A recent study in The Lancet Oncology showed that AI found 20 percent more cancers on mammogram screens than radiologists alone.
The Lancet Oncology
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.
Anita Freeman, left, hugs her sister Elizabeth Martin at a bakery in Huntington Beach, California, in 2009 — a year before Elizabeth's cancer diagnosis. (Courtesy Anita Freeman)
(Courtesy Anita Freeman)
For five weeks, Anita Freeman watched her sister writhe in pain. The colon cancer diagnosed four years earlier became metastatic.
"I still wouldn't wish that ending on my worst enemy."
At this tormenting juncture, her 66-year-old sister, Elizabeth Martin, wanted to die comfortably in her sleep. But doctors wouldn't help fulfill that final wish.
"It haunts me," Freeman, 74, who lives in Long Beach, California, says in recalling the prolonged agony. Her sister "was breaking out of the house and running in her pajamas down the sidewalk, screaming, 'Help me. Help me.' She just went into a total panic."
Finally, a post-acute care center offered pentobarbital, a sedative that induced a state of unconsciousness, but only after an empathetic palliative care doctor called and insisted on ending the inhumane suffering. "We even had to fight the owners of the facility to get them to agree to the recommendations," Freeman says, describing it as "the only option we had at that time; I still wouldn't wish that ending on my worst enemy."
Her sister died a week later, in 2014. That was two years before California's medical aid-in-dying law took effect, making doctors less reliant on palliative sedation to peacefully end unbearable suffering for terminally ill patients. Now, Freeman volunteers for Compassion & Choices, a national grassroots organization based in Portland, Oregon, that advocates for expanding end-of-life options.
Palliative sedation involves medicating a terminally ill patient into lowered awareness or unconsciousness in order to relieve otherwise intractable suffering at the end of life. It is not intended to cause death, which occurs due to the patient's underlying disease.
In contrast, euthanasia involves directly and deliberately ending a patient's life. Euthanasia is legal only in Canada and some European countries and requires a health care professional to administer the medication. In the United States, laws in seven states and Washington, D.C. give terminally ill patients the option to obtain prescription medication they can take to die peacefully in their sleep, but they must be able to self-adminster it.
Recently, palliative sedation has been gaining more acceptance among medical professionals as an occasional means to relieve suffering, even if it may advance the time of death, as some clinicians believe. However, studies have found no evidence of this claim. Many doctors and bioethicists emphasize that intent is what distinguishes palliative sedation from euthanasia. Others disagree. It's common for controversy to swirl around when and how to apply this practice.
Elizabeth Martin with her sister Anita Freeman in happier times, before metastatic cancer caused her tremendous suffering at the end of her life.
(Courtesy Anita Freeman)
"Intent is everything in ethics. The rigor and protocols we have around palliative sedation therapy also speaks to it being an intervention directed to ease refractory distress," says Martha Twaddle, medical director of palliative medicine and supportive care at Northwestern University's Lake Forest Hospital in Lake Forest, Illinois.
Palliative sedation should be considered only when pain, shortness of breath, and other unbearable symptoms don't respond to conventional treatments. Left to his or her own devices, a patient in this predicament could become restless, Twaddle says, noting that "agitated delirium is a horrible symptom for a family to witness."
At other times, "we don't want to be too quick to sedate," particularly in cases of purely "existential distress"—when a patient experiences anticipatory grief around "saying goodbye" to loved ones, she explains. "We want to be sure we're applying the right therapy for the problem."
Encouraging patients to reconcile with their kin may help them find inner peace. Nonmedical interventions worth exploring include quieting the environment and adjusting lighting to simulate day and night, Twaddle says.
Music-thanatology also can have a calming effect. It is live, prescriptive music, mainly employing the harp or voice, tailored to the patient's physiological needs by tuning into vital signs such as heart rate, respiration, and temperature, according to the Music-Thanatology Association International.
"When we integrated this therapeutic modality in 2003, our need for using palliative sedation therapy dropped 75 percent and has remained low ever since," Twaddle observes. "We have this as part of our care for treating refractory symptoms."
"If palliative sedation is being employed properly with the right patient, it should not hasten death."
Ethical concerns surrounding euthanasia often revolve around the term "terminal sedation," which "can entail a physician deciding that the patient is a lost cause—incurable medically and in substantial pain that cannot adequately be relieved," says John Kilner, professor and director of the bioethics programs at Trinity International University in Deerfield, Illinois.
