Modeling balance control during sustained waking allows posturographic sleepiness testing.

Abstract

We develop a method to quantify sleepiness. Sleepiness is a major risk factor in traffic and occupational accidents, but lack of convenient tests precludes monitoring impending sleepiness. Posturographic balance testing could address this need because sleepiness increases postural sway. It is, however, unclear how sleepiness influences balance control. Our results, for 12 subjects, show that balance control is more susceptible to increasing time awake (TA) compared to neuromuscular processes. This conclusion is reached since during sustained waking the control process slows down by 3.4% per hour of increased TA. This slowdown accounts for 65% of the variance in diurnal balance. We quantified balance control by modeling the body as an inverted pendulum and by expressing the control as the critical time interval for open-loop control (Deltat(c)) of the center-of-mass movements of this pendulum. To estimate the subjects' TA, we regressed the Deltat(c) scores recorded during sustained waking against increasing TA, and equated separate Deltat(c) test scores with the diurnal Deltat(c) scores. We estimated TA with 68% positive predictive value. The results encourage implementing balance modeling into a device that performs clinical or industrial balance testing because the model-based Deltat(c) score responded to increasing TA.