Artificial Intelligence is getting better than humans at detecting breast cancer
A recent study in The Lancet Oncology showed that AI found 20 percent more cancers on mammogram screens than radiologists alone.
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.
Prostate Cancer Treatments Are Racing Ahead. Here’s a Big Reason Why.
Dr. Stacy Loeb (left) and Dr. Heather Cheng were both recipients of a special early-career grant from the Prostate Cancer Foundation for young researchers.
In his lab at UCLA, Dr. Charles Sawyer discovered two drugs for metastatic prostate cancer that are now in routine use all over the world.
At the University of Washington at Seattle, Dr. Heather Cheng was part of a team that discovered the connection between BRCA2 mutations and advanced prostate cancer, and she recently opened a prostate cancer genetics clinic – a new frontier in the field.
At UT Southwestern Medical Center in Dallas, Dr. Nima Sharifi's pioneering research showed why certain drugs don't work in castrate-resistant prostate cancer, and now new therapies are being developed instead.
"We have good reason to believe that investing in young scientists is the way to go."
What Do These Researchers Share in Common?
They were all under 40 when they received a special grant for early-career scientists from the Prostate Cancer Foundation, the leading philanthropic organization that funds prostate cancer research. Experts say that the foundation's dedicated support for young innovators has been a game changer in contributing to the discovery of newer and better therapies for prostate cancer patients.
Howard Soule, the foundation's Executive Vice President and Chief Science Officer, was aware that many of the people who leave behind major legacies in science typically make their discoveries before age 40, like Albert Einstein, who was in his thirties when he published his paper on general relativity.
So back in 2007, the PCF decided to ramp up its support for young researchers.
"We have good reason to believe that investing in young scientists is the way to go, so we've created a program at PCF that is I believe is unique in the field," says Soule.
The Young Investigator Awards Program rigorously screens a pool of roughly 150 applicants for 20 to 25 awards that consist of funding for three years – and that's just the start.
"It's much more than sending them money," says Soule. "We celebrate them at annual meetings, we have a networking center with no equal in the field, and throughout the years of their three-year-award and basically forever, we create community. We are a safe place for them to land, they share data with us that's unpublished, and we provide a lot of feedback and stewardship to their donors."
Spotlighting Recipients: From the Study of Tumors to Twitter
Heather Cheng was in her thirties when she received her award three years ago. "It's been very, very helpful in allowing me to do the type of work I am really excited about doing," she says.
At the time, she had recently joined the faculty at the University of Washington after completing an MD/PhD medical scientist training program, internal medicine residency and hematology/oncology fellowship, and she was considering what new direction to take in her research. Several patients captured her imagination who were diagnosed at a very young age with metastatic prostate cancer, and "even though we had cool new drugs to extend life, these particular patients' cancers blew through everything."
"This is a new intersection because genetics has not been discussed in the context of care for men with prostate cancer that much."
She decided to make a niche out of understanding the connection between often early-onset aggressive prostate cancer and familial genetic risk, in order to improve treatment options for these patients. In 2016, Cheng launched a new clinic and invited any men to visit who have a family history of cancer and who are interested in genetic testing, or who have a known mutation and want to learn about treatment opportunities, or who want to know if their cancer tumor can be inherited.
"It's an open door to have a discussion because the technology and treatment potentials are so new," Cheng says. "There's a lot to learn."
It used to be that a doctor would ask a male patient about his family history, and if a mother had breast cancer at a young age, for example, and several other family members met the criteria for a genetic risk, then perhaps the patient had inherited a mutation in a cancer risk gene. But what to do next was unclear.
Now, doctors are taking men with a diagnosis of prostate cancer, sequencing their inherited DNA or their tumors, and finding out if they have mutations that could guide their treatment plan. For example, medications called PARP inhibitors have shown encouraging early results for men with a BRCA2 gene mutation and are now in clinical trials for treating prostate cancer.
"This is a new intersection because genetics has not been discussed in the context of care for men with prostate cancer that much," Cheng says. "This has changed practice because changes to national cancer guidelines have happened in less than five years. The change has happened so quickly that the field is not completely prepared for implementation and clinical logistics."
