Synthetic Whole-Head MRI Brain Tumor Segmentation Dataset

Citation Author(s):
National Taiwan University of Science and Technology
National Taiwan University, College of Medicine
Submitted by:
Zolnamar Dorjsembe
Last updated:
Wed, 07/26/2023 - 04:11
Data Format:
Research Article Link:
0 ratings - Please login to submit your rating.


The demand for artificial intelligence (AI) in healthcare is rapidly increasing. However, significant challenges arise from data scarcity and privacy concerns, particularly in medical imaging. While existing generative models have achieved success in image synthesis and image-to-image translation tasks, there remains a gap in the generation of 3D semantic medical images. To address this gap, we introduce Med-DDPM, a diffusion model specifically designed for semantic 3D medical image synthesis, effectively tackling data scarcity and privacy issues.

The novelty of Med-DDPM lies in its incorporation of semantic conditioning, enabling precise control during the image generation process. Our model outperforms Generative Adversarial Networks (GANs) in terms of stability and performance, generating diverse and anatomically coherent images with high visual fidelity. Comparative analysis against state-of-the-art augmentation techniques demonstrates that Med-DDPM produces comparable results, highlighting its potential as a data augmentation tool for enhancing model accuracy.

In conclusion, Med-DDPM pioneers 3D semantic medical image synthesis by delivering high-quality and anatomically coherent images. Furthermore, the integration of semantic conditioning with Med-DDPM holds promise for image anonymization in the field of biomedical imaging, showcasing the capabilities of the model in addressing challenges related to data scarcity and privacy concerns. Our code and model weights are publicly accessible on our GitHub repository at, facilitating reproducibility.


We created a synthetic Dataset with our proposed method Med-DDPM, containing 1000 whole head synthetic MRIs and their corresponding mask images. The masks have three labels: 0 for background, 1 for the head, and 2 for the tumor area. For binary segmentation, users can easily modify the head label to the background label and the tumor label to 1. To obtain additional synthetic data, users can download our code and pretrained model weights from our repository at and run our inference script.


required those dataset to train my ai model and to check the capability of my AI model

Submitted by Aditya Nanda on Thu, 01/25/2024 - 09:18

I used this dataset with deep learning for my graduation project.

Submitted by Atakan Kaplan on Wed, 05/29/2024 - 06:22

I need this dataset to train my deep neural network. Thanks.

Submitted by Dong Wang on Tue, 06/11/2024 - 21:29