Study finds using AI to read mammograms reduces risk of developing aggressive breast cancer
Some aggressive breast cancers can be difficult to detect and don’t always show up on mammograms, but a new study from Sweden finds that artificial intelligence can help.
Mammograms are used to detect breast cancers earlier, when they are easier to manage and treat.. When radiologists used artificial intelligence to aid their readings, breast cancer detection rates increased, resulting in less aggressive and advanced cases, a large trial in Sweden has found.
The study – published in Thursday’s issue of the medical journal The Lancet – More than 100,000 women underwent mammography screening. Half were supported by AI and the remaining mammograms were reviewed by two different radiologists, a standard practice in much of Europe known as double reading. It is not commonly used in Canada, where a radiologist usually examines the mammogram.
The study looked at rates of interval cancer, a term doctors use for aggressive tumors that show up between routine mammograms. They can be difficult to detect and studies have shown that they They are more likely to be aggressive With a poor prognosis.
The study showed that in the groups where AI screening was implemented, the rate of interval cancer decreased by 12 percent.
The lead author, Dr. Christina Long, said the major reduction is “a huge sigh of relief.”
“The great results we’ve seen (are) very promising,” said Lang, a breast radiologist and associate professor of diagnostic radiology at Lynn University in Malmo, Sweden.
Screening improvements
In 2024, an estimated 30,500 Canadians will be diagnosed breast cancerAccording to the Public Health Agency of Canada, which also noted that the chance of survival for at least five years after diagnosis was about 89 per cent.
Last year, the Canadian Task Force for Preventive Health Care recommended people should be able to get mamOgrams start from the age of 40. In practice such recommendations have not been widely adopted, although some Canadian provinces are moving toward that standard.
In Sweden, population-based breast cancer screening is offered to people aged 40 to 74 years.
The frequency of screening also varies – in most Canadian jurisdictions, screening typically occurs every 24 months. In Sweden it is between 18 and 24 months.
A new study shows artificial intelligence is showing major promise in detecting breast cancer, but researchers admit it has limitations. The technology can produce false positives – flagging too many mammograms as abnormal when there is actually no cancer.
During the two-year study, AI-assisted mammograms were divided into two different groups. Those who were low risk required only one radiologist to examine them, while those considered high risk required two.
The researchers reported that numerically, AI-assisted screening resulted in 11 fewer late cancers than standard screening (82 versus 93, or 12 percent).
“This is really a way to improve overall screening testing,” Long said.
He acknowledged that the study showed a reduction in late cancer, but long-term studies are needed to find out how AI-assisted screening could affect mortality rates.
Screening for the study took place at a single center in Sweden, which the researchers acknowledged is a limitation. Second, the race and ethnicity of the participants were not recorded.
next steps, long That said, the cost-effectiveness for Swedish researchers will have to be determined.
AI could save public health dollars
Dr. Jean Seely, Chief of Breast Imaging at the Ottawa Hospital and a A University of Ottawa radiology professor says the value of AI-assisted mammograms is clear.
Seely, who was not involved in the study, said interval cancer is typically stage 3, meaning treatment alone would cost more than $100,000 to $200,000 per patient.
“This is something that is not only saving lives, but also saving public health dollars,” Seely said.
He has used research funds to buy A.I.Software to assist in reading mammograms.
“It’s like re-reading and it helps me focus,” Seely said. “We’re all human, right? We can get distracted sometimes.”
Oncologist says investment in AI is needed
Montreal oncologist Dr. Gerald Batiste, who was not involved in the study, says there’s now no question that AI-assisted mammography is a valuable tool.
His hospital, Jewish General of Montreal, is currently considering AI options for breast imaging.
“We must invest in AI solutions,” Batiste said. “These studies confirm that they detect things that…the human eye cannot see.”
As for what happens next, Batiste called on health ministers to talk to their counterparts in charge of innovation and technology.
“Health care and regulators need to move at the speed of cancer, not the speed of bureaucracy,” he said.
