Artificial intelligence (AI) is being tested at Aberdeen Royal Infirmary to help radiologists review thousands of mammograms annually.
June, a trial participant and healthcare assistant, will have surgery due to the pilot’s assistance in identifying her early-stage breast cancer.
Low-level X-rays, called mammograms, are used in breast cancer screenings to track and find changes that are too small to be felt or seen.
According to the NHS, around 1,300 lives are saved annually in the UK.
And even while more women in Scotland accepted invitations to undergo routine breast screenings throughout the three years leading up to 2022, fewer radiologists are available to analyze the results.
Artificial intelligence (AI), a technology that enables computers to carry out specific tasks that would traditionally need human intelligence, is already widely used in a variety of fields.
The more practical applications of AI are already being seen in healthcare, despite high-profile academics’ recent warnings that the technology could wipe out humans.
Many scientists and medical professionals view AI as a strong tool to complement practitioners rather than replace them because of its ability to speed up the process of medication and disease discovery.
Between 250 and 300 patients will be called back out of the 5,000 mammograms that radiologists examine on average each year, with 30 to 40 of them needing more in-depth care.
According to Dr Gerald Lip, clinical director of the North East Scotland Breast Screening Programme, “There is a possibility that you could miss cancers with that number.”
Scotland’s AI Strategy, which was released in 2021, included a reference to NHS Grampian’s Gemini initiative, a partnership between the NHS, the University of Aberdeen, and private businesses.
The AI model Mia utilized in the trial was created by Kheiron Medical Technologies, and Microsoft provided the cloud computing facilities needed to run it.
Dr. Lip and other radiologists are testing artificial intelligence (AI) as an additional check at the conclusion of mammography scan evaluations because National Screening Counsel regulations currently prohibit the automatic deployment of AI in screenings.
Following Dr. Lip’s explanation of how the AI tool assisted in identifying a problem area, June, a trial participant who had previously undergone a procedure identical to this, had a biopsy performed to remove a small portion of her breast tissue for testing.
Right now, he said, “What we’re seeing is a lady who has mammograms on both her left and right sides; you’re looking for differences.”
Radiologists can view and examine changes between the two images that the AI found by pressing a button.
Dr. Lip indicated a particular location that the AI program had marked as the primary cause for concern.
“In screening, you want to pick up things when they are small before they become big,” he continued.
June explained to Click a few weeks after her biopsy that employing AI rather than an additional set of human eyes made the procedure feel less invasive.
You are aware that people are viewing your photographs on the screen, she said. “However, when it’s an AI, that slight sense that someone is watching has vanished.”
As a result of June’s biopsy results, she will require more surgery.
They have definitely found it at an earlier stage this time, June said, adding that the biopsy revealed that she does have early-stage cancer. But I’m going to get a mastectomy since I’ve had a history with it.I don’t want to receive that kind of care. However, the fact that it is being caught is also comforting.
A significant study of Scotland’s breast screening program from the previous year revealed that many radiology and advanced practice staff members are approaching or have reached retirement age.
The service, which depends on “super reader” radiologists to examine a lot of findings, is at risk of becoming “vulnerable” due to the decline of these radiologists, according to the report.
The Scottish government’s research found that using AI to replace one human reader might “cover half of the screening image reading burden of approximately 1.72 million images read each year,” where two radiologists are required to read and report results.
But may the technologies Aberdeen is testing eventually take the place of employees?
“I believe the purpose of this evaluation is to determine the best way we can work with AI, whether it be replacing one of the radiologists, part-reading some of the routine mammograms, or improving our cancer detection as a safety net,” Dr. Lip stated.
Over 30 NHS trusts in the UK will deploy the technology, according to Peter Kecskemethy, co-founder of Kheiron.
The technology appears to continue playing a significant role in assisting doctors in saving lives, with trusts in England actively investigating how AI may enable better, faster outcomes for breast cancer patients.
