Evaluating sleepiness using force platform posturography

Abstract

We have investigated the feasibility to use posturography as a method to estimate sleep deprivation. This manuscript presents a proof-of-concept of this idea. Twenty-one healthy subjects aged 20-37 years participated in the study. The subjects were deprived of sleep for up to 36 h. Their postural stability was measured as a function of sleep deprivation time. As a reference the critical fusion frequency method for measuring sleepiness was used. The 163 posturographic parameters used for analyzing the posturographic data were found from the literature. Of these parameters, the fractal dimension of the sway path, the most common frequency of the sway, the time-interval for open-loop control of stance, and the most common amplitude of the sway showed the highest linear correlations with sleep deprivation time. Using these four parameters we were able to estimate the sleep deprivation time with an accuracy better than 5 h for 80% of the subjects.

Robust evaluation of time since awakening using force platform posturography

“…In addition, for direct comparison with previous findings reported by Haeggström et al (2006), we Figure 2. In general, the fractal dimension variable is a statistical measure of the complexity in the data and, using the definition employed by Haeggström et al (2006), it will be calculated as the slope of a straight line fitted to the first region (up to the critical time) in the stabilogram diffusion plot in a log-log scale. In comparison to the original proposition of the stabilogram diffusion plot applied to the COP analysis (Collins and De Luca, 1993), this calculation corresponds to two times the Hurst exponent in that region.…”

“…However, because they yielded similar results, we showed only results corresponding to the variable AREA, the area of the ellipse computed by Principal Component Analysis such that there is a 95% probability that a new observation will lie inside the ellipse (Oliveira et al, 1996). In addition, for direct comparison with previous findings reported by Haeggström et al (2006), we Figure 2. In general, the fractal dimension variable is a statistical measure of the complexity in the data and, using the definition employed by Haeggström et al (2006), it will be calculated as the slope of a straight line fitted to the first region (up to the critical time) in the stabilogram diffusion plot in a log-log scale.…”

“…This procedure resulted in four experimental conditions: RO, RC, FO and FC. Condition RO is the "normal" or "control" condition that has been tested in previous studies (Haeggström et al, 2006;Morad et al, 2007). The 8 trials of each data collection time point were performed in the following order: two RO trials, two RC trials, two FO trials and two FC trials.…”

“…It is thus an important traffi c safety issue to fi nd a robust, objective (as opposed to subjective reports such as the Stanford Sleepiness Scale) (Hoddes et al, 1973) and practical way to estimate the amount of time a person has been awake (time since awakening or TA). Posturography is an interesting candidate (Haeggström et al, 2006;Morad et al, 2007;Forsman et al, 2010;Smith et al, 2012;Tietavainen et al, 2013) because it yields objective measures and can be performed by means of cheap and easily transportable devices such as force platforms. Basically, posturography is the measurement of human posture (Winter, 1995).…”

“…However, simple measurements of the amount of sway, such as the standard deviation, area, and velocity of COP trajectories during conventional normal upright standing, are not always signifi cantly correlated with TA and novel measurements, such as fractal dimension and critical time from a stabilogram diffusion plot, have been proposed because they achieved a higher correlation with TA (Haeggström et al, 2006;Morad et al, 2007;Smith et al, 2012;Tietavainen et al, 2013).…”