National Medical Imaging Platform (NMIP)

The National Medical Imaging Platform is part of a programme of work by the AI Imaging team at the NHS AI Lab, helping to support the rapid development of the best AI technologies for improving patient health and care.

Following the success of our National COVID-19 Chest Imaging Database, created to support the NHS pandemic response, we are now preparing to develop a National AI Medical Imaging Platform (NMIP). This platform will include even larger volumes of data and different types of medical imaging, giving AI developers access to the kind of high-quality data needed to develop and test AI products for a range of health and social care needs.

Why is a medical imaging platform needed?

Artificial intelligence technologies applied to imaging, such as cancer screening, are amongst the most advanced uses of AI in healthcare. Adoption of these technologies has significant potential to both improve care outcomes and reduce costs for the NHS across the UK.

The National Medical Imaging Platform will make it possible to collect imaging data on a national scale, and provide this data for the development and testing of AI screening technology. The platform aims to provide a database of medical data and imaging of sufficient size and quality to:

  • enable the creation of safe AI innovations and research that will improve patient health
  • provide developers with the high quality training and testing data sets needed to produce the best AI
  • test for and reduce the impact of biases that can result from AI tools that have been trained on unrepresentative or small sample data sets
  • give researchers a single source of national medical data that will lead to a better understanding of disease

How you can get involved

Data collection and access

We are collaborating with NHS trusts and healthcare providers to ensure that when we start collecting data to the NMIP, we will create a database that is large and diverse. How an AI technology performs is inherently linked to the characteristics of the data on which it was trained and validated. This is why it is important to train and validate AI technologies on sufficiently sized datasets that are representative of the UK population. This gives researchers and technology companies the confidence that the results of the test will be a good reflection of how their model might perform in a real world healthcare setting.

When the NMIP is developed we will create a Data Access Environment (DAE) that provides not only the data, but also the tools with which AI developers and researchers can analyse the data and train and validate their AI technologies all in the same secure and controlled environment.

In the meantime, you can contribute chest-imaging data and make access requests to our National COVID-19 Chest Imaging Database.

Engaging with people and communities

We are engaging with the people using or affected by our imaging database to understand concerns and answer questions.

Contact us

If you have questions about the NMIP or NCCID, please get in touch with the AI Imaging team at imaging@nhsx.nhs.uk.