In a collaborative effort between the Massachusetts Institute of Technology (MIT) and other U.S. universities, scientists have demonstrated an AI system that can detect Parkinson's disease based purely on a person's breathing patterns. Parkinson's disease is notoriously difficult to diagnose due to its dependence on the appearance of motor symptoms such as tremors, stiffness, and sluggishness, but these symptoms typically appear several years after the commencement of the disease. The tool may assess whether a person has Parkinson's disease based on their nocturnal breathing patterns, which are the patterns of breathing that occur during sleep. Over the years, cerebrospinal fluid and neuroimaging have been studied as potential screening tools for Parkinson's disease. However, these techniques are invasive, expensive, and require access to specialised medical facilities, preventing them from being used in routine testing that would otherwise enable early diagnosis or ongoing disease monitoring. Researchers have demonstrated that an AI-based Parkinson's assessment may be performed each night at home while the patient sleeps and without physical contact. For this purpose, scientists created a device resembling a home Wi-Fi router, but instead of giving internet access, it emits radio signals, analyses how they are reflected off the surrounding space, and then, without physical touch, extracts the subject's breathing patterns. There is no effort necessary from the patient or caregiver, as the breathing signal is then passively transmitted to the neural network for Parkinson's assessment. The finding has major significance for the development of Parkinson's drugs and clinical care, and it can be broadened in the near future to detect other health problems.