App helps keep your pills straight

Summary:Researchers at the National Institutes of Health have developed a simple software that can identify pills from a camera phone image in under a second.

There are thousands of prescription pills, but only few colors and shapes to choose from.

To avoid dangerous mix-ups in hospitals and at home, Jesus Caban at the U.S. National Institutes of Health and colleagues have developed software that can identify a pill from a camera phone image, New Scientist reports.

Tools to identify pills on smartphones do exist – such as apps from and WebMD – and they help distinguish between pills using descriptions that you have to type in. In clinical settings, however, this might be too time consuming.

Caban’s new software extracts the shape, color, and imprint of a pill from a picture you can take on your phone.

The team tested the system on images of 568 prescribed pills, taken from different angles and in various light conditions.

The app identified the drugs with 91 percent accuracy, and within a second. (The team hopes accuracy will improve once the system learns to recognize a pill from a wider range of angles.)

And the technique is simple enough to become a smartphone app for use at home.

The work was presented at the International Conference on Image Processing last week.

[From New Scientist]

Image by .v1ctor. via Flickr

This post was originally published on

Topics: Innovation


Janet Fang has written for Nature, Discover and the Point Reyes Light. She is currently a lab technician at Lamont-Doherty Earth Observatory. She holds degrees from the University of California, Berkeley and Columbia University. She is based in New York. Follow her on Twitter.

zdnet_core.socialButton.googleLabel Contact Disclosure

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Related Stories

The best of ZDNet, delivered

You have been successfully signed up. To sign up for more newsletters or to manage your account, visit the Newsletter Subscription Center.
Subscription failed.