Skip to main content

Federated learning may provide a solution for future digital health challenges

A new position paper from London Medical Imaging and AI Centre for Value Based Healthcare researchers, partnering organisations such as Nvidia and OWKIN, and 14 other institutions, proposes the use of federated learning, a machine learning technique that trains an algorithm across multiple decentralised data points, to provide a solution to securely utilising large volumes of clinical data and help realise the full potential of machine learning in healthcare.  

Published recently in Nature Digital Medicine (NDM), the paper reveals the solutions federated learning may provide for the future of digital health, and the challenges that arise around quality, hegemony, and security of patient data. 

AI Centre CTO and senior contributor, Dr. Jorge Cardoso, said “federated learning is currently the best approach for scalable, safe, robust and fair AI in a healthcare setting.”

The paper highlights that as existing medical data sits in “data silos” with restricted access, a federated learning model could be the key to realising the potential of AI, as it enables machine learning from non-co-located data, whilst addressing privacy and data governance challenges.

In practice, this means a centralised AI algorithm can learn from locally stored anonymised data at multiple NHS Trusts by building a consensus model, from multiple data sources, without any patient data leaving the secure data enclaves at each hospital.

Federated learning ensures data is safe and patient’s privacy is preserved, while providing unparalleled access to high-fidelity data – the opportunities are immense.
– Dr Jorge Cardoso

Already through federated learning clinicians have improved diagnostic tools for imaging analysis and pharma companies are decreasing cost and time to market with collaborative and accelerated drug discovery.

Dr. Nicola Rieke, Senior Solution Architect and Researcher at NVIDIA, said “The federated learning paradigm expands AI development to diverse data to create robust models and realise greater value in clinical practices. NVIDIA’s work in federated learning is centred on preserving patient data privacy while democratising the ability to build AI collaboratively, resulting in AI models that generalise at a global scale.”

The researchers said federated learning has “the potential to increase the accuracy and robustness of healthcare AI, while reducing costs and improving patient outcomes”, which in turn could be vital for precision medicine and ultimately improve medical care in the coming decades.

The AI Centre, in partnership with 11 NHS Trusts, is in the process of deploying an ambitious Federated Learning Interoperability Platform (FLIP), to support pioneering artificial intelligence systems across the NHS.