Artificial Intelligence Model Can Detect Parkinson’s From Breathing Patterns

07/09/2022

A new artificial intelligence model can detect Parkinson’s disease by observing a person’s breathing patterns. The algorithm can also determine the severity of Parkinson’s disease and track progression over time. While Parkinson’s is known as notoriously difficult to diagnose (as it relies on the appearance of motor symptoms such as stiffness, slowness, and tremors, but these symptoms often appear several years after the disease onset).

Now, Professors in the Department of Electrical Engineering and Computer Science (EECS) at MIT have developed a new artificial intelligence model that can detect Parkinson’s just from reading a person’s breathing patterns. The tool is a neural network, a series of connected algorithms that mimic the operations of how a human brain works, capable of determining Parkinson’s from breathing patterns that occur while sleeping. The neural network is also able to discern the severity of someone’s disease and track the disease’s progression over time.

For years, researchers investigated the potential of detecting Parkinson’s using neuroimaging and cerebrospinal fluid. But such methods are costly, invasive, and require access to specialized medical centers, making them unsuitable for frequent testing that could otherwise provide early diagnosis or continuous tracking of the disease’s progression. To accomplish this, a device was developed with the appearance of a home Wi-Fi router which emits radio signals, analyzes their reflections off the surrounding environment, and extracts the subject’s breathing patterns without any physical contact. The breathing data is then fed to the network to assess Parkinson’s in a passive manner, with zero effort necessary from the patient and caregiver.