Home Blogs & News

The Aridhia DRE Supporting Cancer Research at the Royal Marsden NHS Foundation Trust

In a previous blog, we introduced Royal Marsden Hospital’s BRIDgE (Biomedical Research Informatics Digital Environment) platform. In this follow up blog, Lisa Scerri, BRIDgE Business Manager at Royal Marsden discusses some of the innovative projects that have taken place on the platform, enabled by the Aridhia DRE.

The Aridhia DRE supporting cancer research at the Royal Marsden NHS Foundation Trust

Aridhia powers the BRIDgE (Biomedical Research Informatics Digital Environment) Trusted Research Environment at the Royal Marsden, enabling innovative data science projects using multi-modal real-world cancer data.

BRIDgE projects have included the Early Detection of Recurrence following Radical Radiotherapy using Machine Learning by Dr Sumeet Hindocha, where routinely collected medical record data was used to classify patients into high and low risk groups for recurrence after having had radical radiotherapy for Non-Small Cell Lung Cancer, with the aim of potentially diagnosing recurrence earlier and therefore improving outcomes, and An Evaluation of Patients with Lung Nodules Using Automated Data Extraction by Dr Ben Hunter who worked with the RM informatics team to use natural language processing on electronic health records, to identify patients with lung nodules in their scan reports and evaluate our current service for diagnosing lung nodules and metastases in different cancer types. Both projects, plus the PillCam led by Dr Elena Cojocaru, who is developing an AI tool using images of the colon collected using a device swallowed by a patient, with a view to eventually utilising this technology in colon screening, were led by Dr Richard Lee’s Early Diagnosis and Detection Centre.

The BRIDgE TRE’s ‘Five Safes’ governance framework enables seamless working between the Royal Marsden and our research partners, such as the Institute of Cancer Research (ICR), and supports a number of projects led by Professor Ceire Costelloe and her team of data scientists from the ICR Health Informatics Unit. Their recent collaboration with Professor Udai Banerji of the Drug Development Unit, supported by our team, used analysis of electronic medical records to study the effects of comorbidity, which is the existence of two or more chronic conditions, and the resulting polypharmacy, which is the existence of several medications, to explore what barriers exist for cancer patients with comorbidity and polypharmacy in trial recruitment, and to identify their additional healthcare needs in clinical trials. Professor Costelloe and her team are also working with The Clatterbridge Cancer Centre and The Christie NHS Foundation Trust to analyse routinely collected data to determine whether patients could benefit from a digital alert for sepsis that is designed specifically for cancer patients.

Dr Lee and Prof Costelloe are also using BRIDgE for a London Data-at-Scale Project looking at the care of patients with hard-to-diagnose cancers. This also utilises the OneLondon Secure Data Environment to link data from patients seen in Rapid Diagnostic Centres, based in acute London hospitals, with their primary care data, to investigate and address reasons for service disparities (including referrals and outcomes) between London regions, for different patient groups, symptoms and cancer sites.

Recently we have sought to enable AI development using multi-modal data by working with Dr Simon Doran, senior imaging scientist from the ICR, to create a pipeline from the hospital PACs system, using XNAT, for the delivery of anonymised images to the Trusted Research Environment. BRIDgE now allows images, digital pathology, genomics and data from the RM electronic healthcare record system to be appropriately anonymised or pseudonymised and presented to research users, within a secure data environment. This aligns with the NHS ‘Data Saves Lives’ strategy, which identifies TREs (also known as secure data environments (SDE)) as ‘a big step forward in how data can be accessed securely in a virtual setting.’ NHS England envisage that SDEs ‘will become the default route for accessing NHS data for research and external uses.’