100k Synthetic Brains (DDPM)

The Synthetic Brain Project is focused on building deep learning models that can synthesize high-resolution morphologically-correct 3D MRI images of human brains.

These models and datasets will help scientists pave the way toward better federated-learning setups, more robust models under limited data, safer AI, and improved anomaly detection.

The AI models were developed by King’s and NVIDIA scientists and engineers as part of The London Medical Imaging & AI Centre for Value-Based Healthcare research funded by UK Research and Innovation and a Wellcome Flagship Programme (in collaboration with University College London). The dataset is composed of over 100,000 T1w 1mm isotropic synthetic brains of different gender, ages, ventricular sizes and size to head ratios.

The code and models will soon be made available as open-source upon publication. Two twin-datasets are being released, utilising two state-of-the-art generative models, one using demonising diffusion probabilistic models (DDPM), and another using transformers. This specific dataset has been generated using DDPMs. The 100k samples have been generated using a uniformly sampled conditioning vector across ages, sexes, and brain size.

To find out more, contact Jorge Cardoso: m.jorge.cardoso@kcl.ac.uk