A new device can detect Parkinson’s disease only from respiratory models. The device, developed by a team led by Dina Katabi of MIT, uses a neural network to discern the presence and severity of neurological disease.
Parkinson’s disease is famous for being diagnosed because it is mainly dependent on the appearance of motor symptoms such as tremors, stiffness, and slowness, but these symptoms often appear several years after the onset of disease. Now, Dina Katabi, Professor Thuan and Nicole Pham in the Department of Electrical Engineering and Computer Science in MIT and the main investigator at MIT Jameel Clinic, and his team has developed an artificial intelligence model that can detect Parkinson only from reading a person’s breathing pattern from someone who is breathing somebody. . The tool in question is a nerve tissue, a series of connected algorithms that mimic the way the human brain works, able to assess whether a person has Parkinson’s from their nocturnal breathing – that is, the pattern of breathing that occurs during sleep. Networks, which are trained by Mit PhD Yuzhe and Postdoc Yuan Yuan students, can also see the severity of a person’s Parkinson’s disease and track the development of their disease from time to time. Which is the first writer in a new paper to describe the work, published by Today in Nature Medicine. Katabi, which is also an affiliate of MIT computer science and artificial intelligence laboratory and Director of the Center for Wireless Network and Cellular Computing, are senior writers. They joined Yuan and 12 colleagues from Rutgers University, University of Rochester Medical Center, Mayo Clinic, Massachusetts General Hospital, and Boston University College of Health and Rehabilition. Over the years, researchers have investigated the potential for detecting Parkinson’s using cerebrospinal fluid and neuroimaging, but such methods are invasive, expensive, and require access to special medical centers, making them not suitable for testing that can often provide early diagnosis or sustainable tracking from disease development. MIT researchers show that the assessment of Parkinson’s artificial intelligence can be done every night at home when the person is sleeping and without touching their bodies. To do this, the team developed a device with the appearance of a home wi-fi router, but instead of giving internet access, this device radiates radio signals, analyzing their reflections from the surrounding environment, and extracting the subject’s breathing patterns without any contact body. The breathing signal is then fed to the nerve tissue to assess Parkinson’s passively, and there is no effort needed from the patient.
posted on: DeFiance