The Man With 100,000 Brains: Driving UK Life Science Breakthroughs with Simulated Data
Supercomputing is already being used in the UK to advance healthcare research, from medical imaging, genomics, and drug discovery. The below healthcare use case emphasises that, with the application of AI, supercomputing is invaluable for unlocking the value of data.
Researchers at King’s College London used the NVIDIA Cambridge-1 supercomputer and MONAI to create a treasure trove of open-source synthetic brain images, accelerating AI in healthcare. Project lead Jorge Cardoso is a founding member of the MONAI open-source consortium and a researcher in AI for medical imaging. In that last role, Cardoso and his team have discovered ways to create realistic, high resolution 3D images of human brains with AI. Through this most recent project, he is making 100,000 synthetic brain images available free to healthcare researchers. It’s a treasure trove that could accelerate understanding of dementia, aging or any sort of brain disease.
It’s a major donation compared to the world’s largest repository of freely available brain images. The UK Biobank currently maintains multiple brain images taken from more than 50,000 participants, curated at an estimated cost of $150 million.
Synthetic Data for Science
The images represent an emerging branch in healthcare of synthetic data, something that’s already widely used in computer vision for consumer and business apps. Those fields already have access to open datasets with millions of real-world images.
By contrast, medical images are relatively scarce, typically only available to researchers connected to large hospitals, given the need to protect patient privacy. Even then, medical images tend to reflect the demographics the hospital serves, not necessarily the broader population.
A fortunate feature of the new AI approach is it can make images to order. Though they’re simulated, the images are highly useful because they preserve key biological characteristics, so they look and act like real brains would.
Massive Images, Up to 10x Speedups
An NVIDIA DGX SuperPOD, Cambridge-1 packs 640 NVIDIA A100 Tensor Core GPUs, each with enough memory to process one or two of the team’s massive images made up of 16 million 3D pixels.
MONAI’s building blocks include domain-specific data loaders, metrics, GPU accelerated transforms and an optimized workflow engine. The software’s smart caching and multi-node scaling can accelerate jobs up to 10x, said Cardoso. He also credited cuDNN and “the whole NVIDIA AI software stack that helped us work much faster.”
Scaling with MONAI on Cambridge-1
This work was undertaken on the NVIDIA Cambridge-1, a supercomputer dedicated to breakthrough AI research in healthcare. MONAI, an AI framework for medical imaging, provided the software.
Together they created an AI factory for synthetic data that let researchers run hundreds of experiments, choose the best AI models and run inference to generate images.
Beyond the Brain
The team is exploring how the models can make 3D images of any part of the human anatomy in any mode of medical imaging, whether this is MRIs, CAT or PET scans.
Synthetic images will help researchers see how diseases evolve over time. Meanwhile the team at KCL is still exploring how to apply the work to body parts beyond the brain and what kinds of synthetic images (MRI, CAT, PET) are most useful.
A compute engine to unlock the value of data
As the OEDC wrote in a recent publication from its Working Group on Compute & Climate, access to computing resources is a critical enabler for the advancement and diffusion of AI. Cardoso’s project is a powerful illustration of how combining innovative techniques like AI and synthetic data generation can turbo-charge progress towards solving some of the UK’s most urgent challenges. Thanks to the compute power provided by Cambridge-1, as well as NVIDIA’s AI software stack, the KCL team has been able to accelerate their work and the benefits it will bring to UK dementia research significantly.
Rory Daniels
Rory joined techUK in June 2023 after three years in the Civil Service on its Fast Stream leadership development programme.
Laura Foster
Laura is techUK’s Associate Director for Technology and Innovation.
Elis Thomas
Elis joined techUK in December 2023 as a Programme Manager for Tech and Innovation, focusing on AI, Semiconductors and Digital ID.