Computer chip powerhouse Nvidia this week released its Clara platform, a hardware/software combo with powerful algorithmic capabilities, based on a ground-breaking computing architecture that could have real impact on how medical professionals identify disease.
The release addresses what Nvidia describes as a “fundamental disconnect” between legacy medical instruments — which typically have a lifespan of over 10 years — and their ability to run modern applications, which benefit from the 1,000x acceleration of GPU computing over the past decade.
What is the NVIDIA Clara Platform?
The Clara Platform is a bleeding edge computing ecosystem based on some of the tech firms premier computing power. It was created primarily to address a great challenge of today’s medical appliances, namely crunching gigabytes of dynamically generated data by the second for use by medical professionals in delivering healthcare to patients.
More than 400 AI healthcare startups have launched in the past five years, and NVIDIA is positioning the Clara platform as the package best able to help them harness AI to transform healthcare workloads. The platform is already being used to speed up MRI image processing, turn 2D images into 3D and more.
Processing the massive sea of data is the greatest challenge of today’s medical instruments. NVIDIA CLARA AGX will dramatically boost the capabilities of legacy medical devices, enabling scaling from entry level tools to the most demanding 3D instruments: https://t.co/0uBkY152Aw pic.twitter.com/jYrbxPPTVh
— NVIDIA AI (@NVIDIAAI) September 13, 2018
Achieving this level of supercomputing has traditionally required three computing architectures: FPGAs, CPUs and GPUs.
Clara AGX simplifies this to a single, GPU-based architecture that delivers the world’s fastest AI inferencing on NVIDIA Tensor Cores; acceleration through CUDA, the world’s most widely adopted accelerated computing platform; and NVIDIA RTX graphics. It can scale from entry-level devices to the most demanding 3D instruments.
Software Developers Given Clara Platform SDK
Nvidia has also released an SDK (Software Developer Kit) for the platform. As NVIDIA’s VP of Healthcare, Kimberly Powell, put it in a release this week: “The Clara SDK provides medical-application developers with a set of GPU-accelerated libraries for computing, graphics and AI; example applications for reconstruction, image processing and rendering; and computational workflows for CT, MRI and ultrasound”.
Additionally, these libraries all “leverage containers and Kubernetes” which simply means that teams can scale up usage of these deployed applications with the latest in software deployment technology. If deployed in conjunction with the cloud, systems can be made for medical institutions of varying sizes and budgets.
“We are using AI to improve workflow for MRI and PET exams,” said Enhao Gong, founder of medical imaging specialists Subtle Medical. “NVIDIA’s Clara platform will enable us to… scale our technology to reduce risks from contrast and radiation.”
ImFusion, a member of NVIDIA’s accelerator programme, can create 3D ultrasound from a traditional 2D acquisition, and then visualize the ultrasound fused with CT. ImFusion developers ported their application to Clara in less than two days, the company said, and take advantage of Clara’s inferencing, cinematic rendering engine and virtualisation capability.
Where Can I Get my Hands on This?
According the firm, the Nvidia Clara Platform is available now to its early access partners, however there is a planned beta release for Q2 2019.
It will be very interesting to see the unleashed results of the physical and medical sciences which hopefully helps the world to advance yet again in fighting against disease and delivering optimal healthcare.