Developed at its Cambridge Research Lab, the InnerEye project helps in developing AI models that leverage the hospital’s own data to automatically show tumours and healthy organs on patient scans. These are then checked and confirmed by a clinical oncologist prior to giving treatment to the patient, said Microsoft.
According to the tech major, the process will reduce the otherwise lengthy treatment planning stage, which is crucial for head and neck cancers, which can multiply quickly if left untreated.
Microsoft claims that InnerEye can help execute contouring process in complex cases 13 times faster than the current approach.
Addenbrooke’s Hospital oncologist and InnerEye co-lead Dr Raj Jena said: “The results from InnerEye are a game-changer. To be diagnosed with a tumour of any kind is an incredibly traumatic experience for patients.
“So as clinicians we want to start radiotherapy promptly to improve survival rates and reduce anxiety. Using machine learning tools can save time for busy clinicians and help get our patients into treatment as quickly as possible.”
Run by Cambridge University Hospitals NHS Foundation Trust, Addenbrooke’s is a teaching hospital, research centre, and also a designated academic health science centre.
Addenbrooke’s Hospital and Microsoft are said to have been collaborating over the last eight years to develop and pilot InnerEye. The hospital will become the first NHS facility to have introduced a deep-learning solution trained on its own data, once the AI tool is deployed.
Content from our partners
Microsoft said that to make sure that all hospitals can use the InnerEye Deep Learning Toolkit, the company has made it available freely as opensource software. However, clinical use of machine learning models is subject to approval from relevant regulators.
Microsoft Research Cambridge principal research manager Javier Alvarez-Valle said: “AI models trained with InnerEye are changing the way cancer is treated, speeding up the process to give patients greater peace of mind and empowering clinical oncologists with an AI assistant.
“The AI works in the background, so clinical oncologists just open up the scans on their computer and they can see what their AI model has highlighted. The clinical oncologist then decides what to do with that information.
“AI models trained with InnerEye will be hosted in Microsoft’s Azure cloud, so all the data is securely held in the UK and only available to the medical staff who need to use it.”
