Sign In

Communications of the ACM

ACM TechNews

A Smartphone, New Software Could Help Save Infants Born Preterm

View as: Print Mobile App Share:
An ophthalmoscope attached to the lens of a smartphone (left), used to estimate the gestational age of newborns.

A new automated algorithm uses images acquired with an ophthalmoscope attached to the lens of a smartphone to estimate gestational age of newborns.

Credit: Biomedical Optics Express

Duke University researchers developed an automated method for analyzing video of the eye to determine the gestational age of newborns, an innovation that could be crucial in saving the lives of premature infants.

Duke's Arjun D. Desai said the method uses a fully automated machine learning algorithm to analyze video obtained via an inexpensive ophthalmoscope attached to the lens of a smartphone camera, which acquires video of blood vessels within the eye's anterior lens capsule; a special algorithm parses the video to identify the highest-quality video frames showing the region of interest for analysis.

Computational techniques like convolutional neural networks and machine learning are then applied to assess image features in the region and calculate gestational age.

Said Desai, "Our work is a first step to developing a fully automatic pipeline for determining gestational age that is accurate and robust to differences across newborns."

From Optical Society of America
View Full Article


Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account