According to new research out of McGill University in Montreal, Microsoft Kinect, a 3-D depth and motion sensing input device for use with the company’s Xbox 360 and Xbox One video game consoles and Windows PCs, could be a cheap, effective, and easy-to-use tool to evaluate walking gait difficulties in people with multiple sclerosis (MS).
A research team led by McGill postdoctoral fellow Farnood Gholami, and supervised by Jozsef Kovecses of the Mechanical Engineering department and Centre for Intelligent Machines, collaborated with Daria Trojan, a physiatrist with the McGill Department of Neurology and Neurosurgery, in determining whether the Kinect could accurately and consistently detect differences in walking gait in MS patients compared to healthy persons.
A paper reporting the research findings, “A Microsoft Kinect-Based Point-of-Care Gait Assessment Framework for Multiple Sclerosis Patients“, was published in the July 2016 edition of the IEEE Journal of Biomedical and Health Informatics. Based on these findings, the investigators established the potential feasibility of using a Microsoft Kinect camera in a clinical setting.
The researchers noted in a McGill news release that the current clinical practice is for patients’ walking gait to be assessed subjectively by their doctors, which carries the potential for distorted or inconsistent evaluations (for instance, two different clinicians may arrive at conflicting evaluations). Use of a camera engineered to detect movement and computer algorithms that quantify walking patterns provides objectivity and a reduced potential for human error.
In the study, Gholami and colleagues used the Kinect device to capture and record walking movement in 10 MS patients previously assessed for gait abnormalities using traditional methods. As controls, 10 age-and-sex-matched individuals were also evaluated.
Using the data collected, the research team developed computer algorithms to quantify gait characteristics of both groups. Researchers found that the Kinect camera’s measurements of gait characteristics, when analyzed with the algorithms, were reproducibly consistent.
Gait characteristics of MS patients obtained by the algorithm were also correlated with clinical measures of gait, and it was determined that the algorithms could mathematically define MS gait characteristics at different severity levels, accurately determining the individual’s level of gait abnormality.
Gholami said in the McGill release that he first became interested in using motion capture technology for clinical purposes while he was a PhD student, but found the equipment available at the time to be very expensive, difficult to use, and non-portable, prohibiting its broad clinical application. However, Microsoft’s development and commercial release of Kinect as a relatively inexpensive consumer product provided him a highly portable tool that appeared to be sufficiently accurate for use in gait assessment. As such, it could help clinicians better diagnose gait pathology, and potentially also used to more accurately evaluate whether a prescribed medication is effective in improving a patient’s gait.
“Our developed framework can likely be used for other diseases causing gait abnormalities as well, for instance Parkinson’s disease. The next step is to conduct a study with a larger group of MS patients including evaluation in a gait laboratory, and using a newer version of the Kinect device that promises to improve accuracy,” Gholami said.
Trojan notes the tool could also be useful in assessing interventions, such as rehabilitation or medication, or to document MS progression as reflected in gait deterioration, and may also be a useful tool in clinical trials.
This research was made possible with funding from the Natural Sciences and Engineering Research Council of Canada.
IEEE Journal of Biomedical and Health Informatics