Researchers at MIT have developed an AI system that can diagnose Parkinson’s disease and track its progression, simply by monitoring someone’s breathing patterns as they sleep. The device looks like an internet router and can be mounted on the wall in a bedroom. It emits radio waves and then a neural network analyzes the reflected waves to assess breathing patterns. Crucially, the technology may be able to assist in diagnosing Parkinson’s disease much earlier than many conventional techniques and it is highly convenient and non-invasive compared with traditional diagnostics. It may also be particularly beneficial in testing new treatments for Parkinson’s as a non-invasive method to monitor disease progression. Diagnosing Parkinson’s in a timely manner is difficult, since in most cases the diagnostic journey does not even start until motor symptoms, such as tremor and stiffness, have become apparent. However, in many cases, the disease may have begun years earlier, meaning that the chance for early intervention has been lost. Researchers have experimented with techniques such as neuroimaging or analyzing cerebrospinal fluid, but these approaches are invasive and inconvenient, particularly for repeated assessments to check disease progression. Now, a new method has the potential to change all this. The approach is completely non-invasive, and involves simply placing a device in the bedroom a patient sleeps in. A neural network then analyzes the patient’s breathing patterns as they sleep, which are known to be dysregulated in Parkinson’s. “A relationship between Parkinson’s and breathing was noted as early as 1817, in the work of Dr. James Parkinson,” said Dina Katabi, one of the developers of the new system. “This motivated us to consider the potential of detecting the disease from one’s breathing without looking at movements. Some medical studies have shown that respiratory symptoms manifest years before motor symptoms, meaning that breathing attributes could be promising for risk assessment prior to Parkinson’s diagnosis.” Aside from early diagnosis of Parkinson’s disease, the new technology is well-suited to allow clinicians to monitor disease progression without repeat visits to the clinic and expensive and invasive interventions. It may be very useful for clinical trials of Parkinson’s treatments, where disease progression or recovery will obviously be key outcomes. “In terms of drug development, the results can enable clinical trials with a significantly shorter duration and fewer participants, ultimately accelerating the development of new therapies,” said Katabi. “In terms of clinical care, the approach can help in the assessment of Parkinson’s patients in traditionally underserved communities, including those who live in rural areas and those with difficulty leaving home due to limited mobility or cognitive impairment.” Source