Another young investigator, New York University urologist and prostate cancer researcher Stacy Loeb, received her award at age 36 two years ago. She realized that no one had scientifically studied how patients are using crowdsourcing platforms like GoFundMe and YouCaring to raise money for their treatments. In her research, she found that there are many more campaigns for breast cancer and that they are more successful in crowdfunding than the prostate cancer campaigns.
"We have identified some gaps in advocacy and awareness for prostate cancer – fewer people know about it or discuss it, but it is a leading cause of death of U.S. men, so it is important to get more people aware," Loeb notes.
In fact, today the PCF releases data from a survey of more than 2,000 U.S. adults that reveals widespread ignorance about the disease. Two-thirds of respondents, for example, did not know that men with early stage prostate cancer experience no symptoms, and many were unaware that screening begins with a simple blood test.
Besides studying patient behavior, Loeb also wanted to better understand how physicians and scientists are using social media, and how their participation on platforms like Twitter could be fostered to promote greater dissemination of knowledge. So she helped start a monthly prostate cancer journal club on Twitter, hosted through the PCF science account. The club features an important new research paper in the field each month, and she invites the authors of the paper to participate in a 48-hour online discussion.
"The Journal Club is a monthly thing at most institutions," she says, "but typically it's one institution with people from one department. What's better about this is we have people who are doctors, nurses, scientists, patients, stakeholders participating from all over the world."
Why Do Young Innovators Have an Edge?
The environment matters, for one.
"We all bring different life experiences to the table, we grew up in different eras, so we have different norms and tools at our disposal that weren't available," says Loeb, who was one of the early adopters of social media in the urology space. She now gives a lecture at the annual PCF retreat on how to use social media to advance one's scientific career.
"The more you're invested into a system, the less you may be able to recognize its limitations."
But the advantage of youth is not just greater familiarity with the newest tools. It's also the existential benefit of not being entrenched in the way it's always been.
"Often there is a healthy skepticism of what's come before," explains Dr. Joseph La Brie, a clinical psychologist and professor at psychology at Loyola Marymount University. "That's connected to not being wedded to a programmatic view of the problem. There's a freshness and creative outlook because they are looking at it with a new set of eyes, and there's a desire to make their mark on the field, to be unique and innovative and not just follow in somebody else's footsteps."
And as Cheng puts it, "The more you're invested into a system, the less you may be able to recognize its limitations."
But it's notoriously difficult for scientists to get funding for innovative ideas without having already published preliminary data, which is what the National Institutes of Health and other funding bodies like to see. Eliminating that hurdle is a big part of why PCF's approach has been so effective, according to a veteran of the field, Johns Hopkins urologist Dr. Kenneth Pienta; his own groundbreaking research has been supported by PCF since he was a young scientist in the '90s.
"Any granting mechanism that allows people to fund ideas without a lot of preliminary data is a good thing," he declares.
Support for creative young minds is crucial across all endeavors, beyond any single disease or discipline. At a recent conference showcasing emerging technology for DARPA, the research arm of the Defense Department, expert panelists in artificial intelligence were asked: What is the single most important thing to focus on over the next decade?
Robotics pioneer Dr. Rodney Brooks may have put it best: "Figure out how to fund some really radical young mavericks and see what happens."
Kira Peikoff was the editor-in-chief of Leaps.org from 2017 to 2021. As a journalist, her work has appeared in The New York Times, Newsweek, Nautilus, Popular Mechanics, The New York Academy of Sciences, and other outlets. She is also the author of four suspense novels that explore controversial issues arising from scientific innovation: Living Proof, No Time to Die, Die Again Tomorrow, and Mother Knows Best. Peikoff holds a B.A. in Journalism from New York University and an M.S. in Bioethics from Columbia University. She lives in New Jersey with her husband and two young sons. Follow her on Twitter @KiraPeikoff.
Your Genetic Data Is The New Oil. These Startups Will Pay to Rent It.
A doctor conducts a test on DNA.
Perhaps you're one of the 12 million people who has spit into a tube in recent years to learn the secrets in your genetic code, like your ancestry, your health risks, or your carrier status for certain diseases. If you haven't participated in direct-to-consumer genetic testing, you may know someone who has.
It's for people who want more control over their genetic data--plus a share of the proceeds when and if that data is used.
Mountains of genomic data have been piling up steeply over the last several years, but according to some experts, not enough research and drug discovery is being done with the data collected, and customers rarely have a say in how their data is used. Now, a slew of ambitious startup companies are bringing together the best of blockchain technology and human genomics to help solve these problems.
