A new stroke rehabilitation computer program uses sensors to recognize and count arm movements with 77% accuracy, according to its creators. Once refined and tested further, they plan to offer the program for free to rehab experts.
The tool, called PrimSeq, was developed by New York University and uses sensors attached to the arms and back to capture upper body movements in three dimensions. More than 51,000 stroke patients’ upper body movements were used to create the program, which was then tested on a group of eight patients.
Loss of limb mobility is a common consequence of strokes. The study authors, from NYU Grossman School of Medicine, said PrimSeq offers much-needed standardized measures of the precise amount of exercise each patient receives.
Correctly identifying stroke counts is key to achieving ideal stroke rehabilitation outcomes, they said.
“Such help is desperately needed because counts from video recordings or other wearable sensors do not offer standardized measures of the precise amount of rehabilitation exercise each patient receives,” said Heidi Schambra, MD, from the school’s department of rehabilitation medicine, in a statement. “Any improvement in exercise dose received must be based on accurate, automated measurements of the type and number of arm movements involved in a given exercise.”
The research team aims to refine their computer model and develop a smaller device that can be worn on the arm and upper body. If additional experiments are successful, they plan to test it in clinical trials.
“I would argue that our tool holds great promise for clinical use because the alternative is not to have accurate counts at all,” Schambra said.
The researchers shared the data used to build the computer program online.
Sitting tai chi improves mental health functional recovery in ischemic stroke
Stroke Patients See Significant Survival Increase with Modified Cardiac Rehabilitation