As part of a programme to boost data collection to the National COVID-19 Chest Imaging Database, a new SMART box delivers regional data with one simple installation.
At the start of the COVID-19 pandemic The NHS AI Lab Imaging programme diverted its attention to the need for a shared chest image resource for the UK. The result of that work was a National COVID-19 Chest Imaging Database (NCCID), which provides UK hospitals with a shared library of chest X-rays, CT scans and MRI images.
These images, from people with both positive and negative COVID-19 status, now provide a large, centralised UK resource that researchers can use to better understand the virus. It also provides developers with datasets on which they can train and develop technology that ensures the best care for people hospitalised with severe infection.
The benefits of collecting chest imaging data are extensive and the NHS AI Lab team is continually looking for ways to increase hospital contributions to the NCCID. A larger database means better quality research and trials.
Despite willingness to participate, hospital staff are understandably low on time. Data collection typically involves labour-intensive steps to curate and send imaging studies via the Image Exchange Portal.
Now, the deployment of a single SMART box can provide access to 15 years’ worth of imaging data across 2.5 million people in the region. One installation like this allows contribution from an entire region to the NCCID.
How the SMART box works
Royal Surrey’s Secure Medical-image Anonymisation Receiver for Trials (SMART) box is a dedicated server that acts as a data collection hub for collecting images. Although it collects data from multiple sites, this collection takes place entirely within the participating sites digital infrastructure.
The SMART box frees up hospital staff time by managing the upload of patient data and automating the collection of the medical images before a nightly transfer to the database. By keeping the process within the hospital infrastructure, it avoids the transfer of identifiable patient information. Hospital trusts have full transparency on the processing of their data and everyone’s data is handled in a consistent way.
Images are de-identified, or made anonymous, at the point of upload. This means that no patient identifiable information will leave hospital sites. The anonymous images become part of data sets that can be used to trial new AI technologies and to validate, or prove the accuracy, of existing ones.
This data has the potential to enable faster patient assessment in A&E, save radiologists’ time, increase the safety and consistency of care across the country, and ultimately save lives. It is being made available to researchers, clinicians, technology companies and all those wanting to investigate the disease and develop solutions that can support people with COVID-19.
The success of the project and the speed with which it was set up means the NHS AI Lab Imaging team will focus on further regional models like this one.
What next for data collection?
As discovered through the NCCID study, the COVID-19 pandemic has highlighted the potential for medical imaging systems powered by AI. The power of large-scale data collection has also already been demonstrated in other screening studies such as breast cancer.
The NHS AI Lab Imaging team is now developing a National Medical Imaging Platform with the purpose of creating a pipeline for patient data to be used within AI technology across conditions other than COVID-19.
Setting up regional hubs, such as the SMART box deployment in Cheshire and Merseyside, will aid the creation of these large-scale national databases. Resources like the National Medical Imaging Platform will support the testing and development of AI technologies that combat a variety of conditions.
One of the findings coming out of the end of this project will definitely be to focus on regional hubs that will be able to coordinate and better centralise the data, a hub just like Cheshire and Merseyside. It takes years sometimes to set up SMART boxes at individual trusts, so doing it regionally is the only way
Dr Mark Halling-Brown, Head of Scientific Computing at Royal Surrey County Hospital
The NCCID database is a centralised UK database containing X-Ray, CT and MRI images from hospital patients across the country. In collaboration with The British Society of Thoracic Imaging (BSTI) and The Surrey NHS Foundation Trust, NCCID is being used to support a better understanding of the COVID-19 virus to develop technology that will help diagnosis of the disease and enable the best care for patients hospitalised with severe infection.
For more about this project or the work of the AI Imaging team, please contact email@example.com.