NGC | Catalog


For downloads and more information, please view on a desktop device.
Logo for clara_pt_covid19_ct_lesion_segmentation


A pre-trained model for volumetric (3D) segmentation of the COVID-19 lesion from CT images.



Use Case



Clara Train

Latest Version



March 25, 2022


9.23 MB

Model Overview

A pre-trained model for volumetric (3D) segmentation of the COVID-19 lesion from CT images.

Note: The 4.1 version of this model is only compatible with the 4.1 version of the Clara Train SDK container

Model Architecture

The model is trained using a 3D SegResNet [1].

Diagram showing the flow from model input, through the model architecture, and to model output


The training was performed with the following:

  • Script:
  • GPU: four (at least) 16GB of GPU memory
  • Actual Model Input: 224 x 224 x 32
  • AMP: False
  • Optimizer: Adam
  • Learning Rate: 1e-3
  • Loss: DiceCELoss

If out-of-memory or program crash occurs while caching the data set, please change the cache_rate in CacheDataset to a lower value in the range (0, 1).


This model was trained on a global dataset with a large experimental cohort collected from across the globe. The CT volumes of 919 independent subjects are provided by NIH with experts’ lesion region annotations.

  • Target: Lesion
  • Task: Segmentation
  • Modality: CT
  • Size: 919 3D volumes (736 Training, 90 Validation, 93 Testing)
  • Challenge: Large ranging foreground size


Dice score is used for evaluating the performance of the model. On the test set, the trained model achieved score of 0.7109 for lesion.


A graph showing the training loss for 2000 epochs (46000 iterations).

Graph that shows training acc over 2000 epochs


A graph showing the validation mean Dice for 2000 epochs (46000 iterations).

Graph that shows validation mean dice getting higher over 2000 epochs until converging around 0.72

How to Use this Model

The model was validated with NVIDIA hardware and software. For hardware, the model can run on any NVIDIA GPU with memory greater than 16 GB. For software, this model is usable only as part of Transfer Learning & Annotation Tools in Clara Train SDK container. Find out more about Clara Train at the Clara Train Collections on NGC.

Full instructions for the training and validation workflow can be found in our documentation.


Input: 1 channel CT image with intensity in HU and arbitary spacing


  1. Resampling spacing to 0.8 x 0.8 x 5mm
  2. Clipping intensity to [-1000, 500] HU
  3. Converting to channel first

Augmentation for training:

  1. Randomly cropping the volume to a fixed size 224 x 224 x 32
  2. Randomly applying spatial flipping
  3. Randomly applying spatial rotation
  4. Randomly shifting intensity of the volume


Output: 2 channels

  • Label 0: background
  • Label 1: lesion

Sliding-window Inference

Inference is performed on 3D volumes in a sliding window manner with a specified stride.


This training and inference pipeline was developed by NVIDIA. It is based on a segmentation model developed by NVIDIA researchers. This research use only software has not been cleared or approved by FDA or any regulatory agency. Clara pre-trained models are for developmental purposes only and cannot be used directly for clinical procedures.


[1] Myronenko, A., 2018, September. 3D MRI brain tumor segmentation using autoencoder regularization. In International MICCAI Brainlesion Workshop (pp. 311-320). Springer, Cham.


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.