ALZHEIMER’S IF FAMILY HISTORY IS ON MOTHER’S SIDE, STUDY REVEALS
On the whole, the AI tool halved the EEG reading time and significantly enhanced the accuracy of the results, Jones noted.
“This indicates that a wealth of unused information exists in clinically acquired EEGs that we can extract automatically — and we can start to develop improved tools, algorithms, and techniques,” Jones added.
The Mayo Clinic’s Neurology AI Program (NAIP) successfully identified specific types of dementia much earlier than what was achievable through human analysis. (iStock)
Conducting such analysis on a large scale would be “extremely challenging” without AI or machine learning technology, as per the neurologist.
A ‘significant advancement’
Harvey Castro, a board-certified emergency medicine physician based in Dallas and a national speaker on AI in healthcare, who was not part of the study, characterized Mayo Clinic’s research as “a significant advancement.”
“This technology can swiftly and accurately analyze brain wave patterns, revealing early indicators of dementia that often elude human detection,” he stated to Fox News Digital.
“A wealth of unused information exists in clinically acquired EEGs that we can extract automatically.”
While Castro does not typically utilize EEGs in his emergency practice due to the lengthy interpretation process, he noted, “AI enables rapid processing of extensive datasets, allowing for prompt, informed decisions regarding a patient’s cognitive health.”
“I foresee this evolving into a new tool I can employ in the ER,” he remarked.
The ultimate aim is to integrate various assessments such as brain scans, blood tests, cognitive evaluations, and brain waves into a unified model for brain health, as stated by a researcher. (iStock)
AI-enhanced EEG analysis has the potential to be a “game-changer” particularly in rural and underserved regions, Castro suggested.
“It presents a cost-effective, non-invasive means to identify cognitive issues early, especially where advanced diagnostic resources like MRIs or PET scans are scarce.”
Future directions
The aspiration is to incorporate this AI-based EEG analysis into a “multimodal” methodology for dementia evaluations, according to Jones.
“This signifies the capability to create a complete model of brain health that combines brain scans, blood tests, cognitive assessments, and EEG data,” he informed Fox News Digital.
ARTIFICIAL INTELLIGENCE DETECTS CANCER WITH 25% GREATER ACCURACY THAN DOCTORS IN UCLA STUDY
The subsequent phase involves implementing the AI tool into standard clinical practices.
“If you arrive for an EEG due to epilepsy or for a sleep study, we would also be able to provide insights into your cognitive health and flag any issues that may require a consultation with a behavioral neurologist,” Jones said.
Although several years of research remain before this technology becomes broadly available, researchers are optimistic. (iStock)
Looking ahead, the neurologist envisions that EEGs could evolve into a “highly scalable and mobile” technology, enabling individuals to conduct cognitive assessments remotely — similar to measuring blood pressure or heart rate at home.
CLICK HERE TO GET THE FOX NEWS APP
Several years of investigation are still required before this technology can be widely accessible, according to Jones.
Risks and challenges
Despite the advantages of this technology, Castro warned that integrating AI into clinical practice comes with challenges.
“Although AI can offer valuable insights, the clinician’s skills and compassion cannot be replaced.”
“Several hurdles must be overcome, including extensive training for healthcare personnel to use these tools effectively and the risk of becoming overly reliant on AI at the expense of clinical judgment,” he expressed to Fox News Digital.
Maintaining a balance between AI usage and the necessary “human element” is crucial, Castro added.
CLICK HERE TO SIGN UP FOR OUR HEALTH NEWSLETTER
“Even though AI can yield important insights, the expertise and empathy of clinicians are irreplaceable for providing comprehensive patient care.
Other factors to consider include ensuring patient data privacy, securing informed consent, and working to mitigate biases in AI algorithms, as stated by the doctor.
In analyzing complex brain waves, the model identified six specific patterns in individuals with Alzheimer’s or Lewy body disease, distinct from those without cognitive troubles. (iStock)
Jones, the neurologist from Mayo Clinic, acknowledged the risks of over-reliance on algorithms but emphasized that the technology is based on “real-world data for practical applications.”
“Its significance is measured by whether it actually aids in patient care — that’s our primary concern,” he stated.
For more health articles, visit www.foxnews/health
The team is “highly cognizant” of potential issues and takes precautions to mitigate them, Jones told Fox News Digital.
“We adhere to established AI and machine learning practices as part of our software design philosophy and the core values of Mayo Clinic.”
