25 Oct 2023
Device using specific wavelengths could benefit all patients regardless of skin tone.UCSD) has developed a smartphone attachment that images a person's pupil to screen for a variety of neurological conditions.
Published in Scientific Reports, the research could improve both the accessibility of neurological screening procedures and their equity.
Tracking how pupil size changes during certain tasks can give an indication of a subject's neurological functions, but is known to be more difficult in people with dark eye colors and darker skin tones, since conventional color cameras struggle to distinguish the pupil from the surrounding dark iris. Normal pupillometers are also very expensive.
The UCSD device has been designed to work effectively regardless of skin color and be suitable for home use using a smartphone camera, potentially offering significantly expanded access to pupil measurements for health screening purposes.
"The key to accurate pupillometry is capturing images in which the pupil can be clearly differentiated from the iris," commented the UCSD team in its paper. "However, segmenting the pupil and the iris is difficult in visible light, especially for individuals with darker irises."
Near-IR cameras can carry out pupil measurements, but are a rarity on smartphones other than high-end models, while most phones have built-in filters that deliberately cut off light outside the visible spectrum to improve image quality for photography.
UCSD's solution is to use filtered far red wavelengths from within the visible spectrum, to significantly differentiate the pupil and the iris even with the phone's on-board RGB camera.
An imaging solution for everyone
"There has been a large issue with medical device design that depends on optical measurements ultimately working only for those with light skin and eye colors, while failing to perform well for those with dark skin and eyes," said Edward Wang from the UCSD Digital Health Technologies Lab.
"By focusing on how we can make this work for all people while keeping the solution simple and low cost, we aim to pave the way to a future of fair access to remote, affordable healthcare."
The UCSD prototype clips onto a smartphone and contains LEDs plus a longpass filter, to ensure that images contain only the target spectrum of 630 to 700 nanometers. The pupil's response to bright stimulus from the phone's flashlight is videoed, with a machine learning model using the recording to assess the relative change and percentage reduction in pupil size.
Trials on 12 volunteers with a range of eye colors validated results from the new device against a standard clinical pupillometer.
UCSD's next steps will include adding eye tracking facilities to account for eye movements, and moving towards large-scale neurological screenings in at-home environments. Billion Labs Inc, a UCSD spin-out developing health monitoring solutions built around the sensors and computational capability of smartphones, will refine and commercialize the new device.
"We created an inexpensive and fair solution to provide these kinds of emerging neurological screenings regardless of the smartphone price, make or model," commented UCSD's Colin Barry, co-founder and CTO of Billion Labs Inc.