Effectiveness of using a Kinect-based Rapid Movement Therapy (RMT) for fall prevention in chronic stroke survivors
Other conference paper


摘要INTRODUCTION: Community-dwelling stroke patients have higher fall prevalence than healthy older adults, possibly due to their impaired capacity to perform rapid balance-recovery reactions [1]. These reactions involve ballistic limb movements to rapidly expand the base of support to prevent falls. Stroke-induced instability often lead to fear of falling, reduced physical activity, and further impairment in limb functions. Traditional stroke rehabilitation programs often focus on improving the paretic limbs’ range-of-motion (ROM) and muscle strength, which neglected the importance of movement speed in preventing falls. Furthermore, these services often require the presence of clinical personnel, which place enormous financial burden on the healthcare system [2]. Moreover, it would be difficult for the training staffs to track the progress of all of their patients objectively and continuously. Therefore, the objective of this study is to develop a 3D interactive Kinect-based Rapid Movement Therapy (RMT) programme to train rapid limb movements in chronic stroke patients for fall prevention. Such system is capable of automatically tracking and providing real-time feedback on the ROM and movement speed of its users. Effectiveness of the system will be evaluated using a randomized controlled trial study. It is hypothesized that Kinect-based RMT can match/exceed the benefits of conventional training in restoring upper/lower limb motor performance and balance/gait control, leading to improved balance confidence and fewer falls.

METHODS: Community-dwelling chronic stroke patients with some limb and balance impairment are randomly assigned to receive twenty 1-hour sessions of either RMT or conventional balance training (control) in seven weeks. Subjects in the RMT group are required to stand in front of a Microsoft Kinect motion sensor and respond as quickly as possible by reaching/stepping in various directions as prompted by a display screen in front of them. Real-time feedback on their performance on their ROM and response time will be given as a motivator to promote motor relearning. All subjects will be evaluated before and after the training period.

RESULTS: Post-assessments of only three subjects have been completed; results from ~20-30 subjects will be presented in June. Preliminary results suggest that Kinect-based RMT may be effective in improving the participant’s upper/lower limb functions (FMA-UE: 20→25 vs. 29→30.5; FMA-LE: 11→13 vs. 20.5→17.5), balance (BBS: 49→52 vs. 51→51.5) and balance confidence (ABC: 81%→88% vs. 76%→83%) in comparison to the two control subjects.

CONCLUSIONS: If proven effective, this affordable and automated training system can be put into routine rehabilitation program in home/clinical settings to reduce fall risk among stroke patients.

ACKNOWLEDGEMENTS: Health and Medical Research Fund (grant #12131911)

1. Mansfield A, et al. Phys Ther. 91:958-969, 2011.
2. Godwin KM, et al. Top Stroke Rehabil. 18:676-684, 2011.
著者Cheng K.C., Junata M., Man H.S., Lai W.K., Tong K.Y.
會議名稱2017 ISPGR World Congress
會議地點Fort Lauderdale

上次更新時間 2020-22-10 於 10:12