But First, Why Is Your Genome So Valuable?
Access to genetic information is an obvious boon to scientific and medical progress. In the right hands, it has the potential to save lives and reduce suffering — by facilitating the development of better, safer, more targeted treatments and by shedding light on the role of genetics in countless diseases and medical conditions.
Research requiring access to direct-to-consumer (DTC) genomic data is already well underway. For example, 23andMe, the popular California-based DTC genetic testing company, has published 107 research articles so far, as of this May, using data from their five million-plus customers around the world. Their website states that, on average, of the 80 percent of their customers who have opted to share their genomic data for research purposes, each "individual contributes to 200 different research studies."
And this July, a new collaboration was announced between 23andMe and GlaxoSmithKline, the London-based pharmaceutical company. GlaxoSmithKline will be using data from 23andMe customers to develop new medical treatments, while 23andMe will receive $300 million from the four-year deal. Both companies are poised to profit significantly from their union.
Should 23andMe's customers share in the gains? Peter Pitts, president of the Center for Medicine in the Public Interest, believes they should. "Are they going to offer rebates to people who opt in, so their customers aren't paying for the privilege of 23andMe working with a for-profit company in a for-profit research project?" Pitts told NBC. So far, 23andMe has not announced any plans to share profits with their customers.
But outside of such major partnerships, many researchers are frustrated by the missed opportunities to dig deeper into the correlations between genetics and disease. That's because people's de-identified genomic information is "essentially lying fallow," siloed behind significant security blockades in the interest of preserving their anonymity. So how can both researchers and consumers come out ahead?
Putting Consumers Back in Control
For people who want more control over their genetic data -- plus a share of the proceeds when and if that data is used -- a few companies have paired consumer genomics with blockchain technology to form a new field called "blockchain genomics." Blockchain is a data storage technology that relies on a network of computers, or peer-to-peer setup, making it incredibly difficult to hack. "It's a closed loop of transactions that gets protected and encrypted, and it cannot be changed," says Tanya Woods, a blockchain thought leader and founder of Kind Village, a social impact technology platform.
The vision is to incentivize consumers to share their genomic data and empower researchers to make new breakthroughs.
"So if I agree to give you something and you agree to accept it, we make that exchange, and then that basic framework is captured in a block. … Anything that can be exchanged can be ledgered on blockchain. Anything. It could be real estate, it could be the transfer of artwork, it could be the purchase of a song or any digital content, it could be recognition of a certification," and so on.
The blockchain genomics companies' vision is to incentivize consumers to share their genomic data and empower researchers to make new breakthroughs, all while keeping the data secure and the identities of consumers anonymous.
Consumers, or "partners" as these companies call them, will have a direct say regarding which individuals or organizations can "rent" their data, and will be able to negotiate the amount they receive in exchange. But instead of fiat currency (aka "regular money") as payment, partners will either be remunerated in cryptocurrency unique to the specific company or they will be provided with individual shares of ownership in the database for contributing DNA data and other medical information.
Luna DNA, one of the blockchain genomics companies, "will allow any credible researcher or non-profit to access the databases for a nominal fee," says its president and co-founder, Dawn Barry. Luna DNA's infrastructure was designed to embrace certain conceptions of privacy and privacy law "in which individuals are in total control of their data, including the ability to have their data be 'forgotten' at any time," she said. This is nearly impossible to implement in pre-existing systems that were not designed with full control by the individual in mind.
One of the legal instruments to which Barry referred was the European Union's General Data Protection Regulation, which "states that the data collected on an individual is owned and should be controlled by that individual," she explained. Another is the California Privacy Act that echoes similar principles. "There is a global trend towards more control by the individual that has very deep implications to companies and sites that collect and aggregate data."
David Koepsell, CEO and co-founder of EncrypGen, told Forbes that "Most people are not aware that your DNA contains information about your life expectancy, your proclivity to depression or schizophrenia, your complete ethnic ancestry, your expected intelligence, maybe even your political inclinations" — information that could be misused by insurance companies and employers. And though DTC customers have been assured that their data will stay anonymous, some data can be linked back to consumers' identities. Blockchain may be the answer to these concerns.
