AI-zheimer’s prediction using AI

by Kylie Chen (’24) | March 1, 2021

Artificial Intelligence Algorithm Helping an Alzheimer’s Patient. Art by Ava Hennen (’22)

Currently, almost six million Americans suffer from Alzheimer’s disease, a number projected to rise to 13.8 million by 2050. Despite the increasingly high numbers of those who live with and die of Alzheimer’s, this disease has no cure and continues to hurt patients and family members, both emotionally and financially. However, a new artificial intelligence (AI) program may be the key to the early diagnosis of Alzheimer’s, opening the door to the discovery of more effective methods of treating both this disease and other neurological illnesses.

Alzheimer’s disease (AD) is a progressive brain disease that causes 60-80% of dementia cases, rendering it the most common cause of dementia. It damages brain cells, causing such cells to lose proper functionality. At first, the microscopic changes are undetectable, but eventually, they lead to worsening dementia symptoms. Once the disease has advanced to its late stages, doing basic, everyday tasks like talking, swallowing, and walking gets increasingly difficult. This disease has devastating effects on the lives of millions in America alone, costing patients billions of dollars for long-term care and making it the sixth leading cause of death in the United States.

This could all change. In October of last year, researchers published a study examining the relationship between linguistic performance and AD. They developed an artificial intelligence program using natural language processing and machine learning and trained it to detect subtle differences in subjects’ diction. How is this study related to Alzheimer’s? Researchers tested eighty people, half of whom developed AD symptoms before turning 85 years old, by asking subjects to describe a picture of a “cookie theft” in writing. The AI program found that the subjects with more repetitive word choice, spelling and capitalization errors, and simpler grammatical structures, which sometimes excluded subjects and other words, developed Alzheimer’s disease. The program predicted AD with around 75% accuracy; in contrast, predictions for AD on clinical scales have roughly 59% accuracy.

Although there is currently no cure for Alzheimer’s, early detection of the disease allows for early intervention, making it easier to slow the progression of the disease. It also has the potential to help scientists develop better methods to effectively treat the disease at its earlier stages. Overall, this advancement has the potential to change the way physicians diagnose and treat all neurological illnesses, not just Alzheimer’s. Since people with different neurological illnesses have unique language patterns, researchers believe that these linguistic patterns could serve as early signs of the diseases. If this is true, we have moved one step closer to better understanding the complex world of neurological illnesses.

Categories: Science

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s