Validating and calibrating the Nintendo Wii balance board to derive reliable center of pressure measures

Abstract

The Nintendo Wii balance board (WBB) has generated significant interest in its application as a postural control measurement device in both the clinical and (basic, clinical, and rehabilitation) research domains. Although the WBB has been proposed as an alternative to the "gold standard" laboratory-grade force plate, additional research is necessary before the WBB can be considered a valid and reliable center of pressure (CoP) measurement device. In this study, we used the WBB and a laboratory-grade AMTI force plate (AFP) to simultaneously measure the CoP displacement of a controlled dynamic load, which has not been done before. A one-dimensional inverted pendulum was displaced at several different displacement angles and load heights to simulate a variety of postural sway amplitudes and frequencies (<1 Hz). Twelve WBBs were tested to address the issue of inter-device variability. There was a significant effect of sway amplitude, frequency, and direction on the WBB's CoP measurement error, with an increase in error as both sway amplitude and frequency increased and a significantly greater error in the mediolateral (ML) (compared to the anteroposterior (AP)) sway direction. There was no difference in error across the 12 WBB's, supporting low inter-device variability. A linear calibration procedure was then implemented to correct the WBB's CoP signals and reduce measurement error. There was a significant effect of calibration on the WBB's CoP signal accuracy, with a significant reduction in CoP measurement error (quantified by root-mean-squared error) from 2-6 mm (before calibration) to 0.5-2 mm (after calibration). WBB-based CoP signal calibration also significantly reduced the percent error in derived (time-domain) CoP sway measures, from -10.5% (before calibration) to -0.05% (after calibration) (percent errors averaged across all sway measures and in both sway directions). In this study, we characterized the WBB's CoP measurement error under controlled, dynamic conditions and implemented a linear calibration procedure for WBB CoP signals that is recommended to reduce CoP measurement error and provide more reliable estimates of time-domain CoP measures. Despite our promising results, additional work is necessary to understand how our findings translate to the clinical and rehabilitation research domains. Once the WBB's CoP measurement error is fully characterized in human postural sway (which differs from our simulated postural sway in both amplitude and frequency content), it may be used to measure CoP displacement in situations where lower accuracy and precision is acceptable.

Figures

Figure 1.

Experimental setup to measure simulated one-dimensional postural sway. (A) The Nintendo Wii balance board (WBB) mounted and centered on the AMTI force plate (AFP); (B) Four (6.8 kg) lead blocks positioned symmetrically on the mechanical system's base to stabilize the inverted pendulum during oscillation; (C) The experimental setup: the mechanical system was mounted and centered on the WBB, which was mounted and centered on the AFP. The mechanical system consisted of a single inverted pendulum supported by springs (15.1 kg), a (16.0 kg) load applied at the CoM height, h, and four lead blocks positioned on the base to stabilize the inverted pendulum during oscillation. The inverted pendulum was displaced at a specified angle, θi, and then released to oscillate in the AP direction. The mechanical system was rotated 90° to acquire one-dimensional sway in the ML direction.

Figure 2.

The Nintendo WBB. (A) Top-view of the WBB shows the usable surface; (B) Bottom-view of the WBB shows the four foot-pegs, located under each of the four corners of the WBB: top right (TR), top left (TL), bottom left (BL), and bottom right (BR). The four force sensors are housed in the four foot-pegs.

Figure 3.

A diagram of the WBB. Accurate X and Y dimensions are essential for accurate CoP calculations: X = 433 mm, Y = 238 mm.

Figure 4.

A simplified diagram of the experimental setup for x-direction CoP displacement. (NOTE: FAFPx′ = FAFPx since these values reflect the acceleration of the CoM, which is the same for both the WBB and AFP).

Figure 5.

The CoPWBB (in blue) and CoPAFP′ (in red) signals for the condition invoking the lowest frequency response and highest sway amplitude. The zoomed-in templates illustrate the WBB's CoP signal error: the difference (in mm) in CoP displacement (CoPWBB − CoPAFP′).

Figure 6.

An individual WBB's (WBB_4) CoP signal error (CoPWBB − CoPAFP′) is plotted for all sway amplitudes and in both the AP (A) and ML (B) directions.

Figure 7.

Effect of calibration on CoPWBB signals. All four plots contain three signals: The CoPWBB signal before calibration (in blue, solid line), the CoPWBB signal after calibration (CoPWBBcalib) (in green, dashed line), and the “gold standard” CoPAFP′ signal (in red, solid line).

Figure 8.

Effect of calibration on CoPWBB signal error measured by RMSEs. This figure shows the distribution of the RMSEs across all sway amplitudes and WBBs, in both the AP (A) and ML (B) directions, both before and after linear calibration of the CoPWBB signals. The three oscillation frequencies (ω) corresponding to the three load heights (h = 900, 1000 and 1100 mm) are 0.6, 0.5, and 0.4 Hz, respectively.

Figure 9.

Bland-Altman plots of a time-domain measure, mean velocity (MV), before and after linear calibration of the CoPWBB signals. Comparison of MV derived from both the CoPWBB and CoPAFP′ signals for every trial (12 WBBs, three load heights, three displacement angles per load height, per sway direction = 108 trials). The solid line represents the mean difference between measurements (MVWBBvs. MVAFP) and the dotted lines represent the 95% limits of agreement (±1.96 times the standard deviation of the mean difference).