Groundbreaking AI program detects Alzheimer’s disease six years before human doctors

AI detects Alzheimer's disease

Typically, when a new development in the artificial intelligence field makes headlines, the news isn’t good. In recent months, AI-powered tech has falsely labeled Chinese executives as criminals, been named as a cause for rising unemployment, and accidentally sprayed Amazon workers with bear mace.

However, researchers recently created an application for machine learning that should be a cause for celebration, not anxiety. Dr. Jae Ho Sohn, a resident at the University of California, San Francisco’s Department of Radiology and Biomedical Imaging, used an AI algorithm to detect Alzheimer’s disease years before it would otherwise be diagnosed.

In developing his program, Dr. Sohn fed his algorithm thousands of records from UCSF patients who have had memory impairment symptoms. In particular, he had the AI analyze patient data pertaining to the amount of glucose their brains consumed as declining glucose consumption is an indicator of Alzheimer’s.

Manage your supply chain from home with Sourcengine

But since glucose consumption levels vary only slightly as the disease progresses, its early development was difficult to track by medical professionals.

But with its ability to detect granular shifts in glucose levels, the algorithm was able to positively identify Alzheimer’s in patients six years before traditional diagnoses. As the effects of the disease can be mitigated if detected early, Dr. Sohn’s algorithm has the potential to improve the quality of life for millions of people worldwide.

How AI is Shaping the Future of Medicine

Although various governments and corporations have used AI tech to optimize surveillance and distribution, its true potential might be in the medical sector.

Alphabet Inc., Google’s parent company, launched a life-sciences division called Verily early in 2015. By 2016, it was already profitable thanks to sales of groundbreaking products like smart contact lenses that check the glucose levels of diabetes sufferers. In 2018, the company announced that it had developed an AI program to predict the onset of heart disease with an eye test.

Similarly, Microsoft and Amazon have both made significant investments in developing AI that can be used to improve patient care. And while the notion of machine learning solutions being a standard part of healthcare may seem like a distant development, it may become commonplace sooner rather than later.

In June, a British Medtech startup called Babylon Health developed an AI program that outperformed human medical students on English medical diagnosis and treatment exams.

Medical AI and the Component Industry

Even though the AI-powered technologies being researched by various tech giants have the potential to save millions of lives, it’s become a sector of interest due to its extremely lucrative nature. In the United States alone, the healthcare industry is a near $3 trillion market. As such, healthcare could be the source of the next great tech boom.

In addition to machine learning experts, hardware manufacturers, and medical researchers, the integration of AI into healthcare also represents a significant market opportunity for component suppliers.

Because, as experts have pointed out, if the health of 7 billion people is dependent on artificial intelligence-based solutions, a whole lot of specialized AI chips will need fabrication to meet demand.