Artificial Intelligence is getting better than humans at detecting breast cancer
Since the early 2000s, AI systems have eliminated more than 1.7 million jobs, and that number will only increase as AI improves. Some research estimates that by 2025, AI will eliminate more than 85 million jobs.
But for all the talk about job security, AI is also proving to be a powerful tool in healthcare—specifically, cancer detection. One recently published study has shown that, remarkably, artificial intelligence was able to detect 20 percent more cancers in imaging scans than radiologists alone.
Published in The Lancet Oncology, the study analyzed the scans of 80,000 Swedish women with a moderate hereditary risk of breast cancer who had undergone a mammogram between April 2021 and July 2022. Half of these scans were read by AI and then a radiologist to double-check the findings. The second group of scans was read by two researchers without the help of AI. (Currently, the standard of care across Europe is to have two radiologists analyze a scan before diagnosing a patient with breast cancer.)
The study showed that the AI group detected cancer in 6 out of every 1,000 scans, while the radiologists detected cancer in 5 per 1,000 scans. In other words, AI found 20 percent more cancers than the highly-trained radiologists.
Scientists have been using MRI images (like the ones pictured here) to train artificial intelligence to detect cancers earlier and with more accuracy. Here, MIT's AI system, MIRAI, looks for patterns in a patient's mammograms to detect breast cancer earlier than ever before. news.mit.edu
But even though the AI was better able to pinpoint cancer on an image, it doesn’t mean radiologists will soon be out of a job. Dr. Laura Heacock, a breast radiologist at NYU, said in an interview with CNN that radiologists do much more than simply screening mammograms, and that even well-trained technology can make errors. “These tools work best when paired with highly-trained radiologists who make the final call on your mammogram. Think of it as a tool like a stethoscope for a cardiologist.”
AI is still an emerging technology, but more and more doctors are using them to detect different cancers. For example, researchers at MIT have developed a program called MIRAI, which looks at patterns in patient mammograms across a series of scans and uses an algorithm to model a patient's risk of developing breast cancer over time. The program was "trained" with more than 200,000 breast imaging scans from Massachusetts General Hospital and has been tested on over 100,000 women in different hospitals across the world. According to MIT, MIRAI "has been shown to be more accurate in predicting the risk for developing breast cancer in the short term (over a 3-year period) compared to traditional tools." It has also been able to detect breast cancer up to five years before a patient receives a diagnosis.
The challenges for cancer-detecting AI tools now is not just accuracy. AI tools are also being challenged to perform consistently well across different ages, races, and breast density profiles, particularly given the increased risks that different women face. For example, Black women are 42 percent more likely than white women to die from breast cancer, despite having nearly the same rates of breast cancer as white women. Recently, an FDA-approved AI device for screening breast cancer has come under fire for wrongly detecting cancer in Black patients significantly more often than white patients.
As AI technology improves, radiologists will be able to accurately scan a more diverse set of patients at a larger volume than ever before, potentially saving more lives than ever.
New Tech Can Predict Breast Cancer Years in Advance
Every two minutes, a woman is diagnosed with breast cancer. The question is, can those at high risk be identified early enough to survive?
New AI software has predicted risk equally well in both white and black women for the first time.
The current standard practice in medicine is not exactly precise. It relies on age, family history of cancer, and breast density, among other factors, to determine risk. But these factors do not always tell the whole story, leaving many women to slip through the cracks. In addition, a racial gap persists in breast cancer treatment and survival. African-American women are 42 percent more likely to die from the disease despite relatively equal rates of diagnosis.
But now those grim statistics could be changing. A team of researchers from MIT's Computer Science and Artificial Intelligence Laboratory have developed a deep learning model that can more accurately predict a patient's breast cancer risk compared to established clinical guidelines – and it has predicted risk equally well in both white and black women for the first time.