By halting sedation at reasonable intervals, the care team can determine whether significant untreatable pain persists. Periodic discontinuation serves as "evidence that the physician is still working to restore the patient rather than merely to usher the patient painlessly into death," Kilner explains. "Indeed, sometimes after a period of unconsciousness, with the body relieved of unceasing pain, the body can recover enough to make the pain treatable."
The medications for palliative sedation "are tried and true sedatives that we've had for a long time, for many years, so they're predictable," says Joe Rotella, chief medical officer at the American Academy of Hospice and Palliative Medicine.
Some patients prefer to keep their eyes open and remain conscious to answer by name, while others tell their doctors in advance that they want to be more heavily sedated while receiving medications to manage pain and other symptoms. "We adjust the dosage until the patient is sleeping at a desired level of sedation," Rotella says.
Sedation is an intrinsic side effect of most medications prescribed to control severe symptoms in terminally ill patients. In general, most people die in a sleepy state, except for instances of sudden, dramatic death resulting from a major heart attack or stroke, says Ryan R. Nash, a palliative medicine physician and director of The Ohio State University Center for Bioethics in Columbus.
"Using those medications to treat pain or shortness of breath is not palliative sedation," Nash says. In addition, providing supplemental nutrition and hydration in situations where death is imminent—with a prognosis limited to hours or days—generally doesn't help prolong life. "If palliative sedation is being employed properly with the right patient," he adds, "it should not hasten death."
Nonetheless, hospice nurses sometimes feel morally distressed over carrying out palliative sedation. Implementing protocols at health systems would help guide them and alleviate some of their concerns, says Gregg VandeKieft, medical director for palliative care at Providence St. Joseph Health's Southwest Washington Region in Olympia, Washington. "It creates guardrails by sort of standardizing and normalizing things," he says.
"Our goal is to restore our patient. It's never to take their life."
The concept of proportionality weighs heavily in the process of palliative sedation. But sometimes substantial doses are necessary. For instance, an opioid-tolerant patient recently needed an unusually large amount of medication to control symptoms. She was in a state of illness-induced confusion and pain, says David E. Smith, a palliative medicine physician at Baptist Health Supportive Care in Little Rock, Arkansas.
Still, "we are parsimonious in what we do. We only use as much therapeutic force as necessary to achieve our goals," Smith says. "Our goal is to restore our patient. It's never to take their life."
Steven Pinker, a cognitive psychologist at Harvard, argues that the ideals of the Enlightenment deserve a vigorous moral defense.
(Photo credit: Rose Lincoln / Harvard University)
The government shutdown. A volatile stock market. Climate change.
It's so easy to get discouraged by the latest headlines, argues Steven Pinker, that we lose sight of the bigger picture: life today is actually improving.
"To appreciate the world, we've got to look at numbers and trends."
Pinker, a cognitive psychologist from Harvard, says in his book "Enlightenment Now" that we're living at the greatest moment of progress in history, thanks to reason, science, and humanism. But today, he says, these ideals are under-appreciated, and we ignore them at our peril.
So he set out to provide a vigorous moral defense of the values of the Enlightenment by examining the evidence for their effectiveness. Across a range of categories from happiness and health to peace and safety, Pinker examines the data and reassures readers that this is a pretty great time to be alive. As we kick off the new year, he's hopeful that our embrace of science and reason will lead to an even more prosperous future. But political and cultural hurdles must still be overcome before the heroic story of human progress can continue to unfold.
Pinker spoke with our Editor-in-Chief Kira Peikoff in advance of the book's paperback release, which hits stores next Tuesday. This interview has been edited and condensed for clarity.
One anecdote you describe in the book was particularly striking: how the public reacted when the polio vaccine was announced. People took the day off work to celebrate, they smiled at each other in the streets, they offered to throw parades. Today, it's hard to imagine such prevalent enthusiasm for a new advance. How can we bring back a culture of respect and gratitude for science?
That's such a good question. And I wish I knew the answer. My contribution is just to remind people of how much progress we've made. It's easy to ignore if your view of the world comes from headlines, but there are some built-in biases in journalism that we have to counteract. Most things that happen all of a sudden are bad things: wars break out, terrorists attack, rampage shootings occur, whereas a lot of the things that make us better off creep up by stealth. But we have to become better aware of them.
