New Pill Recognition Software Shows Promise

By Steve Tokar

Saturday, 2 AM.  A 911 dispatcher receives a frantic call: "I found my roommate passed out. I think she may have overdosed!" 

But on what? Imagine if, when first responders arrive on the scene, they point a cell phone camera at pills scattered near the victim, and within seconds, they can know what drugs are likely involved.

UCSF and its faculty are perfectly positioned to be contributing the great ideas that become standards a few years from now.
Clay Johnston, MD, PhD, on digital health innovations

“In overdose cases like this, where there is uncertainty as to the identity of the pill, a new application could provide real-time identification in the field or emergency room,” says Rahul C. Deo, MD, PhD, an assistant professor at UCSF’s Cardiovascular Research Institute.

Deo is collaborating with UCSF medical student David Ouyang to develop a mobile application that can rapidly identify pills based on color, shape, and imprint. The project has received a significant boost from the T1 Catalyst Mobile Health Translational Project Award offered by the Clinical and Translational Science Institute (CTSI) at UCSF.

Read more