A new digital platform is transforming how researchers access and analyse cancer patient data. Image Lisa Scerri, BRC Informatics Business Partner Data collected in the health and care system is invaluable. It helps us provide patient care, plan services, carry out research and develop innovative treatments. BRIDgE (Biomedical Research Informatics Digital Environment) is a new digital platform that aims to improve patient outcomes and transform clinical practice at The Royal Marsden, allowing researchers to analyse cancer data in secure, collaborative, virtual spaces. Lisa Scerri, BRC Informatics Business Partner, says: “Traditionally, if a researcher wanted to access patient data, they’d have to download the dataset to their computer, which could lead to breaches in patient confidentiality. With BRIDgE, researchers can access our data remotely through a secure, collaborative ‘workspace’ for data analysis and artificial intelligence [AI] development.” At The Royal Marsden, the Trust collect routine clinical care data during treatment and store it securely in the cloud. When used in BRIDgE, this data is de-identified to protect patient confidentiality, and access is carefully restricted to individuals for cancer research purposes. Furthermore, the data cannot be downloaded from workspaces, and research outputs are checked before being released to ensure that patients cannot be re-identified. “Projects must be approved by the Trust’s Committee for Clinical Research,” says Lisa. “Once generated, the dataset is preserved in the BRIDgE data catalogue, FAIR, so that it can be reused.” There are several reasons why researchers may want to analyse this data, such as understanding how many people have a particular disease to see if a treatment is working, or to investigate a way of improving cancer care for the future. Dr Sumeet Hindocha, Clinical Oncology Specialist Registrar, used the platform to investigate how AI can help predict outcomes for non-small-cell lung cancer patients who have received radiotherapy. He says: “International guidance suggests that patients with lung cancer receive routine checks after treatment to detect recurrence. But there are no recommendations on the frequency of these follow-ups. "Through BRIDgE, we were able to gather de-identified data and use this to compare different machine-learning models to determine which could most accurately identify patients at risk of recurrence following curative radiotherapy. “Our results suggest that this technology could be used to help personalise and improve the surveillance of patients following treatment based on their risk.” BRIDgE is managed by the Trust’s information team, supported by the NIHR BRC (BRC) at The Royal Marsden and The Institute of Cancer Research, London.