Updating README.md with metadata and consistency improvements

README.md

@@ -1,24 +1,35 @@
 ---
 license: other
 license_name: nvidia-open-model-license-agreement
-license_link: >-
-  https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
+license_link: https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
+pipeline_tag: image-to-image
+library_name: monai
+tags:
+  - nvidia
+  - medical-imaging
+  - ct
+  - synthetic-data
+  - generation
 ---
 
-# Model Overview
+# NV-Generate-CT
+
+![Generation Demo](https://raw.githubusercontent.com/NVIDIA-Medtech/.github/main/profile/generate.gif)
 
 ## Description:
-NVIDIA MAISI (Medical AI for Synthetic Imaging) is a state-of-the-art three-dimensional (3D) Latent Diffusion Model designed for generating high-quality synthetic CT images with or without anatomical annotations. This AI model excels in data augmentation and creating realistic medical imaging data to supplement limited datasets due to privacy concerns or rare conditions. It can also significantly enhance the performance of other medical imaging AI models by generating diverse and realistic training data.
+NVIDIA NV-Generate-CT is a state-of-the-art three-dimensional (3D) Latent Diffusion Model designed for generating high-quality synthetic CT images with or without anatomical annotations. This AI model excels in data augmentation and creating realistic medical imaging data to supplement limited datasets due to privacy concerns or rare conditions. It can also significantly enhance the performance of other medical imaging AI models by generating diverse and realistic training data.
+
+NV-Generate-CT offers several key features:
 
-MAISI offers several key features:
+- Generates high-resolution 3D CT images up to 512 × 512 × 768 voxels
+- Supports variable voxel sizes ranging from 0.5mm to 5.0mm
+- Capable of annotating up to 127 anatomical classes, including organs and tumors
+- Allows controllable anatomy size for 10 specific classes
+- Produces paired segmentation masks
 
-Generates high-resolution 3D CT images up to 512 × 512 × 768 voxels
-Supports variable voxel sizes ranging from 0.5mm to 5.0mm
-Capable of annotating up to 127 anatomical classes, including organs and tumors
-Allows controllable anatomy size for 10 specific classes
-Produces paired segmentation masks
+By providing these capabilities, NV-Generate-CT is a valuable tool for researchers advancing AI applications in healthcare. However, it is important to note that this model is intended for research purposes only and not for clinical usage.
 
-By providing these capabilities, MAISI is a valuable tool for researchers advancing AI applications in healthcare. However, it is important to note that this model is intended for research purposes only and not for clinical usage.
+**Training & Fine-tuning**: Visit [GitHub](https://github.com/NVIDIA-Medtech/NV-Generate-CTMR) for training scripts, ControlNet fine-tuning, VAE training, and advanced configuration guides with comprehensive documentation.
 
 ## Terms of Use
 
@@ -91,50 +102,14 @@ Inference Engine: Triton <br>
 **[Preferred/Supported] Operating System(s):** <br>
 * Linux <br>
 
-## Model Version(s): 
-0.3.1  <br>
-
-# Training & Evaluation: 
-## Training Dataset:
-Internal ONLY
-~35 Datasets
-Name, JIRA/SWIPAT, Commercial, and # of Data Tracked
-"MAISI" Sheet: https://docs.google.com/spreadsheets/d/14frhzELquSF_-tF7yGFDBHmSdnp-9-5pmbONQx8iQWk/edit?usp=sharing
-https://docs.google.com/spreadsheets/d/1hmv-O-f6tdgndsRnoqCgcunR2uQ9IySDhZWmjsXwgbM/edit?usp=sharing
-
-## Evaluation Dataset:
-Internal ONLY
-~35 Datasets
-Name, JIRA/SWIPAT, Commercial, and # of Data Tracked
-"MAISI" Sheet: https://docs.google.com/spreadsheets/d/14frhzELquSF_-tF7yGFDBHmSdnp-9-5pmbONQx8iQWk/edit?usp=sharing
-https://docs.google.com/spreadsheets/d/1hmv-O-f6tdgndsRnoqCgcunR2uQ9IySDhZWmjsXwgbM/edit?usp=sharing
-
-** Data Collection Method by dataset <br>
-* Hybrid: Human, Automatic/Sensors <br>
-
-** Labeling Method by dataset <br>
-* Hybrid: Human, Automatic/Sensors <br>
-
-**Properties:** Custom internal and public datasets of 60,000 3D volumes from multiple scanner types.  <br>
-
-## Evaluation Dataset:
-
-** Data Collection Method by dataset <br>
-* Hybrid: Human, Automatic/Sensors <br>
-
-** Labeling Method by dataset <br>
-* Hybrid: Human, Automatic/Sensors <br>
-
-**Properties:** Custom internal and public datasets of organs from multiple scanner types. <br>
-
 ## Inference:
 **Engine:** PyTorch<br>
 **Test Hardware:** 
-A100 with at least 80GB memory for 512x512x512 images<br>
-H100 with at least 80GB memory for 512x512x512 images<br>
+A100<br>
+H100<br>
 
 ## Additional Information:
-The current list of classes available within MAISI:
+The current list of classes available:
   "liver": 1,
   "spleen": 3,
   "pancreas": 4,
@@ -260,6 +235,17 @@ The current list of classes available within MAISI:
   "bone lesion": 128,
   "airway": 132
 
+## Resources
+
+- **Training & Development**: [GitHub Repository](https://github.com/NVIDIA-Medtech/NV-Generate-CTMR) - Complete training pipeline (VAE, diffusion model, ControlNet), fine-tuning guides, and comprehensive development documentation
+- **Interactive Demo**: [MAISI on build.nvidia.com](https://build.nvidia.com/nvidia/maisi) - Try toy examples online with instant generation
+- **Sister Model**: [NV-Generate-MR](https://huggingface.co/nvidia/NV-Generate-MR) - MR image generation variant
+- **Research Papers**:
+  - [MAISI: Medical AI for Synthetic Imaging (WACV 2025)](https://arxiv.org/pdf/2409.11169)
+  - [MAISI-v2: Accelerated 3D High-Resolution Medical Image Synthesis with Rectified Flow](https://arxiv.org/pdf/2508.05772)
+- **Built with**: [MONAI](https://monai.io/) - Medical Open Network for AI
+- **Clara Medical Collection**: [View all NVIDIA medical AI models](https://huggingface.co/collections/nvidia/clara-medical)
+
 ## Ethical Considerations:
 NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications.  When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.  Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).