https://wwwfr.uni.lu/snt/news_events/best_paper_award_for_research_on_stroke_rehabilitation
Publié le lundi 09 avril 2018
A team of researchers at the University of Luxembourg’s Interdisciplinary Centre for Security, Reliability and Trust (SnT) has developed a new system for supporting stroke survivors in their rehabilitation and recovery. Their framework uses low-cost depth sensors, such as Microsoft’s Kinect, to monitor and correct posture, providing patients with real-time feedback from the comfort of their own home.
Globally, stroke is the second leading cause of death, and survivors are often left with long-term problems caused by injury to the brain. Its estimated cost is €65 billion per year in Europe, and yet 90% of strokes or secondary stroke events are preventable with appropriate management of risk factors. Physical exercise in particular is key to both secondary stroke prevention and regaining strength, coordination and movement after a stroke.
Consistent repetition is essential, and yet post-stroke patients often have difficulty performing their exercises at home without specialist supervision. For example, in order to complete a movement such as raising an arm to a certain height, patients might compensate by bending their spine, raising a shoulder or lifting an ankle.
To support stroke survivors in self-managing their physical recovery, the Signal Processing and Communications Research Group’s Computer Vision team at SnT, led by Dr. Djamila Aouada, has developed a system that measures a patient’s posture during exercise, detecting undesirable movements that might slow their recovery or even cause injury. The system monitors the position of their back and the balance between left and right limbs. It gives on-screen real-time feedback, with a range of colours from green to red signifying good or bad posture and identifying the location of the problem.
They presented their work, Flexible Feedback System for Posture Monitoring and Correction, at the 2017 IEEE Fourth International Conference on Image Information Processing (ICIIP 2017) in Shimla, India, and recently received the Computer Vision track’s Best Paper Award.
“Fatigue, and low motivation, confidence and skill levels are all big factors behind why stroke survivors do not exercise as regularly as they need to,” says Renato Baptista, PhD Candidate at SnT and lead author of the paper. “This means that we need to do more to enable them to do their exercises at home without the frustration that comes from being unsure of the accuracy of what they are doing. Given the huge amount already spent on stroke treatment and the fact that stroke incidence is increasing, there’s a clear need for affordable self-management solutions to support medical experts in their work.”
The work, which is part of the European project STARR (Decision SupporT and self-mAnagement system for stRoke survivoRs), also has potential applications for daily life. “Many of us are very sedentary, sitting at desks for long periods, unaware of what poor posture is doing to our bodies,” continues Baptista. “And when we are active, for example at the gym or playing a musical instrument, expert feedback on posture isn’t always available. This is where human motion analysis can play a big role in improving our quality of life.”
No comments:
Post a Comment