Decoding Memory: AI’s Potential in Early Alzheimer’s Detection

Decoding Memory: AI’s Potential in Early Alzheimer’s Detection

Alzheimer’s disease affects around 5.5 million people in the U.S. and an estimated 55 million worldwide. What if we had the technology to stop it before it steals precious memories?

Alzheimer’s is a progressive type of dementia that destroys memories and thinking skills. While incurable, early detection is crucial for managing symptoms like forgetfulness, confusion, difficulty with planning and problem-solving, and personality changes. 

According to the Alzheimer Society, AD messes with the brain by making two protein substances called amyloid and tau clump together. These clumps, called plaques and tangles, disrupt normal brain function. Over time, Alzheimer’s also causes certain parts of the brain to shrink and messes up the chemicals that help brain cells communicate.

“There are several factors contributing to the challenges in treating Alzheimer’s Disease (AD), with three primary ones being: our incomplete understanding of Alzheimer’s disease root causes, the difficulty in achieving a definitive diagnosis, and the likelihood that earlier diagnosis could enable more effective prevention-focused treatments rather than diagnosing after considerable neurodegeneration has taken place,” says Julie Suhr, Professor and Director of Clinical Training at Ohio University.

According to the National Institute of Health Alzheimer’s is fixed with a lot of stigma, leading to a profound negative impact on both people diagnosed with AD and their caregivers.

“I think the most pervasive stigma is that these changes are just inevitable parts of aging and that there’s nothing anyone can do to stop them. Some of these beliefs are due to poor public knowledge of dementia in general and AD specifically. For example, while there are many changes in cognitive skills associated with aging, neuropsychological testing can help to determine whether these are normal and expected changes versus early signs of dementia, and early identification can lead to early treatment,” says Suhr. “Secondly, we cannot stop AD itself, there are many things you can do that will improve a person’s quality of life and keep them safe and independent for as long as possible if it is identified early enough.” 

According to the CDC, early detection means you can access available treatments sooner, build a supportive care network, tap into vital support services, and grant possible opportunities to join clinical trials. However, diagnosing Alzheimer’s relies on a combination of methods. 

Research published by the Alzheimer’s Association has unveiled a method that uses artificial intelligence (AI) algorithms to discover novel biomarkers for Alzheimer’s disease. These biomarkers would be new biological indicators that signal the presence or risk of the disease.

Neurologists at  Cleveland Clinic use advanced imaging tests like computed tomography (CT) and positron emission tomography (PET) scans to detect Alzheimer’s-related changes in your brain. These scans detect unusual activities or buildup of specific proteins related to Alzheimer’s disease (AD). They can also use advanced MRI scans, which are detailed pictures of your brain, to check the health of your brain tissue and rule out other conditions that might be causing similar symptoms to Alzheimer’s.

While these imaging tests offer valuable insights, they can be expensive and limited in their ability to predict or prevent Alzheimer’s. To address this, researchers are exploring new avenues. At Cleveland Clinic, a team led by Feixiong Cheng is finding a way to use artificial intelligence (AI) tools. 

This innovative approach aims to use artificial intelligence (AI) to analyze vast amounts of patient data, including brain scans, to identify novel drug targets for Alzheimer’s disease. By pinpointing these previously unknown targets, researchers hope to develop more effective treatments that could slow or even prevent the progression of this devastating disease.

In an article published by Cleveland Clinic, Feixiong Cheng, the principal investigator at the Genomic Medicine Institute reported,

“Advances in capable and intelligent computer-based algorithms offer the opportunity to harness large-scale data to pinpoint functional variants and risk genes that drive Alzheimer’s disease,” said Cheng. “That allows us to identify targets for treatment development.” 

While doctors use electronic health records (EHRs) to spot signs of COVID-19, they can also use them to look for signs of Alzheimer’s. With more data, this method could also help identify patterns in AD patients. The most valuable signs are those that make tests more accurate when combined with existing methods. Artificial intelligence (AI) is increasingly used to identify signs of the disease from medical images.

“Therapeutic approaches guided by a patient’s specific makeup should be much more effective than one-size-fits-all approaches. But, the existing genetic and genomic data available to us have not yet been fully utilized to explore targeted therapeutic development for AD, due in large part to the limitations of traditional analysis methods,” reported Dr. Cheng from the Cleveland Clinic. “However, advances in capable and intelligent computer-based algorithms offer the opportunity to harness large-scale data to pinpoint functional variants and risk genes that drive AD.”

While researchers are studying how to use AI to find new signs of Alzheimer’s disease, a major challenge is using datasets that may not fully represent the diversity of the disease. This might produce unreliable outcomes. To get around this, some advocate for larger patient recruitment and investigate how these AI discoveries may be applied to aid patients in the real world.

“What is interesting to me as a neuropsychologist is that there are decades of research showing that neuropsychological tests can identify very subtle cognitive decline, and studies have shown they can be useful in diagnosing AD,” says Suhr. “But these tests require time, beyond the time it takes to administer a brief and gross cognitive screener like the Mini-Mental State Examination (MMSE) or Montreal Cognitive Assessment (MOCA). These tests are very reliable and can be interpreted at an individual level, so long as you also know other sociodemographic characteristics of that person.”

The fight against Alzheimer’s disease is diverse, with researchers using cutting-edge AI in addition to traditional neuropsychological testing. Each finding adds a new piece to the puzzle, bringing us closer to the ultimate goal: early detection and successful intervention. 

Despite the hurdles, the scientific community is fuelled by unshakable hope and commitment, propelling us towards a future in which Alzheimer’s burden is reduced and better results are attainable.

Comments are closed.