Clinics across the country are starting to offer patients a new service: having their mammograms read not only by a radiologist, but also by an artificial intelligence model. Hospitals and companies that provide these tools tout their ability to speed up radiologists’ work and detect cancer earlier than standard mammograms alone.
Currently, mammograms identify about 87 percent of breast cancers. They are more likely to miss cancer in younger women and those with dense breasts. They sometimes lead to false positives that require additional testing to rule out cancer, and can also give rise to precancerous conditions that may never cause serious problems but nonetheless lead to treatment because it is impossible to predict the risk of not treating them.
“It’s not perfect science by any means,” said Dr. John Lewin, chief of breast imaging at Smilow Cancer Hospital and Yale Cancer Center.
Experts are excited about the prospect of improving the accuracy of breast cancer screening, which 300,000 women are diagnosed with each year in the United States. But they also worry about whether these AI tools will work well with a wide range of patients and whether they can significantly improve breast cancer survival.
How does AI analysis work?
Mammograms contain a wealth of information about breast tissues and ducts. Certain patterns, such as bright white spots with irregular edges, may be a sign of cancer. In comparison, thin white lines can indicate calcifications that may be benign or require further testing. Other patterns may be difficult for humans to differentiate from normal breast tissue.
AI models can, in some cases, “see what we can’t see,” said Dr. Katerina Dodelzon, a radiologist specializing in breast imaging at NewYork-Presbyterian/Weill Cornell Medical Center.
When an image is analyzed through an AI program, the software highlights suspicious areas that need special attention from a radiologist. Some models can also evaluate images to help busy radiologists prioritize which exams to review first.
“I easily read 100 screening mammograms a day,” said Dr. Carolyn Malone, a radiologist in the breast division of the John Theurer Cancer Center at Hackensack University Medical Center. “I can start reading the ones that the AI thinks are more complex.”
In one of the largest studies of AI mammography, a model used in Sweden improved breast cancer detection by 20 percent. In a trial of 80,000 women, the software detected six cases of cancer per 1,000 women, while radiologists found five per 1,000 women.
A 2022 Danish study showed that an AI model also reduced false positives, meaning fewer women had to return to the doctor for additional testing after a mammogram identified a suspicious spot.
But it’s unclear whether AI analysis will actually reduce deaths from breast cancer, or whether it will simply inflate survival figures by detecting more cancers earlier. And radiologists don’t know how the European results will translate to the United States, or how well the models will work in a more diverse population.
“There is a need for more diverse training and testing of these AI tools and algorithms in order to develop them across different races and ethnicities,” said Dr. Dodelzon. “AI is just a tool that learns based on what it sees.”
Some experts also worry about using these tools before they have been thoroughly vetted, drawing comparisons to computer-aided detection technology hailed in the 1980s as a way to detect breast cancers earlier. breast. A major study later showed that the technology did not make mammogram results more accurate.
When it comes to AI analysis of mammograms, “we may not know for a few years if our performance has declined,” Dr. Lewin said.
There are also some things that AI can’t yet do well, like differentiating between surgical scars and tumors. “For that, you just need a human,” Dr. Malone said. Radiologists, especially those trained in breast imaging, can rely on patients’ medical histories and their own expertise to identify these abnormalities, she said.
Is it worth paying for an AI mammogram?
The Food and Drug Administration has approved about two dozen AI products for mammography. Some of them are deployed to patients in a small number of clinics and tested by other hospitals who wish to ensure the value provided by these tools before offering them to patients.
There are currently no billing codes that radiologists can use to bill insurers for the technology. This means that some centers may increase the cost for patients, charging between $40 and $100 out of pocket for an AI analysis. Other hospitals may absorb the cost and offer the additional analysis for free. Still others may hold the technology for research until they are more sure of the value it can bring to patients.
It will take some time for AI to become part of routine care, which would lead insurance companies to consider reimbursing their costs. Until then, most patients don’t need AI for their mammograms, Dr. Dodelzon said, although it could provide more reassurance to those who are particularly worried about their results.
News Source : www.nytimes.com
Gn Health