It's unlikely that we're going to have replications of the great Salk event, which happened on a particular day, but I think we have to take lessons from cognitive science, from the work of people like Daniel Kahneman and Amos Tversky, showing how misled we can be by images and narratives and that to appreciate the world, we've got to look at numbers and trends.
The cover of "Enlightenment Now," which comes out in paperback next week.
You mention that the President's Bioethics Council under Bush was appointed to deal with "the looming threat of biomedical advances." Do you think that professional bioethicists are more of a hindrance than a help when it comes to creating truly enlightened science policy?
I do. I think that there are some problems in the culture of bioethics. And of course, I would not argue against that the concept of bioethics. Obviously, we have to do biomedical research and applications conscientiously and ethically. But the field called Bioethics tends to specialize in exotic thought experiments that tend to imagine the worst possible things that can happen, and often mire research in red tape that results in a net decrease in human welfare, whereas the goal of bioethics should be to enhance human welfare.
In an op-ed that I published in the Boston Globe a few years ago, I said, deliberately provocatively, that the main moral imperative of bioethics is to get out of the way since there's so much suffering that humans endure from degenerative diseases, from cancer, from heart disease and stroke. The potential for increasing happiness and well-being from biomedical research is just stupendous. So before we start to drag out Brave New World for the umpteenth time, or compare every advance in genetics to the Nazis, we should remember the costs of people dying prematurely from postponing advances in biomedical research.
Later in the book, you mention how much more efficient the production of food has become due to high-tech agriculture. But so many people today are leery of advances in the food industry, like GMOs. And we will have to feed 10 billion people in 2050. Are you concerned about how we will meet that challenge?
Yes, I think anyone has to be, and all the more reason we should be clear about what is simultaneously best for humans and for the planet, which is to grow as much food on this planet as possible. That ideal of density -- the less farmland the better -- runs up against the ideal of the organic farming and natural farming, which use lots of land. So genetically modified organisms and precision agriculture of the kind that is sometimes associated with Israel -- putting every last drop of water to use, delivering it when it's needed, using the minimum amount of fertilizer -- all of these technologically driven developments are going to be necessary to meet that need.
"The potential for increasing happiness and well-being from biomedical research is just stupendous."
You also mention "sustainability" as this big buzz word that you say is based on a flawed assumption that we will run out of resources rather than pivot to ingenious alternatives. What's the most important thing we can do as a culture to encourage innovation?
It has to be an ideal. We have restore it as what we need to encourage, to glorify in order to meet the needs of humanity. Governments have to play a role because lots of innovation is just too risky with benefits that are too widely diffuse for private companies and individuals to pursue. International cooperation has to play a role. And also, we need to change our environmental philosophy from a reflexive rejection of technology to an acknowledgement that it will be technology that is our best hope for staving off environmental problems.
And yet innovation and technology today are so often viewed fearfully by the public -- just look at AI and gene editing. If we need science and technology to solve our biggest challenges, how do we overcome this disconnect?
Part of it is simply making the argument that is challenging the ideology and untested assumptions behind traditional Greenism. Also, on the part of the promoters of technology themselves, it's crucial to make it not just clear, but to make it a reality that technology is going to be deployed to enhance human welfare.
That of course means an acknowledgement of the possible harms and limitations of technology. The fact that the first widely used genetically modified crop was soybeans that were resistant to herbicides, to Roundup -- that was at the very least a public relations disaster for genetically modified organisms. As opposed to say, highlighting crops that require less insecticide, less chemical fertilizers, less water level. The poster children for technology should really be cases that quite obviously benefit humanity.
"One of the surprises from 'Enlightenment Now' was how much moral progress depends on economic progress."
Finally, what is one emerging innovation that you're excited about for 2019?
I would say 4th generation nuclear power. Small modular reactors. Because everything depends on energy. For poor countries to get rich, they are going to have to consume far more energy than they do now and if they do it via fossil fuels, especially coal, that could spell disaster. Zero-carbon energy will allow poor countries to get richer -- and rich countries to stay rich without catastrophic environmental damage.
One of the surprises from "Enlightenment Now" was how much moral progress depends on economic progress. Rich countries not only allow the citizens to have cool gadgets, but all kinds of good things happen when a country gets rich, like Norway, Netherlands, Switzerland. Countries that are richer on average are more democratic, are less likely that to fight wars, are more feminist, are more environmentally conscientious, are smarter -- that is, they have a greater increase in IQ. So anything that makes a country get richer, and that's going to include a bunch of energy, is going to make humanity better off.