NVIDIA Broadens Clara Platform

NVIDIA remains at the forefront of developing key artificial intelligence systems for radiology with their latest launch of the Clara Software Development Kit (SDK), which will enable third-party developers to build enhanced imaging applications with a series of accelerated libraries.

The SDK is an expansion of NVIDIA’s Clara Platform, a GPU-based system comprised of both computing architecture and software development, which was introduced back in September as a part of its Project Clara initiative. Project Clara’s objective is to help computing devices to operate in synchrony by expediting image quality and speed in MRI, ultrasound, and CT scanners.

“Every radiology practice has its own instruments, has its own patients, demographics and its own way of practice,” said Abdul Hamid Halabi, Global Business Development Lead at NVIDIA. “So although we’re seeing a flood of new algorithms come through, we do believe we need to provide radiologists with the tools to take those AIs and localize them for their own patients. It’s going to help most radiologists unlock the value of the data that they’re sitting on today.”

In conjunction with SDK, NVIDIA will also be offering a Transfer Learning Toolkit, a program that’s expected to roll out in 2019 and allows developers to customize pre-trained models to best fit a hospital or clinic’s needs by integrating partial patient data into a larger algorithm through AI-assisted annotation.

NVIDIA is partnering with both the Wexner Medical Center at The Ohio State University (OSU) and the National Institutes of Health (NIH) to implement imaging tools and algorithm services for clinical purposes. At Wexner, NVIDIA is building the “first in-house marketplace for clinical medical imaging in the U.S.,” according to MedCity News. It will feature an app store with a variety of imaging algorithms including ones that can detect brain hemorrhage or coronary artery disease. At the NIH, AI will be used to accelerate research for brain and liver cancer treatments and creating systems that integrate imaging, genomic, and clinical data for more precise clinical practices.