GPU processing and AI-enabled workflows have the potential to provide massive benefits to medical image analysis and diagnosis. XNAT and NVIDIA are collaborating to incorporate XNAT's study management and containerized processing capabilities with the NVIDIA Clara healthcare application platform, and are demonstrating the fruits of that collaboration at RSNA 2019.
The Clara suite of TensorFlow AI models can be trained to draw segmentations of organs and identify tumor tissue within those organs. XNAT has developed a series of integrations and tools that enables a containerized workflow where XNAT study data can be manually segmented by radiologists using a new "Rapid Reader" application, then organized into collections and used to configure and train models in TensorFlow. Trained models can then be deployed on new data.
"What excites me most about this collaboration," says Brad Genereaux, Medical Imaging Alliance Manager with NVIDIA, "is how well they complement one another in providing hospitals and data scientists with the ecosystem they need to make this real. Powered by NVIDIA Clara and fed data by XNAT, we collectively unlock and enable a rapid, open, on-premise, packaged-in-a-box, web-based swiss-army-knife of AI tooling.”
Find more information and a schedule of events and presentations by NVIDIA and XNAT at this link: https://www.nvidia.com/en-us/events/rsna/
If you would like to schedule time to meet with XNAT and NVIDIA Clara representatives at RSNA, drop us a line at info@xnat.org.