|NVIDIA And King’s College London Uses Cambridge-1 To build AI Models To Generate Synthetic Brain Images|
August 1, 2021
NVIDIA and King’s College London have revealed new information about one of the first projects to be run on Cambridge-1, the UK’s most powerful supercomputer. The UK’s most powerful supercomputer, Cambridge-1, was announced in October last year and cost $100 million to build.
King’s College London uses Cambridge-1 to create AI models that can generate synthetic brain images by learning from tens of thousands of MRI (Magnetic resonance imaging) brain scans of people of all ages and disorders.
The company’s early collaborations with AstraZeneca, GSK, Guy’s and St Thomas’ NHS Foundation Trust, King’s College London, and Oxford Nanopore Technologies include:-
Scientists will be able to distinguish healthy brains from diseased brains due to this new research, providing them a more sophisticated knowledge of how diseases appear and potentially allowing for earlier and more accurate diagnoses. Jorge Cardoso, a senior lecturer of artificial medical intelligence at King’s College London, mentioned that Cambridge-1 allows accelerated generation of synthetic data that gives researchers at King’s College London a better understanding of how different factors affect the brain, anatomy, and pathology. Jorge also added that you could ask their model to generate an almost infinite amount of data with prescribed ages and diseases. With this, they can start tackling problems such as how diseases affect the brain and when abnormalities might start existing.
- Developing a deeper understanding of brain diseases similar to dementia.
- Using AI to design new drugs.
- Improving the accuracy of being able to find disease-causing variations in human genomes.
AI for healthcare is proliferating in the UK, with a range of startups and larger pharmaceutical companies turning to mine the vast quantities of data available to discover potential drugs, further understand certain diseases, and hence, improve and personalize patient care.
The use of synthetic data has the extra benefit of ensuring patient privacy since the images were AI-generated. This also allows King’s to open the research to the broader UK healthcare community.
The AI model was created by data scientists and engineers from King’s and NVIDIA. It’s one of the numerous ongoing initiatives on Cambridge-1. Drug discovery and genome sequencing are among the digital biology projects proposed by other top UK healthcare organizations.
With 80 NVIDIA DGXTM A100 systems integrating NVIDIA A100 GPUs, BlueField®-2 DPUs, and NVIDIA HDR InfiniBand networking, Cambridge-1 is the UK’s most powerful supercomputer.
The synthetic data model developed by King’s College London will be shared with the more extensive research and startup community.