Virtual reality system based on Kinect for the elderly in fall prevention.

Background: Daily life movements require balance ability. Good balance control is closely related to body stability and its development. Therefore, balance training is necessary for any age group.

Objective: This study proposes the combination of Kinect and virtual reality to build an information platform of interactive scenarios, for practice and evaluation of balance ability. Real-time monitoring of SpO_2, pulse rate, velocity and reaction time during the training process is presented for the training activities of elderly physical function.

Methods: Based on the indicators of balance ability, this information platform sets out various training activities to improve balance ability, making the supposedly tedious process fun and vivid and leading to much better training results.

Results: The data (SpO_2, pulse rate, velocity, reaction time) collected from this platform can be sorted and analysed, and the results used to evaluate the performance of balance training, and referenced for follow-up planning in the future. The real-time pulse rate and SpO_2 measurement information indicating the training activities for the elderly to maintain physical function has a positive significance. A noninvasive and unconstrained real-time method to detect the pulse rate and SpO_2 during exercises is presented. The results of balance assessment scale testing of BBS and TUG for the experimental group show that effective balance really improved. The difference between the experimental group and the control group was achieved by using paired t-test. The data were analysed by the descriptive statistics on significant level of P < 0.01.

Conclusions: From the training results of the participants, we know that the information platform developed for enhancing balance ability is in line with practical needs. Furthermore, the fun and interesting game-like exercises it introduces are very helpful in improving balance ability, and certainly in preventing falls.