A recent study in The Lancet Oncology showed that AI found 20 percent more cancers on mammogram screens 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.
Tom Patterson holding an image of Acinetobacter baumannii and Steffanie Strathdee holding an image of a bacteriophage.
An attacking rogue hippo, giant jumping spiders, even a coup in Timbuktu couldn't knock out Tom Patterson, but now he was losing the fight against a microscopic bacteria.
Death seemed inevitable, perhaps hours away, despite heroic efforts to keep him alive.
It was the deadly drug-resistant superbug Acinetobacter baumannii. The infection struck during a holiday trip with his wife to the pyramids in Egypt and had sent his body into toxic shock. His health was deteriorating so rapidly that his insurance company paid to medevac him first to Germany, then home to San Diego.
Weeks passed as he lay in a coma, shedding more than a hundred pounds. Several major organs were on the precipice of collapse, and death seemed inevitable, perhaps hours away despite heroic efforts by a major research university hospital to keep Tom alive.
Tom Patterson in a deep coma on March 14, 2016, the day before phage therapy was initiated.
(Courtesy Steffanie Strathdee)
Then doctors tried something boldly experimental -- injecting him with a cocktail of bacteriophages, tiny viruses that might infect and kill the bacteria ravaging his body.
It worked. Days later Tom's eyes fluttered open for a few brief seconds, signaling that the corner had been turned. Recovery would take more weeks in the hospital and about a year of rehabilitation before life began to resemble anything near normal.
In her new book The Perfect Predator, Tom's wife, Steffanie Strathdee, recounts the personal and scientific ordeal from twin perspectives as not only his spouse but also as a research epidemiologist who has traveled the world to track down diseases.
Part of the reason why Steff wrote the book is that both she and Tom suffered severe PTSD after his illness. She says they also felt it was "part of our mission, to ensure that phage therapy wasn't going to be forgotten for another hundred years."
Tom Patterson and Steffanie Strathdee taking a first breath of fresh air during recovery outside the UCSD hospital.
(Courtesy Steffanie Strathdee)
From Prehistoric Arms Race to Medical Marvel
Bacteriophages, or phages for short, evolved as part of the natural ecosystem. They are viruses that infect bacteria, hijacking their host's cellular mechanisms to reproduce themselves, and in the process destroying the bacteria. The entire cycle plays out in about 20-60 minutes, explains Ben Chan, a phage research scientist at Yale University.
They were first used to treat bacterial infections a century ago. But the development of antibiotics soon eclipsed their use as medicine and a combination of scientific, economic, and political factors relegated them to a dusty corner of science. The emergence of multidrug-resistant bacteria has highlighted the limitations of antibiotics and prompted a search for new approaches, including a revived interest in phages.
Most phages are very picky, seeking out not just a specific type of bacteria, but often a specific strain within a family of bacteria. They also prefer to infect healthy replicating bacteria, not those that are at rest. That's what makes them so intriguing to tap as potential therapy.
Tom's case was one of the first times that phages were successfully infused into the bloodstream of a human.
Phages and bacteria evolved measures and countermeasures to each other in an "arms race" that began near the dawn of life on the planet. It is not that one consciously tries to thwart the other, says Chan, it's that countless variations of each exists in the world and when a phage gains the upper hand and kills off susceptible bacteria, it opens up a space in the ecosystem for similar bacteria that are not vulnerable to the phage to increase in numbers. Then a new phage variant comes along and the cycle repeats.
Robert "Chip" Schooley is head of infectious diseases at the University of California San Diego (UCSD) School of Medicine and a leading expert on treating HIV. He had no background with phages but when Steff, a friend and colleague, approached him in desperation about using them with Tom, he sprang into action to learn all he could, and to create a network of experts who might provide phages capable of killing Acinetobacter.
"There is very little evidence that phage[s] are dangerous," Chip concluded after first reviewing the literature and now after a few years of experience using them. He compares broad-spectrum antibiotics to using a bazooka, where every time you use them, less and less of the "good" bacteria in the body are left. "With a phage cocktail what you're really doing is more of a laser."
Collaborating labs were able to identify two sets of phage cocktails that were sensitive to Tom's particular bacterial infection. And the FDA acted with lightning speed to authorize the experimental treatment.
A bag of a four-phage "cocktail" before being infused into Tom Patterson.
(Courtesy Steffanie Strathdee)
Tom's case was scientifically important because it was one of the first times that phages were successfully infused into the bloodstream of a human. Most prior use of phages involved swallowing them or placing them directly on the area of infection.
The success has since sparked a renewed interest in phages and a reexamination of their possible role in medicine.
Over the two years since Tom awoke from his coma, several other people around the world have been successfully treated with phages as part of their regimen, after antibiotics have failed.
The Future of Phage Therapy
The experience treating Tom prompted UCSD to create the Center for Innovative Phage Applications and Therapeutics (IPATH), with Chip and Steff as co-directors. Previous labs have engaged in basic research on phages, but this is the first clinical center in North America to focus on translating that knowledge into treating patients.
In January, IPATH announced the first phase 2 clinical trial approved by the FDA that will use phages intravenously. The viruses are being developed by AmpliPhi Biosciences, a San Diego-based company that supplied one of the phages used to treat Tom. The new study takes on drug resistant Staph aureus bacteria. Experimental phage therapy treatment using the company's product candidates was recently completed in 21 patients at seven hospitals who had been suffering from serious infections that did not respond to antibiotics. The reported success rate was 84 percent.
The new era of phage research is applying cutting-edge biologic and informatics tools to better understand and reshape the viruses to better attack bacteria, evade resistance, and perhaps broaden their reach a bit within a bacterial family.
Genetic engineering tools are being used to enhance the phages' ability to infect targeted bacteria.
"As we learn more and more about which biological activities are critical and in which clinical settings, there are going to be ways to optimize these activities," says Chip. Sometimes phages may be used alone, other times in combination with antibiotics.
Genetic engineering using tools are being used to enhance the phages' ability to infect targeted bacteria and better counter evolving forms of bacterial resistance in the ongoing "arms race" between the two. It isn't just theory. A patient recently was successfully treated with a genetically modified phage as part of the regimen, and the paper is in press.
In reality, given the trillions of phages in the world and the endless encounters they have had with bacteria over the millennia, it is likely that the exact phages needed to kill off certain bacteria already exist in nature. Using CRISPR to modify a phage is simply a quick way to identify the right phage useful for a given patient and produce it in the necessary quantities, rather than go search for the proverbial phage needle in a sewage haystack, says Chan.
One non-medical reason why using modified phages could be significant is that it creates an intellectual property stake, something that is patentable with a period of exclusive use. Major pharmaceutical companies and venture capitalists have been hesitant to invest in organisms found in nature; but a patentable modification may be enough to draw their interest to phage development and provide the funding for large-scale clinical trials necessary for FDA approval and broader use.
"There are 10 million trillion trillion phages on the planet, 10 to the power of 31. And the fact is that this ongoing evolutionary arms race between bacteria and phage, they've been at it for a millennia," says Steff. "We just need to exploit it."
