Nadezhda Korostyleva

← Selected work

Wireless EOG driver drowsiness monitor

Blink rate is a validated marker of sleepiness. A wearable that measures it can warn a driver before the driver notices.

Junior design, University of South Florida
April to May 2024
Top three of fifteen teams, 2024 Florida Medical Manufacturing Consortium

The idea

Electrooculography measures the corneo-retinal potential, so eye movement and blinks appear directly in the signal. Blink frequency is a clinically validated sleepiness marker, which makes it a reasonable thing to monitor continuously in a driver — the measurement is unobtrusive and the signal is large enough to work with outside a lab.

images/eog-device.jpg
The wearable itself, ideally worn or next to something for scale.
The EOG wearable.

What we built

The device acquires the EOG signal, extracts blink events, and streams the result over Bluetooth Low Energy to a paired phone application. I built the app and the full BLE communication layer in C++, with an interface designed to be readable at a glance, since anything demanding attention defeats the purpose of a drowsiness warning.

images/eog-app.png
Screenshot of the mobile app interface.
Companion application showing live blink data over BLE.
images/eog-signal.png
A plot of raw EOG with detected blinks marked, if you still have the data. This is the figure that proves the device worked.
EOG trace with detected blink events.

We presented the project at the 2024 Florida Medical Manufacturing Consortium, where it was selected as one of the top three of fifteen teams in the cohort.

Next: hardware workshops →