The Lowdown
Study results published in Radiology described how the AI software read mammogram images from more than 60,000 patients at Massachusetts General Hospital to identify subtle differences in breast tissue that pointed to potential risk factors, even in their earliest stages. The team accessed the patients' actual diagnoses and determined that the AI model was able to correctly place 31 percent of all cancer patients in the highest-risk category of developing breast cancer within five years of the examination, compared to just 18 percent for existing models.
"Each image has hundreds of thousands of pixels identifying something that may not necessarily be detected by the human eye," said MIT professor Regina Barzilay, one of the study's lead authors. "We all have limited visual capacities so it seems some machines trained on hundreds of thousands of images with a known outcome can capture correlations the human eye might not notice."
Barzilay, a breast cancer survivor herself, had abnormal tissue patterns on mammograms in 2012 and 2013, but wasn't diagnosed until after a 2014 image reading, illustrating the limitations of human processing alone.
MIT professor Regina Barzilay, a lead author on the new study and a breast cancer survivor herself.
(Courtesy MIT)
Next up: The MIT team is looking at training the model to detect other cancers and health risks. Barzilay recalls how a cardiologist told her during a conference that women with heart diseases had a different pattern of calcification on their mammograms, demonstrating how already existing images can be used to extract other pieces of information about a person's health status.
Integration of the AI model in standard care could help doctors better tailor screening and prevention programs based on actual instead of perceived risk. Patients who might register as higher risk by current guidelines could be identified as lower risk, helping resolve conflicting opinions about how early and how often women should receive mammograms.
Open Questions: While the results were promising, it's unknown how well the model will work on a larger scale, as the study looked at data from just one institution and used mammograms supplied by just one hospital. Some risk factor information was also unavailable for certain patients during the study, leaving researchers unable to fully compare the AI model's performance to that of the traditional standard.
One incentive to wider implementation and study, however, is the bonus that no new hardware is required to use the AI model. With other institutions now showing interest, this software could lead to earlier routine detection and treatment of breast cancer — resulting in more lives saved.
Sarah Mancoll was 22 years old when she noticed a bald spot on the back of her head. A dermatologist confirmed that it was alopecia aerata, an autoimmune disorder that causes hair loss.
Of 213 new drugs approved from 2003 to 2012, only five percent included any data from pregnant women.
She successfully treated the condition with corticosteroid shots for nearly 10 years. Then Mancoll and her husband began thinking about starting a family. Would the shots be safe for her while pregnant? For the fetus? What about breastfeeding?
Mancoll consulted her primary care physician, her dermatologist, even a pediatrician. Without clinical data, no one could give her a definitive answer, so she stopped treatment to be "on the safe side." By the time her son was born, she'd lost at least half her hair. She returned to her Washington, D.C., public policy job two months later entirely bald—and without either eyebrows or eyelashes.
After having two more children in quick succession, Mancoll recently resumed the shots but didn't forget her experience. Today, she is an advocate for including more pregnant and lactating women in clinical studies so they can have more information about therapies than she did.
"I live a very privileged life, and I'll do just fine with or without hair, but it's not just about me," Mancoll said. "It's about a huge population of women who are being disenfranchised…They're invisible."
About 4 million women give birth each year in the United States, and many face medical conditions, from hypertension and diabetes to psychiatric disorders. A 2011 study showed that most women reported taking at least one medication while pregnant between 1976 and 2008. But for decades, pregnant and lactating women have been largely excluded from clinical drug studies that rigorously test medications for safety and effectiveness.
An estimated 98 percent of government-approved drug treatments between 2000 and 2010 had insufficient data to determine risk to the fetus, and close to 75 percent had no human pregnancy data at all. All told, of 213 new pharmaceuticals approved from 2003 to 2012, only five percent included any data from pregnant women.
But recent developments suggest that could be changing. Amid widespread concerns about increased maternal mortality rates, women's health advocates, physicians, and researchers are sensing and encouraging a cultural shift toward protecting women through responsible research instead of from research.
"The question is not whether to do research with pregnant women, but how," Anne Drapkin Lyerly, professor and associate director of the Center for Bioethics at the University of North Carolina at Chapel Hill, wrote last year in an op-ed. "These advances are essential. It is well past time—and it is morally imperative—for research to benefit pregnant women."
