Addenbrooke’s Hospital in Cambridge along with 20 other hospitals from across the world and healthcare technology leader, NVIDIA, have used artificial intelligence (AI) to predict COVID-19 patients’ oxygen needs.
The research published in Nature Medicine, used an algorithm to analyse chest x-rays and anonymised electronic health data from hospital patients with COVID-19 symptoms.
Once the algorithm learned from the data, the analysis was used to build an AI tool called EXAM, which was found to predict patients’ oxygen needs within 24 hours of their arrival in the emergency department, with a sensitivity of 95% and a specificity of over 88%.
Through federated learning, the outcomes of around 10,000 COVID-19 patients worldwide were analysed, including 250 who came to Addenbrooke’s Hospital in the first wave of the pandemic in 2020.
The research was supported by the National Institute for Health Research (NIHR) Cambridge Biomedical Research Centre (BRC).
US hospital Mass General Brigham and the NIHR Cambridge BRC are now working with NVIDIA Inception startup Rhino Health to run prospective studies on EXAM.
WHY IT MATTERS
The study is one of the largest, most diverse clinical federated learning studies to date, bringing together collaborators across North and South America, Europe and Asia.
“Federated learning has transformative power to bring AI innovation to the clinical workflow,” said Professor Fiona Gilbert, who led the study in Cambridge and is honorary consultant radiologist at Addenbrooke’s Hospital.
THE LARGER CONTEXT
Another AI trial is taking place at Addenbrooke’s Hospital in Cambridge using an algorithm to diagnose dementia from an MRI brain scan before symptoms develop. It is also the first hospital in the world to use InnerEye, an AI deep-learning tool from Microsoft Research Cambridge that accelerates the treatment of cancer patients.
Cambridge University Hospitals NHS foundation trust (CUH), which runs Addenbrooke’s and the Rosie hospitals, was the first UK healthcare trust to be awarded Stage 7 on the Electronic Medical Record Adoption Model(EMRAM) by HIMSS, recognising its use of technology, data and analytics.
ON THE RECORD
First author on the study, Dr Ittai Dayan, from Mass General Bingham in the US, said: “By developing the EXAM model using federated learning and objective, multimodal data from different continents, we were able to build a generalisable model that can help frontline physicians worldwide.”
Dr Mona G Flores, NVIDIA global head for medical AI, said: “Federated learning allowed researchers to collaborate and set a new standard for what we can do globally, using the power of AI. This will advance AI not just for healthcare but across all industries looking to build robust models without sacrificing privacy.”
Professor Gilbert added: “Creating software to match the performance of our best radiologists is complex, but a truly transformative aspiration.”
Source: Read Full Article