Both blockchain technology and the DTC genetic testing arena have a glaring diversity problem.
"The security that's provided by blockchain is tremendous," Woods says. "It's a significant improvement … and as we move toward more digitized economies around the world, these kinds of solutions that are providing security, validity, trust — they're very important."
In the case of blockchain genomics companies like EncrypGen, Luna DNA, Longenesis, and Zenome, each partner who joins would bring a digital copy of their genetic readout from DTC testing companies (like 23andMe or AncestryDNA). The blockchain technology would then be used to record how and for what purposes researchers interact with it. (To learn more about blockchain, check out this helpful visual guide by Reuters.)
Obstacles in the Path to Success
The cryptocurrency approach as a method of payment could be an unattractive lure to consumers if only a limited number of people make transactions in a given currency's network. And the decade-old technology underlying it -- blockchain -- is not yet widely supported, or even well-understood, by the public at large.
"People conflate blockchain with cryptocurrency and bitcoin and all of the concerns and uncertainty thereof," Barry told us. "One can think of cryptocurrency as a single expression of the vast possibilities of the blockchain technology. Blockchain is straightforward in concept and arcane in its implementation."
But blockchain, with its Gini coefficient of 0.98, is one of the most unequal "playing fields" around. The Gini coefficient is a measure of economic inequality, where 0 represents perfect equality and 1 represents perfect inequality. Around 90 percent of bitcoin users, for example, are male, white or Asian, between the ages of 18 and 34, straight, and from middle and upper class families.
The DTC genetic testing arena, too, has a glaring diversity problem. Most DTC genetic test consumers, just like most genetic study participants, are of European descent. In the case of genetic studies, this disparity is largely explained by the fact that most research is done in Europe and North America. In addition to being over 85 percent white, individuals who purchase DTC genetic testing kits are highly educated (about half have more than a college degree), well off (43 percent have a household income of $100,000 or more per year), and are politically liberal (almost 65 percent). Only 14.5 percent of DTC genetic test consumers are non-white, and a mere 5 percent are Hispanic.
Since risk of genetic diseases often varies greatly between ethnic groups, results from DTC tests can be less accurate and less specific for those of non-European ancestry — simply due to a lack of diverse data. The bigger the genetic database, wrote Sarah Zhang for The Atlantic, the more insights 23andMe and other DTC companies "can glean from DNA. That, in turn, means the more [they] can tell customers about their ancestry and health…" Though efforts at recruiting non-white participants have been ongoing, and some successes have been made at improving ancestry tools for people of color, the benefits of genomic gathering in North America are still largely reaped by Caucasians.
So far, it's not yet clear who or how many people will choose to partake in the offerings of blockchain genomics companies.
So one chief hurdle for the blockchain genomics companies is getting the technology into the hands of those who are under-represented in both blockchain and genetic testing research. Women, in particular, may be difficult to bring on board the blockchain genomics bandwagon — though not from lack of interest. Although women make up a significant portion of DTC genetic testing customers (between 50 and 60 percent), their presence is lacking in blockchain and the biotech industry in general.
At the North American Bitcoin Conference in Miami earlier this year, only three women were on stage, compared to 84 men. And the after-party was held in a strip club.
"I was at that conference," Woods told us. "I don't know what happened at the strip club, I didn't observe it. That's not to say it didn't happen … but I enjoyed being at the conference and I enjoyed learning from people who are experimenting in the space and developing in it. Generally, would I have loved to see more women visible? Of course. In tech generally I want to see more women visible, but there's a whole ecosystem shifting that has to happen to make that possible."
Luna's goal is to achieve equal access to a technology (blockchain genomics) that could potentially improve health and quality of life for all involved. But in the merging of two fields that have been unequal since their inception, achieving equal access is one tall order indeed. So far, it's not yet clear who or how many people will choose to participate. LunaDNA's platform has not yet launched; EncrypGen released their beta version just last month.
Sharon Terry, president and CEO of Genetic Alliance — a nonprofit organization that advocates for access to quality genetic services — recently shared a message that reflects the zeitgeist for all those entering the blockchain genomics space: "Be authentic. Tell the truth, even about motives and profits. Be transparent. Engage us. Don't leave us out. Make this real collaboration. Be bold. Take risks. People are dying. It's time to march forward and make a difference."