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.
Podcast: Trusting Science with Dr. Sudip Parikh, CEO of AAAS
The "Making Sense of Science" podcast features interviews with leading experts about health innovations and the big ethical and social questions they raise. The podcast is hosted by Matt Fuchs, editor of the award-winning science outlet Leaps.org.
As Pew research showed last month, many Americans have less confidence in science these days - our collective trust has declined to levels below when the pandemic began. But leaders like Dr. Sudip Parikh are taking important steps to more fully engage people in scientific progress, including breakthroughs that could benefit health and prevent disease. In January 2020, Sudip became the 19th Chief Executive Officer of the American Association for the Advancement of Science (AAAS), an international nonprofit that seeks to advance science, engineering and innovation throughout the world, with 120,000 members in 91 countries. He is the executive publisher of Science, one of the top academic journals in the world, and the Science family of journals.
Listen to the episode
Listen on Apple | Listen on Spotify | Listen on Stitcher | Listen on Amazon | Listen on Google
In this episode, Sudip and I talk about:
- Reasons to be excited about health innovations that could come to fruition in the next several years.
- Sudip's thoughts about areas of health innovation where we should be especially cautious.
- Strategies for scientists and journalists to instill greater trust in science.
- How to tap into and nurture kids' passion for STEM subjects.
- The best roles for experts to play in society and the challenges they face.
And we pack several other fascinating topics into our 35 minutes. Here are links to check out and learn more about Sudip Parikh and AAAS:
- Sudip Parikh's official bio - https://www.aaas.org/person/sudip-parikh
- Sudip Parikh, Why We Must Rebuild Trust in Science, Trend Magazine, Feb. 9, 2021 - https://www.pewtrusts.org/en/trend/archive/winter-...
- Follow Sudip on Twitter - https://twitter.com/sudipsparikh
- AAAS website - https://www.aaas.org/
- AAAS podcast - https://www.science.org/podcasts
- The latest issue of Science - https://www.science.org/
- Science Journals homepage - https://www.science.org/journals
- AAAS Mentor Resources - https://www.aaas.org/stemmentoring
- AAAS Science Journalism Awards - https://sjawards.aaas.org/enter
- Pew Research Center Report, Americans' Trust in Scientists, Other Groups Declines, Feb. 15, 2022 https://www.pewresearch.org/science/2022/02/15/ame...
For millions of people with macular degeneration, treatment options are slim. The disease causes loss of central vision, which allows us to see straight ahead, and is highly dependent on age, with people over 75 at approximately 30% risk of developing the disorder. The BrightFocus Foundation estimates 11 million people in the U.S. currently have one of three forms of the disease.
Recently, ophthalmologists including Daniel Palanker at Stanford University published research showing advances in the PRIMA retinal implant, which could help people with advanced, age-related macular degeneration regain some of their sight. In a feasibility study, five patients had a pixelated chip implanted behind the retina, and three were able to see using their remaining peripheral vision and—thanks to the implant—their partially restored central vision at the same time.
Should people with macular degeneration be excited about these results?
“Every week, if not every day, patients come to me with this question because it's devastating when they lose their central vision,” says retinal surgeon Lynn Huang. About 40% of her patients have macular degeneration. Huang tells them that these implants, along with new medications and stem cell therapies, could be useful in the coming years.
“The goal here is to replace the missing photoreceptors with photovoltaic pixels, basically like little solar panels,” Palanker says.
That implant, a pixelated chip, works together with a tiny video camera on a specially designed pair of eyeglasses, which can be adjusted for each patient’s prescription. The video camera relays processed images to the chip, which electrically stimulates inner retinal neurons. These neurons, in turn, relay information to the brain’s visual cortex through the optic nerve. The chip restores patients’ central sight, but not completely. The artificial vision is basically monochromatic (whitish-yellowish) and fairly blurry; patients were still legally blind even after the implant, except when using a zoom function on the camera, but those with proper chip placement could make out large letters.
“The goal here is to replace the missing photoreceptors with photovoltaic pixels, basically like little solar panels,” Palanker says. These pixels, located on the implanted chip, convert light into pulsed electrical currents that stimulate retinal neurons. In time, Palanker hopes to improve the chips, resulting in bigger boosts to visual acuity.
The pixelated chips are surgically implanted during a process Palanker admits is still “a surgical learning curve.” In the study, three chips were implanted correctly, one was placed incorrectly, and another patient’s chip moved after the procedure; he did not follow post-surgical recommendations. One patient passed away during the study for unrelated reasons.
University of Maryland retinal specialist Kenneth Taubenslag, who was not involved in the study, said that subretinal surgeries have become less common in recent years, but expects implants to spur improvements in these techniques. “I think as people get more experience, [they’ll] probably get more reliable placement of the implant,” he said, pointing out that even the patient with the misplaced chip was able to gain some light perception, if not the same visual acuity as other patients.
Retinal implants have come under scrutiny lately. IEEE Spectrum reported that Second Sight, manufacturer of the Argus II implant used for people with retinitis pigmentosa, a genetic disease that causes vision loss, would no longer support the product. After selling hundreds of the implants at $150,000 apiece, company leaders announced they’d “decided to pursue an orderly wind down” of Second Sight in March 2020 in the wake of financial issues. Last month, the company announced a merger, shifting its focus to a new retinal implant, raising questions for patients who have Argus II implants.
Retinal surgeon Eugene de Juan of the University of California, San Francisco, was involved with early studies of the Argus implants, though his participation ended over a decade ago, before the device was marketed by Second Sight. He says he would consider recommending future implants to patients with macular degeneration, given the promise of the technology and the lack of other alternatives.
“I tell my patients that this is an area of active research and development, and it's getting better and better, so let's not give up hope,” de Juan says. He believes cautious optimism for Palanker’s implant is appropriate: “It's not the first, it's not the only, but it's a good approach with a good team.”