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.
Deaf Scientists Just Created Over 1000 New Signs to Dramatically Improve Ability to Communicate
For the deaf, talent and hard work may not be enough to succeed in the sciences. According to the National Science Foundation, deaf Americans are vastly underrepresented in the STEM fields, a discrepancy that has profound economic implications.
The problem with STEM careers for the deaf and hard-of-hearing is that there are not enough ASL signs available.
Deaf and hard-of-hearing professionals in the sciences earn 31 percent more than those employed in other careers, according to a 2010 study by the National Technical Institute for the Deaf (NTID) in Rochester, N.Y., the largest technical college for deaf and hard-of-hearing students. But at the same time, in 2017, U.S. students with hearing disabilities earned only 1.1 percent of the 39,435 doctoral degrees awarded in science and engineering.
One reason so few deaf students gravitate to science careers and may struggle to complete doctoral programs is the communication chasm between deaf and hard-of-hearing scientists and their hearing colleagues.
Lorne Farovitch is a doctoral candidate in biomedical science at the University of Rochester of New York. Born deaf and raised by two deaf parents, he communicated solely in American Sign Language (ASL) until reaching graduate school. There, he became frustrated at the large chunk of his workdays spent communicating with hearing lab mates and professors, time he would have preferred spending on his scientific work.
The problem with STEM careers for the deaf and hard-of-hearing is that there are not enough ASL signs available, says Farovitch. Names, words, or phrases that don't exist in ASL must be finger spelled — the signer must form a distinct hand shape to correspond with each letter of the English alphabet, a tedious and time-consuming process. For instance, it requires 12 hand motions to spell out the word M-I-T-O-C-H-O-N-D-R-I-A. Imagine repeating those motions countless times a day.
To bust through this linguistic quagmire, Farovitch, along with a team of deaf STEM professionals, linguists, and interpreters, have been cooking up signs for terms like Anaplasma phagocytophilum, the tick-borne bacterium Farovitch studies. The sign creators are then videotaped performing the new signs. Those videos are posted on two crowd-sourcing sites, ASLcore.org and ASL Clear.
The beauty of ASL is you can express an entire concept in a single sign, rather than by the name of a word.
"If others don't pick it up and use it, a sign goes extinct," says Farovitch. Thus far, more than 1,000 STEM terms have been developed on ASL Clear and 500 vetted and approved by the deaf STEM community, according to Jeanne Reis, project director of the ASL Clear Project, based at The Learning Center for the Deaf in Framingham, Mass.
The beauty of ASL is you can express an entire concept in a single sign, rather than by the name of a word. The signs are generally intuitive and wonderfully creative. To express "DNA" Farovitch uses two fingers of each hand touching the tips of the opposite hand; then he draws both the hands away to suggest the double helix form of the hereditary material present in most organisms.
"If you can show it, you can understand the concept better,'' says the Canadian-born scientist. "I feel I can explain science better now."
The hope is that as ASL science vocabulary expands more, deaf and hard-of-hearing students will be encouraged to pursue the STEM fields. "ASL is not just a tool; it's a language. It's a vital part of our lives," Farovitch explains through his interpreter.
The deaf community is diverse—within and beyond the sciences. Sarah Latchney, PhD, an environmental toxicologist, is among the approximately 90 percent of deaf people born to hearing parents. Hers made sure she learned ASL at an early age but they also sent Latchney to a speech therapist to learn to speak and read lips. Latchney is so adept at both that she can communicate one-on-one with a hearing person without an interpreter.
Like Favoritch, Latchney has developed "conceptually accurate" ASL signs but she has no plans to post them on the crowd-sourcing sites. "I don't want to fix [my signs]; it works for me," she explains.
Young scientists like Farovitch and Latchney stress the need for interpreters who are knowledgeable about science. "When I give a presentation I'm a nervous wreck that I'll have an interpreter who may not have a science background," Latchney explains. "Many times what I've [signed] has been misinterpreted; either my interpreter didn't understand the question or didn't frame it correctly."
To enlarge the pool of science-savvy interpreters, the University of Rochester will offer a new masters degree program: ASL Interpreting in Medicine and Science (AIMS), which will train interpreters who have a strong background in the biological sciences.
