recent news of my team
GTC 2020
Clara Train SDK 3.0 is released with new features of AutoML, DeepGrow, and more.
In a federated learning collaboration, the American College of Radiology, Diagnosticos da America, Partners HealthCare, Ohio State University and Stanford Medicine developed better predictive models to assess breast tissue density with NVIDIA Clara Federated Learning SDK. [link]
MONAI
On April 21 2020, NVIDIA and King’s College London Announce MONAI Open Source AI Framework for Healthcare Research
RSNA 2019
During RSNA 2019, NVIDIA announced the CLARA Train SDK 2.0.
- Federated learning module enable researcher to collaborate and build AI Models without sharing private data.
- Achieved 55x faster vs. Tensorflow Native and 28x faster vs. Clara Train SDK 1.0.
- Automatic Mixed Precision (AMP)
- Deterministic training
- Smart Cache
Documentation … [link]
MONAI
During MICCAI 2019, NVIDIA and KCL initial the consortium for Medical Open Network for AI – Toolkit for Healthcare Imaging (MONAI).
MICCAI 2019
In MICCAI 2019, DLMED team published 10 papers.
- Xiaosong’s paper on “Interactive 3D Segmentation Editing and Refinement via Gated Graph Neural Network” won the best paper award at GLMI workshop.
- Can’s paper on “iSMORE: an iterative self super-resolution algorithm” won the best paper at SASHIMI workshop.
Clara AI Toolkit
In GTC 2019, our CEO Jensen Huang announced the NVIDIA Clara AI Toolkit. NVIDIA Clara™ is a computational platform for developers that makes it easy to build, manage, and deploy intelligent medical imaging workflows and instruments.