NGC | Catalog
CatalogModelsMONAI BraTS MRI Segmentation

MONAI BraTS MRI Segmentation

For downloads and more information, please view on a desktop device.
Logo for MONAI BraTS MRI Segmentation


A pre-trained model for volumetric (3D) segmentation of brain tumor subregions from multimodal MRIs based on BraTS 2018 data.



Latest Version



April 4, 2023


36.04 MB

Model Overview

A pre-trained model for volumetric (3D) segmentation of brain tumor subregions from multimodal MRIs based on BraTS 2018 data. The whole pipeline is modified from clara_pt_brain_mri_segmentation.

The model is trained to segment 3 nested subregions of primary brain tumors (gliomas): the "enhancing tumor" (ET), the "tumor core" (TC), the "whole tumor" (WT) based on 4 aligned input MRI scans (T1c, T1, T2, FLAIR).

  • The ET is described by areas that show hyper intensity in T1c when compared to T1, but also when compared to "healthy" white matter in T1c.
  • The TC describes the bulk of the tumor, which is what is typically resected. The TC entails the ET, as well as the necrotic (fluid-filled) and the non-enhancing (solid) parts of the tumor.
  • The WT describes the complete extent of the disease, as it entails the TC and the peritumoral edema (ED), which is typically depicted by hyper-intense signal in FLAIR.

Model workflow


The training data is from the Multimodal Brain Tumor Segmentation Challenge (BraTS) 2018.

  • Target: 3 tumor subregions
  • Task: Segmentation
  • Modality: MRI
  • Size: 285 3D volumes (4 channels each)

The provided labelled data was partitioned, based on our own split, into training (200 studies), validation (42 studies) and testing (43 studies) datasets.


The data list/split can be created with the script scripts/

python scripts/ --path your-brats18-dataset-path

Training configuration

This model utilized a similar approach described in 3D MRI brain tumor segmentation using autoencoder regularization, which was a winning method in BraTS2018 [1]. The training was performed with the following:

  • GPU: At least 16GB of GPU memory.
  • Actual Model Input: 224 x 224 x 144
  • AMP: True
  • Optimizer: Adam
  • Learning Rate: 1e-4
  • Loss: DiceLoss


4 channel aligned MRIs at 1 x 1 x 1 mm

  • T1c
  • T1
  • T2


3 channels

  • Label 0: TC tumor subregion
  • Label 1: WT tumor subregion
  • Label 2: ET tumor subregion


Dice score was used for evaluating the performance of the model. This model achieved Dice scores on the validation data of:

  • Tumor core (TC): 0.8559
  • Whole tumor (WT): 0.9026
  • Enhancing tumor (ET): 0.7905
  • Average: 0.8518
Training Loss and Dice

A graph showing the training loss and the mean dice over 300 epochs

Validation Dice

A graph showing the validation mean dice over 300 epochs

MONAI Bundle Commands

In addition to the Pythonic APIs, a few command line interfaces (CLI) are provided to interact with the bundle. The CLI supports flexible use cases, such as overriding configs at runtime and predefining arguments in a file.

For more details usage instructions, visit the MONAI Bundle Configuration Page.

Execute training:
python -m monai.bundle run training --meta_file configs/metadata.json --config_file configs/train.json --logging_file configs/logging.conf
Override the train config to execute multi-GPU training:
torchrun --standalone --nnodes=1 --nproc_per_node=8 -m monai.bundle run training --meta_file configs/metadata.json --config_file "['configs/train.json','configs/multi_gpu_train.json']" --logging_file configs/logging.conf

Please note that the distributed training-related options depend on the actual running environment; thus, users may need to remove --standalone, modify --nnodes, or do some other necessary changes according to the machine used. For more details, please refer to pytorch's official tutorial.

Override the train config to execute evaluation with the trained model:
python -m monai.bundle run evaluating --meta_file configs/metadata.json --config_file "['configs/train.json','configs/evaluate.json']" --logging_file configs/logging.conf
Execute inference:
python -m monai.bundle run evaluating --meta_file configs/metadata.json --config_file configs/inference.json --logging_file configs/logging.conf


[1] Myronenko, Andriy. "3D MRI brain tumor segmentation using autoencoder regularization." International MICCAI Brainlesion Workshop. Springer, Cham, 2018.


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


Copyright (c) MONAI Consortium

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.