Medical Marvel: Duke’s GPU-Powered Imaging Wins $150,000 Global Impact Award

by Tonie Hansen

Duke University researchers have received NVIDIA’s $150,000 Global Impact Award for their work using GPU-powered imaging techniques that give surgeons a 3D, stereoscopic live feed while performing highly sensitive ocular microsurgery.

Nearly 30 applications from 13 nations were submitted for the annual award, which recognizes researchers using NVIDIA technology for groundbreaking work that addresses social, humanitarian and environmental problems.

The Duke team, led by engineering professor Joseph Izatt, are using GPUs to transform how optical microsurgery is conducted. State-of-the-art microscopes produce 3D images of the eye every few microseconds to give surgeons immediate feedback during operations.

Global Impact Award presented to Dr. Joseph Izatt (center)
Global Impact Award presented to Dr. Joseph Izatt (center)

Far Surpassing Standard Practice

Traditionally, optical microsurgery has required the patient get a pre-op scan, generating images the surgeon would use to map out disease and plan surgery.

Then, there’d be a post-op scan to ensure the operation was a success.State-of-the-art microscopes use optical coherence tomography (OCT), an advanced imaging technique that produces 3D images in five to six seconds.

Izatt’s work goes much further. It takes complete 3D volumetric images that are updated every tenth of a second and rendered from two different angles. The result is a real-time stereoscopic display into both microscope eyepieces. Rather than doing pre- and post-operative images to gauge their success, surgeons can have immediate feedback as they operate.

This process means surgeons “are slicing and dicing the data in real time,” Izatt said during a presentation at NVIDIA’s GTC 2016.

The Duke team uses our GeForce GTX TITAN Black GPU, CUDA programming libraries and 3D Vision technology to power their solution. It’s already been deployed in more than 90 surgeries at the Duke Eye Center and the Cleveland Clinic Cole Eye Institute, and could be available for commercial use within a couple years.

Honorable Mention

NVIDIA gave honorable mention to reseachers at  Imperial College London, where a team of researchers is using computational methods to help diagnose traumatic brain injuries. Led by Ben Glocker, a lecturer at the college’s computing department, the team is deploying sophisticated image analysis tools, powered by GPUs and deep learning.

NVIDIA invites submissions for the 2017 award through the end of October.

More Global Impact Award 2016 nominees:

GPUs Help Monitor Rising Sea Levels with Pinpoint Accuracy

How Haiti’s Earthquake Inspired New Ways to Map Structural Safety Using GPUs

How GPUs Are Helping Map Worldwide Poverty