Cambridge hospital Addenbrooke has become the first in the world to use Inner Eye, a Microsoft AI software able to pinpoint tumours, to treat patients. The software, which is proving itself incredibly precise at being able to identify tumours from scans, is helping to speed up cancer treatments, cutting backlogs and saving lives.
Inner Eye is reducing the time it takes clinicians to prepare radiotherapy treatment to the targeted needs of patients by up to 90%. While the software is demonstrating high levels of accuracy, its results are still checked over and confirmed by a clinical oncologist before treatment is signed off on and started.
Recent statistics suggest that as many as 50% of the people in the UK will develop a form of cancer at some point in their life. Half of those will require radiotherapy treatment, which uses focused, high-intensity radiation beams to damage tumour DNA, while avoiding healthy tissue and organs around them.
Left untreated, many cancers grow quickly, potentially spreading to other parts of the body. Reducing the treatment planning stage, which is currently lengthy, has the potential to save numerous lives.
Inner Eye was developed over eight years, through a joint project between Microsoft and Addenbrooke’s and is free to use for other hospitals. It will now be introduced into more NHS trusts, with hospitals ‘training’ the AI tool on their own data sets to improve accuracy.
Addenbrooke’s oncologist Raj Jena comments on Inner Eye’s potential with:
“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.”
The standard process for preparing radiotherapy treatment involves creating a series of 3D scans of a patient. These are made up of stacks of 2D images, dozens deep. Each is examined and marked up by a radiation oncologist, clinical oncologist or specialised technician in a process referred to as ‘contouring’.
A contour line must be manually drawn around the tumours and key healthy organs in a target area, by experts using specialist software. This planning stage can often take several hours, but Inner Eye can do the job thirteen times faster.
Commenting on the development, Imperial College London professor Pat Price, who is also chairwoman of the charity Action Radiotherapy, said:
“This is just one brilliant example of the quiet but amazing technological revolution that has unfolded in radiotherapy in recent years and could dramatically improve cancer survival rates.”
Waiting times for cancer treatment were an issue before the Covid-19 pandemic, which has drastically exacerbated the issue. Between April and September last year, 4.4 million fewer cancer scans were performed compared to the same period in 2019. That means potentially hundreds of thousands of people missing out on potentially life-saving diagnosis and treatment.
One in seven patients are currently waiting for more than three months for scans. Reducing that wait by weeks could make all the difference between being able to effectively treat cancers, or them developing past the point of no return.
Over future years, Inner Eye and other new radiation technologies could make all the difference in that critical race against time.