Since the Americans with Disabilities Act was enacted in 1990, opportunities in higher education for deaf and hard-of-hearing students have opened up in the form of federally funded financial aid and the creation of student disability services on many college campuses. Still, only 18 percent of deaf adults have graduated from college, compared to 33 percent of the general population, according to a survey by the U.S. Census Bureau in 2015.
The University of Rochester and the Rochester Institute of Technology, home to NTID, have jointly created two programs to increase the representation of deaf and hard-of-hearing professionals in the sciences. The Rochester Bridges to the Doctorate Program, which Farovitch is enrolled in, prepares deaf scholars for biomedical PhD programs. The Rochester Postdoctoral Partnership readies deaf postdoctoral scientists to successfully attain academic research and teaching careers. Both programs are funded by the National Institutes of Science. In the last five years, the University of Rochester has gone from zero deaf postdoctoral and graduate students to nine.
"Deafness is not a problem, it's just a difference."
It makes sense for these two private universities to support strong programs for the deaf: Rochester has the highest per capita population of deaf or hard-of-hearing adults younger than 65 in the nation, according to the U.S. Census. According to the U.S. Department of Education, there are about 136,000 post-secondary level students who are deaf or hard of hearing.
"Deafness is not a problem, it's just a difference," says Farovitch. "We just need a different way to communicate. It doesn't mean we require more work."
Pregnant and Breastfeeding Women Might Have a New Reason to Ditch Artificial Sweeteners
Women considering pregnancy might have another reason to drop artificial sweeteners from their diet, if a new study of mice proves to apply to humans as well. It highlights "yet another potential health impact of zero-calorie sweeteners," according to lead author Stephanie Olivier-Van Stichelen.
The discovery was serendipitous, not part of the original study.
It found that commonly used artificial sweeteners consumed by female mice transfer to pups in the womb and later through milk, harming their development. The sweeteners affected the composition of bacteria in the gut of the pups, making them more vulnerable to developing diabetes, and greatly reduced the liver's capacity to neutralize toxins.
The discovery was serendipitous, not part of the original study, says John Hanover, the senior author and a cell biologist at the NIH National Institute of Diabetes and Digestive and Kidney Diseases. The main study looked at how a high sugar diet in the mother turns genes on and off in the developing offspring.
It compared them with mothers fed a low sugar diet, replacing sugar with a mix of sucralose and acesulfame-K (AK), two non-nutrient artificial sugars that are already used extensively in our food products and thought to be safe.
While the artificial sweeteners had little effect on the mothers, the trace amounts that were transferred through the placenta and milk had a profound effect on the pups. Hanover believes the molecules are changing gene expression during a crucial, short period of development.
"Somewhat to our surprise, we saw in the pups a really dramatic change in the microbiome" of those whose mothers were fed the artificial sweeteners, Hanover told leapsmag. "It looked like the neonates were much, much more sensitive than their mothers to the sucralose and AK." The unexpected discovery led them to publish a separate paper.
"The protective microbe Akkermansia was largely missing, and we saw a pretty dramatic shift in the ratio of two bacteria that are normally associated with metabolic disease," a precursor to diabetes, he explains. Akkermansia is a bacteria that feeds on mucus in the gut and helps remodel the tissue to an adult state over the first several months of life in a mouse. A similar process takes several years in humans, as the infant is weaned off of breast milk as the primary food source.
The good news is the body seems to remove these artificial sweeteners fairly quickly, probably within a week.
Another problem the researchers saw in the animals was "a particularly striking change in the metabolism of the detoxification systems" in the liver, says Hanover. A healthy liver is dark red, but a high dose of the artificial sweeteners turned it white, "which is a sign of massive problems."
The study was conducted in mice and Hanover cautions the findings may not apply to humans. "But in general, the microbiome changes that one sees in the rodent model mimics what we see in humans...[and] the genes that are turned on in the mouse and the human are very similar."
Hanover acknowledges the quantity of artificial sweeteners used in the study is on the high end of human consumption, roughly the equivalent of 20 cans of diet soda a day. But the sweeteners are so ubiquitous in consumer products, from foods to lipstick, and often not even mentioned on the label, that it is difficult to measure just how much a person consumes every day.
The good news is the body seems to remove these artificial sweeteners fairly quickly, probably within a week. Until further studies provide a clearer picture, women who want to err on the side of caution can choose to reduce if not eliminate their exposure to artificial sweeteners during pregnancy and breastfeeding.