2023-06-12 04:56:31
- Mark Cislak and Liv McMahon
- BBC Click
A mammogram uses low-level X-rays to detect breast cancer
An experiment is currently underway at the Aberdeen Royal Infirmary in Scotland to see if artificial intelligence can help radiologists who examine thousands of mammograms each year to detect breast cancer.
This experience helped discover that Ms. Gowon, a health worker and one of the participants in the study, had early-stage breast cancer and is now preparing to undergo surgery to remove one of her breasts.
The mammogram uses a low level of X-rays to detect breast cancer, by monitoring any changes that are difficult to feel or see because they are very small.
According to the UK’s National Health Service (NHS), mammograms help save the lives of 1,300 women every year.
While the number of women undergoing routine breast cancer screening following receiving an invitation from the NHS has risen in Scotland in the three years leading up to 2022, the number of radiologists checking the results is shrinking.
What is artificial intelligence?
It is noteworthy that artificial intelligence technology, which involves computers carrying out certain tasks that usually require human intelligence, is already widely used in a number of industries.
And while statements by prominent experts in the field expressing their fears that artificial intelligence may lead to the annihilation of humanity recently made headlines, the most practical applications of the technology have already proven their effectiveness in the health sector.
Its potential to speed up disease discovery and drug manufacture means that many scientists and clinicians see AI as a powerful tool that can help rather than replace specialists.
Aberdeen experience
image copyright Grampian NHS
A trial using artificial intelligence in breast cancer screening is currently underway at the Royal Aberdeen Hospital.
According to Public Health Scotland data, the number of breast cancer cases detected through mammograms for women of all ages rose to 1,830 between 2021 and 2022 in Scotland.
Of the average 5,000 women whose mammogram results are reviewed by radiologists each year, between 250 and 300 are re-examined, and of these, regarding 30 to 40 are in need of more accurate examinations or treatment.
“There is a chance that some cancers will be missed because of this high number,” says Dr Gerard Leib, medical director of breast cancer screening in North East Scotland.
The GEMMINI project – a joint project between the NHS, the University of Aberdeen and the private sector aiming to understand the impact of the use of AI on breast cancer screening – is mentioned in the AI strategy Scotland launched in 2021.
Kheiron Medical Technologies developed an artificial intelligence model called “Mia”, which was used in the Aberdeen hospital experiment, and Microsoft provided the necessary cloud computing services to support it.
Since the current rules set by the National Medical Examination Commission prohibit the automatic use of AI in a medical examination, Dr. Lipp and other radiologists are currently experimenting with it as an additional test in the final stage of reviewing mammograms.
One of the participants in the trial, who had previously undergone surgery related to breast cancer, Joon, had breast tissue removed following Dr. Lip told her how the AI tool had helped her detect a disturbing change.
Dr. Lip explains to Joon the changes the AI software detected in her mammograms
Dr Lieb explained to BBC Click how the AI works using the results of an anonymous woman’s mammogram.
“What we’re seeing now are two mammograms taken from the right side and the left side, and what we’re looking for are the differences.”
With the press of a button, radiologists can see and examine the differences the AI has detected between the two images.
Dr. Lieb pointed to an area that the AI program had circled and identified it as the main area of concern.
Areas of concern of breast tissue in a mammogram image, circled by the AI program
“When testing is done, the aim is to identify problems in their infancy and before they escalate,” he added.
A few weeks following her breast biopsy, June told Click that using artificial intelligence instead of a second person made the procedure seem less invasive.
“You know that the images are on the screen, and people are looking at them. When it uses artificial intelligence, that feeling that someone is looking at those images disappears,” Joon said.
The results of a biopsy of Goon’s tissue indicated that she would need to undergo surgery once more.
Jewon added, “The biopsy showed that I had early-stage cancer, so they must have caught it at an earlier stage this time. Because I have a previous history.” [مع السرطان]I’m going to have a mastectomy.”
“It’s not the treatment I want, but at the same time it’s reassuring that the infection has been detected.”
“loss Chronic in crews
An extensive review of the breast cancer screening program in Scotland published last year suggested that an overwhelming number of radiologists were nearing or already reaching retirement age.
She stated that the reduction in the number of highly skilled radiologists who had to be relied upon to review so many results meant that the department risked being left in a “fragile position”.
The Royal College of Radiologists has also warned that the UK is facing a “chronic staff shortage”, with patients now being forced to wait for far too long before essential cancer tests or treatments.
A radiologist reviews mammograms at the Royal Aberdeen Infirmary
Given that two radiologists are needed to read and report the results, the Scottish government’s review stated that replacing a human reader with artificial intelligence might do half the task of reading regarding 1.72 million scans a year.
But can the technology that Aberdeen Hospital is currently testing replace humans?
“In my opinion, the goal of that assessment is to determine the best way we can take advantage of AI, whether it is to replace one of the radiologists, to do a partial read of some normal mammograms, or to act as a network,” says Dr. Leib. Aman’s goal is to improve the detection of cancer.”
Khyron co-founder Peter Kiskimethe says the technology will be used by more than 30 NHS departments across the UK and millions of people will be able to benefit from it.
As some health departments in England are currently studying how to take advantage of artificial intelligence to give better and faster results for breast cancer patients, it seems that technology will continue to play a major role in helping doctors save patients’ lives.
1686611578
#Breast #cancer #artificial #intelligence #doctors #detect