A Mount Sinai-led team of researchers has enhanced an artificial intelligence (AI)-powered algorithm to analyze video recordings of clinical sleep tests, ultimately improving accurate diagnosis of a common sleep disorder affecting more than 80 million people worldwide. The study findings were published in the journal Annals of Neurology on January 9.
REM sleep behavior disorder (RBD) is a sleep condition that causes abnormal movements, or the physical acting out of dreams, during the rapid eye movement (REM) phase of sleep. RBD that occurs in otherwise healthy adults is called “isolated” RBD. It affects more than one million people in the United States and, in nearly all cases, is an early sign of Parkinson’s disease or dementia.
RBD is extremely difficult to diagnose because its symptoms can go unnoticed or be confused with other diseases. A definitive diagnosis requires a sleep study, known as a video-polysomnogram, to be conducted by a medical professional at a facility with sleep-monitoring technology. The data are also subjective and can be difficult to universally interpret based on multiple and complex variables including sleep stages and amount of muscle activity. Although video data is systematically recorded during a sleep test, it is rarely reviewed and is often discarded after the test has been interpreted.
Previous limited work in this area had suggested that research-grade 3D cameras may be needed to detect movements during sleep because sheets or blankets would cover the activity. This study is the first to outline the development of an automated machine learning method that analyzes video recordings routinely collected with a 2D camera during overnight sleep tests. This method also defines additional “classifiers” or features of movements, yielding an accuracy rate for detecting RBD of nearly 92 percent.
“This automated approach could be integrated into clinical workflow during the interpretation of sleep tests to enhance and facilitate diagnosis, and avoid missed diagnoses,” said corresponding author Emmanuel During, MD, Associate Professor of Neurology (Movement Disorders), and Medicine (Pulmonary, Critical Care and Sleep Medicine), at the Icahn School of Medicine at Mount Sinai. “This method could also be used to inform treatment decisions based on the severity of movements displayed during the sleep tests and, ultimately, help doctors personalize care plans for individual patients.”
The Mount Sinai team replicated and expanded a proposal for an automated machine learning analysis of movements during sleep studies that was created by researchers at the Medical University of Innsbruck in Austria. This approach uses computer vision, a field of artificial intelligence that allows computers to analyze and understand visual data including images and videos. Building on this framework, Mount Sinai experts used 2D cameras, which are routinely found in clinical sleep labs, to monitor patient slumber overnight. The dataset included analysis of recordings at a sleep center of about 80 RBD patients and a control group of about 90 patients without RBD who had either another sleep disorder or no sleep disruption. An automated algorithm that calculated the motion of pixels between consecutive frames in a video was able to detect movements during REM sleep. The experts reviewed the data to extract the rate, ratio, magnitude, and velocity of movements, and ratio of immobility. They analyzed these five features of short movements to achieve the highest accuracy to date by researchers, at 92 percent.
Researchers from the Swiss Federal Technology Institute of Lausanne (École Polytechnique Fédérale de Lausanne) in Lausanne, Switzerland contributed to the study by sharing their expertise in computer vision.
