an AI could predict disease via breathing

In France, more than 200,000 people are affected by Parkinson’s disease according to the France Parkinson association.

This disease is a neurodegenerative disease characterized by the destruction of a specific population of neurons involved in movement control. These are dopamine neurons.

Symptoms of the disease are characterized by slowness in movement, rigidity and tremor at rest. So far, the causes of the disease remain unknown. In addition, doctors are only able to make the diagnosis when the symptoms appear.

Parkinson’s: detecting the disease before symptoms appear

Today, the diagnosis of Parkinson’s disease is difficult to make because there are no easy-to-study biomarkers that can testify to the onset of the disease or its evolution.

It is on the basis of this problem that a team of researchers from MIT (Massachusetts Institute of Technology) decided to study breathing during sleep in patients. Dina Katabi, one of the scientists of the study, specifies that: “certain medical studies have shown that respiratory symptoms appear years before the motor symptoms. Thanks to the development of this new artificial intelligence tool, doctors will be able to detect the disease early by studying the breathing of patients.

The researchers tested this new artificial intelligence on more than 7,000 patients, including 757 with Parkinson’s. The tool is a series of algorithms that mimic the functioning of a human brain. It thus allows to diagnose disease but also to assess its severity and progression by analyzing the person’s breathing during sleep.

The device records a person’s breathing patterns during sleep, effortlessly patient or caregiver. This differs from current methods of detecting the disease.

Thus the algorithm developed by MIT could make it possible to detect disease long before the first symptoms appear.