CIDER: Enabling Robustness-Power Tradeoffs on a Computational Eyeglass

Mobile Sensing of Alertness, Sleep and Circadian Rhythm

Human biology is deeply rooted in the daily 24-hour temporal period. Our biochemistry varies significantly and idiosyncratically over the course of a day. Staying out of sync with one's circadian rhythm can lead to many complications over time, including a higher likelihood for cardiovascular disease, cancer, obesity, and mental health problems [1]. Constant changes in daily rhythm due to shift work has been shown to increase risk factors for cancer, obesity, and Type 2 diabetes. Moreover, the advent of technology and the resultant always-on ethos can cause rhythm disruption on personal and societal levels for about 70% of the population [2].

Circadian disruption can also cause a serious deficit in cognitive performance. In particular, alertness - a key biological process underlying our cognitive performance - reflects circadian rhythms [3]. Sleep deprivation and circadian disruption can result in poor alertness and reaction time [3]. The decline in cognitive performance after 20 to 25 hours of wakefulness is equivalent to a Blood Alcohol Concentration (BAC) of 0.10% [4]. To compare, in New York State, a BAC of more than 0.05% is considered "impaired" and 0.08% is considered "intoxicated" [5]. In other words, the effects of sustained sleep deprivation and circadian disruption on cognitive performance is similar (or worse) to being intoxicated.

iLid: Low-power Sensing of Fatigue and Drowsiness Measures on a Computational Eyeglass

The ability to monitor eye closures and blink patterns has long been known to enable accurate assessment of fatigue and drowsiness in individuals. Many measures of the eye are known to be correlated with fatigue including coarse-grained measures like the rate of blinks as well as fine-grained measures like the duration of blinks and the extent of eye closures. Despite a plethora of research validating these measures, we lack wearable devices that can continually and reliably monitor them in the natural environment. In this work, we present a low-power system, iLid, that can continually sense fine-grained measures such as blink duration and Percentage of Eye Closures (PERCLOS) at high frame rates of 100fps. We present a complete solution including design of the sensing, signal processing, and machine learning pipeline; implementation on a prototype computational eyeglass platform; and extensive evaluation under many conditions including illumination changes, eyeglass shifts, and mobility. Our results are very encouraging, showing that we can detect blinks, blink duration, eyelid location, and fatigue-related metrics such as PERCLOS with less than a few percent error.