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.
The Nose Knows: Dogs Are Being Trained to Detect the Coronavirus
Asher is eccentric and inquisitive. He loves an audience, likes keeping busy, and howls to be let through doors. He is a six-year-old working Cocker Spaniel, who, with five other furry colleagues, has now been trained to sniff body odor samples from humans to detect COVID-19 infections.
As the Delta variant and other new versions of the SARS-CoV-2 virus emerge, public health agencies are once again recommending masking while employers contemplate mandatory vaccination. While PCR tests remain the "gold standard" of COVID-19 tests, they can take hours to flag infections. To accelerate the process, scientists are turning to a new testing tool: sniffer dogs.
At the London School of Hygiene and Tropical Medicine (LSHTM), researchers deployed Asher and five other trained dogs to test sock samples from 200 asymptomatic, infected individuals and 200 healthy individuals. In May, they published the findings of the yearlong study in a preprint, concluding that dogs could identify COVID-19 infections with a high degree of accuracy – they could correctly identify a COVID-positive sample up to 94% of the time and a negative sample up to 92% of the time. The paper has yet to be peer-reviewed.
"Dogs can screen lots of people very quickly – 300 people per dog per hour. This means they could be used in places like airports or public venues like stadiums and maybe even workplaces," says James Logan, who heads the Department of Disease Control at LSHTM, adding that canines can also detect variants of SARS-CoV-2. "We included samples from two variants and the dogs could still detect them."
Detection dogs have been one of the most reliable biosensors for identifying the odor of human disease. According to Gemma Butlin, a spokesperson of Medical Detection Dogs, the UK-based charity that trained canines for the LSHTM study, the olfactory capabilities of dogs have been deployed to detect malaria, Parkinson's disease, different types of cancers, as well as pseudomonas, a type of bacteria known to cause infections in blood, lungs, eyes, and other parts of the human body.
COVID-19 has a distinctive smell — a result of chemicals known as volatile organic compounds released by infected body cells, which give off an odor "fingerprint."
"It's estimated that the percentage of a dog's brain devoted to analyzing odors is 40 times larger than that of a human," says Butlin. "Humans have around 5 million scent receptors dedicated to smell. Dogs have 350 million and can detect odors at parts per trillion. To put this into context, a dog can detect a teaspoon of sugar in a million gallons of water: two Olympic-sized pools full."
According to LSHTM scientists, COVID-19 has a distinctive smell — a result of chemicals known as volatile organic compounds released by infected body cells, which give off an odor "fingerprint." Other studies, too, have revealed that the SARS-CoV-2 virus has a distinct olfactory signature, detectable in the urine, saliva, and sweat of infected individuals. Humans can't smell the disease in these fluids, but dogs can.
"Our research shows that the smell associated with COVID-19 is at least partly due to small and volatile chemicals that are produced by the virus growing in the body or the immune response to the virus or both," said Steve Lindsay, a public health entomologist at Durham University, whose team collaborated with LSHTM for the study. He added, "There is also a further possibility that dogs can actually smell the virus, which is incredible given how small viruses are."
In April this year, researchers from the University of Pennsylvania and collaborators published a similar study in the scientific journal PLOS One, revealing that detection dogs could successfully discriminate between urine samples of infected and uninfected individuals. The accuracy rate of canines in this study was 96%. Similarly, last December, French scientists found that dogs were 76-100% effective at identifying individuals with COVID-19 when presented with sweat samples.
Grandjean Dominique, a professor at France's National Veterinary School of Alfort, who led the French study, said that the researchers used two types of dogs — search and rescue dogs, as they can sniff sweat, and explosive detection dogs, because they're often used at airports to find bomb ingredients. Dogs may very well be as good as PCR tests, said Dominique, but the goal, he added, is not to replace these tests with canines.
In France, the government gave the green light to train hundreds of disease detection dogs and deploy them in airports. "They will act as mass pre-test, and only people who are positive will undergo a PCR test to check their level of infection and the kind of variant," says Dominique. He thinks the dogs will be able to decrease the amount of PCR testing and potentially save money.
Since the accuracy rate for bio-detection dogs is fairly high, scientists think they could prove to be a quick diagnosis and mass screening tool, especially at ports, airports, train stations, stadiums, and public gatherings. Countries like Finland, Thailand, UAE, Italy, Chile, India, Australia, Pakistan, Saudi Arabia, Switzerland, and Mexico are already training and deploying canines for COVID-19 detection. The dogs are trained to sniff the area around a person, and if they find the odor of COVID-19 they will sit or stand back from an individual as a signal that they've identified an infection.
While bio-detection dogs seem promising for cheap, large-volume screening, many of the studies that have been performed to date have been small and in controlled environments. The big question is whether this approach work on people in crowded airports, not just samples of shirts and socks in a lab.
"The next step is 'real world' testing where they [canines] are placed in airports to screen people and see how they perform," says Anna Durbin, professor of international health at the John Hopkins Bloomberg School of Public Health. "Testing in real airports with lots of passengers and competing scents will need to be done."
