NGC | Catalog
CatalogContainersClara Train SDK

Clara Train SDK

Logo for Clara Train SDK
Description
This domain optimized developer application framework is based on the open source SDKs of MONAI and NVIDIA FLARE. It brings ease of use and a simple CLI-based interface, allowing to quickly get started with annotating, training, and adapting AI models.
Publisher
NVIDIA
Latest Tag
v4.1
Modified
April 4, 2023
Compressed Size
6.62 GB
Multinode Support
Yes
Multi-Arch Support
No
v4.1 (Latest) Security Scan Results

Linux / amd64

Sorry, your browser does not support inline SVG.

NVIDIA Clara Train SDK for Medical Imaging

Clara Train is a domain optimized developer application framework that is based upon the open source SDKs of MONAI and NVIDIA FLARE.

Clara Train brings ease of use and a simple CLI-based interface, allowing users to quickly get started for annotating, training, and adapting AI models.

It includes capabilities for AI-Assisted Annotation, making any medical viewer AI capable. Clara Train also provides pre-trained models to kick start AI development with techniques like Transfer Learning, Federated Learning, and AutoML.

The pre-trained models are provided as Model Applications, packaged as MMARs (Medical Model Archives), to users, providing an intuitive config based environment for data scientists and researchers to get kick-started with AI Development.

Getting Started

Clara Train SDK documentation can be accessed here.

To learn more about how to run Clara Train on a cloud service provider, refer to the documentation here.

What is New

Clara Train v4.1 is built on an open-source foundation. For AIAA and training framework, we leverage MONAI Label and MONAI. For Federated Learning, we leverage NVIDIA FLARE.

With 4.1, MONAI Label now includes continuous learning allowing end users to annotate and trigger a training loop from client viewer.

The MMAR API is a new feature in Clara Train 4.1 to allow for creating an MMAR from python code. For more information, see the documentation section here.

Clara Train brings the latest and greatest features from MONAI - like Vision Transformers for 3D Imaging; self supervised learning and AutoML for network architecture search. See more information about these features here.

This version also includes a digital pathology pipeline for a fully convolutional classification network that works with whole-slide images and performs optimized data loading using cuCIM , which can tile large datasets on-demand and process them through a CUDA-enabled pipeline.

Below are the latest pre-trained models being released by NVIDIA. Note the 4.1 versions of these models are only compatible with the 4.1 version of the Clara Train SDK container.

  1. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_pathology_metastasis_detection
  2. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_fed_learning_brain_tumor_mri_segmentation
  3. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_spleen_ct_annotation
  4. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_covid19_3d_ct_classification
  5. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_covid19_ct_lung_segmentation
  6. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_brain_mri_segmentation
  7. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_liver_and_tumor_ct_segmentation
  8. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_covid19_ct_lesion_segmentation
  9. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_deepgrow_2d_annotation
  10. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_deepgrow_3d_annotation
  11. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_spleen_ct_segmentation
  12. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_chest_xray_classification
  13. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_net_arch_search_segmentation
  14. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_unetr_ct_btcv_segmentation
  15. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/med/models/clara_pt_self_supervised_learning_segmentation

Relevant Documentation

  1. For Clara Train SDK, please refer to this section in the Clara Train Getting Started guide.
  2. For AIAA, please refer to this section in AIAA Getting Started guide.
  3. For NVIDIA FLARE resources, please refer to this page.
  4. For MONAI resources, please refer to this MONAI webpage.

License

End User License Agreement is included with the product. Licenses are also available along with the model application zip file. By pulling and using the Clara Train SDK container and downloading models, you accept the terms and conditions of these licenses.

Suggested Reading

  1. Getting Started Guide at https://docs.nvidia.com/clara/clara-train-sdk/pt/model.html
  2. https://developer.nvidia.com/blog/tag/clara/

Use the NVIDIA Devtalk forum for questions regarding this Software