"In excluding pregnant women from drug trials to protect them from experimentation, we subject them to uncontrolled experimentation."
To that end, the American College of Obstetricians and Gynecologists' Committee on Ethics acknowledged that research trials need to be better designed so they don't "inappropriately constrain the reproductive choices of study participants or unnecessarily exclude pregnant women." A federal task force also called for significantly expanded research and the removal of regulatory barriers that make it difficult for pregnant and lactating women to participate in research.
Several months ago, a government change to a regulation known as the Common Rule took effect, removing pregnant women as a "vulnerable population" in need of special protections -- a designation that had made it more difficult to enroll them in clinical drug studies. And just last week, the U.S. Food and Drug Administration (FDA) issued new draft guidances for industry on when and how to include pregnant and lactating women in clinical trials.
Inclusion is better than the absence of data on their treatment, said Catherine Spong, former chair of the federal task force.
"It's a paradox," said Spong, professor of obstetrics and gynecology and chief of maternal fetal medicine at University of Texas Southwestern Medical Center. "There is a desire to protect women and fetuses from harm, which is translated to a reluctance to include them in research. By excluding them, the evidence for their care is limited."
Jacqueline Wolf, a professor of the history of medicine at Ohio University, agreed.
"In excluding pregnant women from drug trials to protect them from experimentation, we subject them to uncontrolled experimentation," she said. "We give them the medication without doing any research, and that's dangerous."
Women, of course, don't stop getting sick or having chronic medical conditions just because they are pregnant or breastfeeding, and conditions during pregnancy can affect a baby's health later in life. Evidence-based data is important for other reasons, too.
Pregnancy can dramatically change a woman's physiology, affecting how drugs act on her body and how her body acts or reacts to drugs. For instance, pregnant bodies can more quickly clear out medications such as glyburide, used during diabetes in pregnancy to stabilize high blood-sugar levels, which can be toxic to the fetus and harmful to women. That means a regular dose of the drug may not be enough to control blood sugar and prevent poor outcomes.
Pregnant patients also may be reluctant to take needed drugs for underlying conditions (and doctors may be hesitant to prescribe them), which in turn can cause more harm to the woman and fetus than had they been treated. For example, women who have severe asthma attacks while pregnant are at a higher risk of having low-birthweight babies, and pregnant women with uncontrolled diabetes in early pregnancy have more than four times the risk of birth defects.
Current clinical trials involving pregnant women are assessing treatments for obstructive sleep apnea, postpartum hemorrhage, lupus, and diabetes.
For Kate O'Brien, taking medication during her pregnancy was a matter of life and death. A freelance video producer who lives in New Jersey, O'Brien was diagnosed with tuberculosis in 2015 after she became pregnant with her second child, a boy. Even as she signed hospital consent forms, she had no idea if the treatment would harm him.
"It's a really awful experience," said O'Brien, who now is active with We are TB, an advocacy and support network. "All they had to tell me about the medication was just that women have been taking it for a really long time all over the world. That was the best they could do."
More and more doctors, researchers and women's health organizations and advocates are calling that unacceptable.
By indicating that filling current knowledge gaps is "a critical public health need," the FDA is signaling its support for advancing research with pregnant women, said Lyerly, also co-founder of the Second Wave Initiative, which promotes fair representation of the health interests of pregnant women in biomedical research and policies. "It's a very important shift."
Research with pregnant women can be done ethically, Lyerly said, whether by systematically collecting data from those already taking medications or enrolling pregnant women in studies of drugs or vaccines in development.
Current clinical trials involving pregnant women are assessing treatments for obstructive sleep apnea, postpartum hemorrhage, lupus, and diabetes. Notable trials in development target malaria and HIV prevention in pregnancy.
"It clearly is doable to do this research, and test trials are important to provide evidence for treatment," Spong said. "If we don't have that evidence, we aren't making the best educated decisions for women."