Early diagnosis of breast cancer in north-east Scotland is about to take a significant step forward thanks to funding from Friends of ANCHOR.
The charity is the key funder for a new digital dataset that will drive the potential of artificial intelligence (AI) in diagnosing breast cancer.
Experts say the treasure trove of anonymised patient information, paired with cutting edge tech that’s already been assessed at NHS Grampian, makes Aberdeen Royal Infirmary an international leader in breast imaging research and service development.
Professor Lesley Anderson, Chair in Health Data Science at the University of Aberdeen, put forward the funding request to Friends of ANCHOR, which signed off £69,408 to pay for the upgraded information, which will be coined the ‘ANCHOR breast imaging dataset’.
'Increasing cancer detection'
Prof Anderson said: “A comprehensive dataset incorporating medical images is crucial to evaluate and develop AI technologies to improve cancer detection. NHS Grampian will be the first site in Scotland to host this type of dataset enabling AI tools to be developed and tested to increase cancer detection and enable more efficient use of resources.
“This funding will lead to substantial advancements in breast cancer screening in the Grampian region and beyond.”
The data within the digital vault - scans of people who have been diagnosed with breast cancer - is already in storage, but it needs to be optimised to allow AI to do its job properly.
Friends of ANCHOR’s funding will do just that, anonymising the scans and linking them with healthcare and clinical data to enhance how AI interprets the information. Thanks to the funding, the dataset will also be expanded to include an additional eight years’ worth of scan images.
The change to the dataset is possible thanks to Friends of ANCHOR’s financial commitment and a £5,000 contribution from NHS Grampian Charity.
The funding win follows the news that an AI tool tested by NHS Grampian clinicians and the University of Aberdeen AI evaluation team identified tiny signs of breast cancer that had been missed by doctors.
The artificial intelligence tool, called ‘Mia’ looked at mammograms from more than 10,000 women. Using Mia along with a human reader would have detected an additional 11 cancers that hadn’t been picked up by the naked eye without bringing any more women back for further investigation.
'Dataset will feed AI fine-tuned information'
Dr Gerald Lip, Clinical Director for the North East of Scotland Breast Screening Programme, who led the project, explained how Friends of ANCHOR’s funding will help spearhead further improvements.
He said: “The success of artificial intelligence in healthcare will rely heavily on a high-quality local dataset. In other words, the data that’s fed to AI needs to be fine-tuned first, and matched with appropriate local healthcare information, right down to what kind of imaging equipment has taken a scan.
“Being able to establish such a database is unique and timely, ensuring the north east of Scotland gets the best tools for its needs.”
It’s an investment Friends of ANCHOR is pleased to support, said chief executive Sarah-Jane Hogg.
“AI has the potential to bring substantial gains in diagnosis and patient care, and Friends of ANCHOR is very pleased to be playing a part in optimising the technology in NHS Grampian.
“It’s thanks to our fundraisers and donors that Friends of ANCHOR can support this request, and I’d like to thank them for making investments like this one possible.
“Prof Anderson, Dr Lip and their colleagues in the Aberdeen collaboration are at the vanguard and we’re continually impressed by their pioneering approach.”
Friends of ANCHOR has funded other initiatives tasked with improving breast cancer outcomes too. Enhanced contrast mammography is a method supported by £25,000 of charity funding, which will in future take patient wait times from 60 days to same-day answers for complex cases. During the pandemic in 2020, the charity also fast-tracked funding for radiofrequency tags and detector probes. A Scotland-first at the time, the tags increase surgical accuracy in operating theatres, and streamlined a patient’s pathway to surgery.