According to Butlin of Medical Detection Dogs, scalability could be a challenge. However, scientists don't intend to have a dog in every waiting room, detecting COVID-19 or other diseases, she said.
"Dogs are the most reliable bio sensors on the planet and they have proven time and time again that they can detect diseases as accurately, if not more so, than current technological diagnostics," said Butlin. "We are learning from them all the time and what their noses know will one day enable the creation an 'E-nose' that does the same job – imagine a day when your mobile phone can tell you that you are unwell."
The Voice Behind Some of Your Favorite Cartoon Characters Helped Create the Artificial Heart
In June, a team of surgeons at Duke University Hospital implanted the latest model of an artificial heart in a 39-year-old man with severe heart failure, a condition in which the heart doesn't pump properly. The man's mechanical heart, made by French company Carmat, is a new generation artificial heart and the first of its kind to be transplanted in the United States. It connects to a portable external power supply and is designed to keep the patient alive until a replacement organ becomes available.
Many patients die while waiting for a heart transplant, but artificial hearts can bridge the gap. Though not a permanent solution for heart failure, artificial hearts have saved countless lives since their first implantation in 1982.
What might surprise you is that the origin of the artificial heart dates back decades before, when an inventive television actor teamed up with a famous doctor to design and patent the first such device.
A man of many talents
Paul Winchell was an entertainer in the 1950s and 60s, rising to fame as a ventriloquist and guest-starring as an actor on programs like "The Ed Sullivan Show" and "Perry Mason." When children's animation boomed in the 1960s, Winchell made a name for himself as a voice actor on shows like "The Smurfs," "Winnie the Pooh," and "The Jetsons." He eventually became famous for originating the voices of Tigger from "Winnie the Pooh" and Gargamel from "The Smurfs," among many others.
But Winchell wasn't just an entertainer: He also had a quiet passion for science and medicine. Between television gigs, Winchell busied himself working as a medical hypnotist and acupuncturist, treating the same Hollywood stars he performed alongside. When he wasn't doing that, Winchell threw himself into engineering and design, building not only the ventriloquism dummies he used on his television appearances but a host of products he'd dreamed up himself. Winchell spent hours tinkering with his own inventions, such as a set of battery-powered gloves and something called a "flameless lighter." Over the course of his life, Winchell designed and patented more than 30 of these products – mostly novelties, but also serious medical devices, such as a portable blood plasma defroster.
Ventriloquist Paul Winchell with Jerry Mahoney, his dummy, in 1951 |
A meeting of the minds
In the early 1950s, Winchell appeared on a variety show called the "Arthur Murray Dance Party" and faced off in a dance competition with the legendary Ricardo Montalban (Winchell won). At a cast party for the show later that same night, Winchell met Dr. Henry Heimlich – the same doctor who would later become famous for inventing the Heimlich maneuver, who was married to Murray's daughter. The two hit it off immediately, bonding over their shared interest in medicine. Before long, Heimlich invited Winchell to come observe him in the operating room at the hospital where he worked. Winchell jumped at the opportunity, and not long after he became a frequent guest in Heimlich's surgical theatre, fascinated by the mechanics of the human body.
One day while Winchell was observing at the hospital, he witnessed a patient die on the operating table after undergoing open-heart surgery. He was suddenly struck with an idea: If there was some way doctors could keep blood pumping temporarily throughout the body during surgery, patients who underwent risky operations like open-heart surgery might have a better chance of survival. Winchell rushed to Heimlich with the idea – and Heimlich agreed to advise Winchell and look over any design drafts he came up with. So Winchell went to work.
Winchell's heart
As it turned out, building ventriloquism dummies wasn't that different from building an artificial heart, Winchell noted later in his autobiography – the shifting valves and chambers of the mechanical heart were similar to the moving eyes and opening mouths of his puppets. After each design, Winchell would go back to Heimlich and the two would confer, making adjustments along the way to.
By 1956, Winchell had perfected his design: The "heart" consisted of a bag that could be placed inside the human body, connected to a battery-powered motor outside of the body. The motor enabled the bag to pump blood throughout the body, similar to a real human heart. Winchell received a patent for the design in 1963.
At the time, Winchell never quite got the credit he deserved. Years later, researchers at the University of Utah, working on their own artificial heart, came across Winchell's patent and got in touch with Winchell to compare notes. Winchell ended up donating his patent to the team, which included Dr. Richard Jarvik. Jarvik expanded on Winchell's design and created the Jarvik-7 – the world's first artificial heart to be successfully implanted in a human being in 1982.
The Jarvik-7 has since been replaced with newer, more efficient models made up of different synthetic materials, allowing patients to live for longer stretches without the heart clogging or breaking down. With each new generation of hearts, heart failure patients have been able to live relatively normal lives for longer periods of time and with fewer complications than before – and it never would have been possible without the unsung genius of a puppeteer and his love of science.