As the population ages globally, conditions that will become more prevalent involve neurodegenerative ones. These types of disorders develop gradually over time, and unfortunately, are not detectable until late into the disease. These include conditions like Alzheimer’s disease, Parkinson’s disease, Lou Gehrig’s and other types of dementia. Though many have limited options for treatment, diagnosing early brain problems is still preferable. If nothing else, this would allow additional therapies to be explored that might prevent progression.
Given these objectives, one possible tool in diagnosing early brain problems involves the use of AI, or artificial intelligence. Researchers and companies are increasingly exploring the use of artificial intelligence in neurological disease. Some are using it to screen for Alzheimer’s in older adults. Others are using it to track eye movements that might hint at brain-related issues. Interestingly, not only are these tests showing promise, but they are detecting dementia years before clinical signs. (For more on the early detection of dementia, check out this Bold Business article.) Because of this, many health experts are quite enthusiastic about the promise that these AI tools may hold.
“This is the first report I have seen that took people who are completely normal and predicted with some accuracy who would have problems years later.” – Dr. Michael Weiner, Alzheimer’s Disease Researcher at the University of California, San Francisco
Language Screening with Artificial Intelligence in Neurological Disease
The Framingham Heart Study has been ongoing for decades. In the 1980’s early data from this longitudinal study completely changed the recommended diets and activities. But the study didn’t stop then. In fact, it has been ongoing, and now it is revealing some additional insights. IBM researchers are finding that AI examining data collected from these individuals predicts early brain problems 75% of the time. And it does so several years in advance.
The study examined 80 people in their 80s with normal brain function 7.5 years prior and followed them over time. The study routinely includes cognitive examinations, and these exams include tests of language abilities. By using AI, it found that those who went onto to develop Alzheimer’s disease had some notable abnormalities early. Specifically, this included repetitive word usage, omissions of prepositions, and increased grammatical and spelling errors. The researchers concluded that the findings support the use of artificial intelligence in neurological disease as a screening tool.
“The interesting thing with mobile artificial intelligence applications is that the more data we obtain, the more accurate our algorithms become at detecting and tracking the presence and progression of these various devastating neurological conditions.” – Etienne de Villers-Sidani, MD, Founder, CEO and Chief Scientific Officer at Innodem
Using AI in Other Conditions with Early Brain Problems
Using artificial intelligence in neurological disease extends well beyond Alzheimer’s disease. The researchers at IBM are not only exploring its use here but in other neurodegenerative conditions. These include illnesses such as Parkinson’s disease, ALS, and other dementias. It also involves psychiatric disorders like bipolar disease and schizophrenia. Each of these conditions show early brain problems well before clinical detection. If AI could detect these changes earlier, then potentially interventions might be employed to slow the disease.
Interestingly, researchers are looking at the use of AI in schizophrenia with these concepts in mind. Like Alzheimer’s disease, schizophrenia shows language changes as part of the condition. However, these changes are different and include more simplistic forms of speech and rapid, loosely-connected thoughts. Studies evaluating individuals with AI has shown that this tool could detect 85 percent of those who would go onto to have schizophrenia. And it did it three years before they met clinical criteria for diagnosis. Results like these are why there is excitement about the future use of artificial intelligence in neurological disease.
“For us, it is a priority to do the science correctly and at scale. We should have many more samples. There are more than 60 million psychiatric interviews in the U.S. each year but none of those interviews are using the tools we have.” – Guillermo Cecchi, an IBM researcher
AI Assessments of Eye Movements Also Show Promise
Language changes are not the only ones that occur in neurodegenerative diseases. Changes in eye movements can also be markers for early brain problems. Understanding this, Innodem Neurosciences, a Montreal start-up, is using AI to evaluate eye changes. Currently, its artificial intelligence in neurological disease detection is being applied to possible Multiple Sclerosis development. And early results look promising. In fact, Innodem recently received $6 million in Series A funding to further their efforts. As a result, the company plans to start exploring AI in other neurological conditions with altered eye movements as well. Specifically, it plans to study Parkinson disease and ALS patients in the future.
Innodem didn’t just stumble upon the use of AI in diagnosing early brain problems with eye movement changes. It first was used in an app called Pigio. Pigio was used to help individuals with communication and mobility issues communicate through eye movements. However, the vast amounts of anonymous data that was collected allowed AI patterns to be identified. Thus, Innodem suspected the use of artificial intelligence in neurological disease detection might show promise. Obviously, other investors believe they’re onto something based on the funding recently provided.
Leveraging AI and Data to Aid Disease Prevention
Advances in technology often bring about amazing discoveries and opportunities. Using artificial intelligence in neurological disease management reflects this statement well. Having the ability to diagnose early brain problems could be a real gamechanger for many. It provides opportunities to prepare, and it invites ways to perform better research. And for conditions that have preventative measures or treatments, it could improve quality of life. All of these reasons explain why the use of AI in neurodegenerative conditions is exciting. Given the early evidence cited above, data analytics and AI will likely be routine diagnostic tools employed in the future. And as data expands, these tools will only get better and better.
Want to make 2021 a better year than 2020? Then check out PROJECT BOLD LIFE: The Proven Formula to Take on Challenges and Achieve Happiness and Success.