Aberdeen Royal Infirmary’s AI Trial Aids Radiologists in Analyzing Mammograms to Detect Breast Cancer
A cutting-edge trial at Aberdeen Royal Infirmary is currently underway to investigate the potential of artificial intelligence (AI) in assisting radiologists with the evaluation of thousands of mammograms annually.
The trial has already yielded promising results, as it successfully identified early-stage breast cancer in June, a healthcare assistant and participant in the study. Consequently, she is now scheduled for surgery.
Mammograms, which utilize low-level X-rays, are employed in breast cancer screenings to detect and monitor minute changes that may be undetectable through physical examination.
According to the NHS, mammograms contribute to saving approximately 1,300 lives each year in the UK.
While the number of women attending routine breast screenings in Scotland has increased over the past three years until 2022, the pool of radiologists available to review the results is diminishing.
Aberdeen’s AI trial
Significant Increase in Screen-Detected Breast Cancers in Scotland, According to Public Health Scotland Data
Public Health Scotland (PHS) data reveals a notable rise in screen-detected breast cancers among women of all ages in Scotland, reaching 1,830 cases between 2021 and 2022.
Radiologists evaluate an average of 5,000 mammograms annually, leading to approximately 250 to 300 patients being recalled for further examination, with 30 to 40 of them requiring closer attention.
Dr. Gerald Lip, clinical director of the North East Scotland Breast Screening Programme, expressed concerns about the possibility of missing cancers with such numbers, highlighting the need for improved detection methods.
NHS Grampian’s Gemini project, a collaborative effort involving the NHS, the University of Aberdeen, and private industry, was cited in Scotland’s AI Strategy upon its launch in 2021.
The trial utilizes Kheiron Medical Technologies’ AI model, Mia, with Microsoft providing cloud computing services to support the project.
As per current guidelines set by the National Screening Council, AI deployment in screenings is not yet automated. Therefore, Dr. Lip and other radiologists are experimenting with AI as an additional check at the end of mammogram scan reviews.
June, a participant in the trial who has undergone similar surgery previously, underwent a biopsy to test a small portion of breast tissue after Dr. Lip explained how the AI tool aided in identifying an area of concern.
By clicking a button, radiologists can examine differences highlighted by the AI between two scans.
Dr. Lip emphasized the importance of detecting issues when they are small, rather than allowing them to grow. He pointed out an area flagged by the AI software as the primary cause for concern.
June, reflecting on her experience, expressed that using AI instead of relying solely on human analysis made the process feel less invasive. She appreciated the sense of privacy provided by the AI technology.
Following the biopsy, June received the news that she once again requires surgery due to an early-stage cancer diagnosis. Despite her reservations, she acknowledged the reassurance that comes with early detection.
“The biopsy showed that I do have an early-stage cancer, they’ve certainly caught it at an earlier stage this time,” June shared. “But because I’ve had previous history with it, I’m going to go in and have a mastectomy. It’s not treatment I want to have. But at the same time, it’s reassuring that it’s being caught.”