The Death Predictor: A Helpful New Tool or an Ethical Morass?
Whenever Eric Karl Oermann has to tell a patient about a terrible prognosis, their first question is always: "how long do I have?" Oermann would like to offer a precise answer, to provide some certainty and help guide treatment. But although he's one of the country's foremost experts in medical artificial intelligence, Oermann is still dependent on a computer algorithm that's often wrong.
Doctors are notoriously terrible at guessing how long their patients will live.
Artificial intelligence, now often called deep learning or neural networks, has radically transformed language and image processing. It's allowed computers to play chess better than the world's grand masters and outwit the best Jeopardy players. But it still can't precisely tell a doctor how long a patient has left – or how to help that person live longer.
Someday, researchers predict, computers will be able to watch a video of a patient to determine their health status. Doctors will no longer have to spend hours inputting data into medical records. And computers will do a better job than specialists at identifying tiny tumors, impending crises, and, yes, figuring out how long the patient has to live. Oermann, a neurosurgeon at Mount Sinai, says all that technology will allow doctors to spend more time doing what they do best: talking with their patients. "I want to see more deep learning and computers in a clinical setting," he says, "so there can be more human interaction." But those days are still at least three to five years off, Oermann and other researchers say.
Doctors are notoriously terrible at guessing how long their patients will live, says Nigam Shah, an associate professor at Stanford University and assistant director of the school's Center for Biomedical Informatics Research. Doctors don't want to believe that their patient – whom they've come to like – will die. "Doctors over-estimate survival many-fold," Shah says. "How do you go into work, in say, oncology, and not be delusionally optimistic? You have to be."
But patients near the end of life will get better treatment – and even live longer – if they are overseen by hospice or palliative care, research shows. So, instead of relying on human bias to select those whose lives are nearing their end, Shah and his colleagues showed that they could use a deep learning algorithm based on medical records to flag incoming patients with a life expectancy of three months to a year. They use that data to indicate who might need palliative care. Then, the palliative care team can reach out to treating physicians proactively, instead of relying on their referrals or taking the time to read extensive medical charts.
But, although the system works well, Shah isn't yet sure if such indicators actually get the appropriate patients into palliative care. He's recently partnered with a palliative care doctor to run a gold-standard clinical trial to test whether patients who are flagged by this algorithm are indeed a better match for palliative care.
"What is effective from a health system perspective might not be effective from a treating physician's perspective and might not be effective from the patient's perspective," Shah notes. "I don't have a good way to guess everybody's reaction without actually studying it." Whether palliative care is appropriate, for instance, depends on more than just the patient's health status. "If the patient's not ready, the family's not ready and the doctor's not ready, then you're just banging your head against the wall," Shah says. "Given limited capacity, it's a waste of resources" to put that person in palliative care.
The algorithm isn't perfect, but "on balance, it leads to better decisions more often."
Alexander Smith and Sei Lee, both palliative care doctors, work together at the University of California, San Francisco, to develop predictions for patients who come to the hospital with a complicated prognosis or a history of decline. Their algorithm, they say, helps decide if this patient's problems – which might include diabetes, heart disease, a slow-growing cancer, and memory issues – make them eligible for hospice. The algorithm isn't perfect, they both agree, but "on balance, it leads to better decisions more often," Smith says.
Bethany Percha, an assistant professor at Mount Sinai, says that an algorithm may tell doctors that their patient is trending downward, but it doesn't do anything to change that trajectory. "Even if you can predict something, what can you do about it?" Algorithms may be able to offer treatment suggestions – but not what specific actions will alter a patient's future, says Percha, also the chief technology officer of Precise Health Enterprise, a product development group within Mount Sinai. And the algorithms remain challenging to develop. Electronic medical records may be great at her hospital, but if the patient dies at a different one, her system won't know. If she wants to be certain a patient has died, she has to merge social security records of death with her system's medical records – a time-consuming and cumbersome process.
An algorithm that learns from biased data will be biased, Shah says. Patients who are poor or African American historically have had worse health outcomes. If researchers train an algorithm on data that includes those biases, they get baked into the algorithms, which can then lead to a self-fulfilling prophesy. Smith and Lee say they've taken race out of their algorithms to avoid this bias.
Age is even trickier. There's no question that someone's risk of illness and death goes up with age. But an 85-year-old who breaks a hip running a marathon should probably be treated very differently than an 85-year-old who breaks a hip trying to get out of a chair in a dementia care unit. That's why the doctor can never be taken out of the equation, Shah says. Human judgment will always be required in medical care and an algorithm should never be followed blindly, he says.
Experts say that the flaws in artificial intelligence algorithms shouldn't prevent people from using them – carefully.
Researchers are also concerned that their algorithms will be used to ration care, or that insurance companies will use their data to justify a rate increase. If an algorithm predicts a patient is going to end up back in the hospital soon, "who's benefitting from knowing a patient is going to be readmitted? Probably the insurance company," Percha says.
Still, Percha and others say, the flaws in artificial intelligence algorithms shouldn't prevent people from using them – carefully. "These are new and exciting tools that have a lot of potential uses. We need to be conscious about how to use them going forward, but it doesn't mean we shouldn't go down this road," she says. "I think the potential benefits outweigh the risks, especially because we've barely scratched the surface of what big data can do right now."
The Friday Five covers five stories in research that you may have missed this week. There are plenty of controversies and troubling ethical issues in science – and we get into many of them in our online magazine – but this news roundup focuses on scientific creativity and progress to give you a therapeutic dose of inspiration headed into the weekend.
Here are the promising studies covered in this week's Friday Five:
- Kids stressing you out? They could be protecting your health.
- A new device unlocks the heart's secrets
- Super-ager gene transplants
- Surgeons could 3D print your organs before operations
- A skull cap looks into the brain like an fMRI
This article originally appeared in One Health/One Planet, a single-issue magazine that explores how climate change and other environmental shifts are making us more vulnerable to infectious diseases by land and by sea - and how scientists are working on solutions.
On a warm summer day, forests, meadows, and riverbanks should be abuzz with insects—from butterflies to beetles and bees. But bugs aren’t as abundant as they used to be, and that’s not a plus for people and the planet, scientists say. The declining numbers of insects, coupled with climate change, can have devastating effects for people in more ways than one. “Insects have been around for a very long time and can live well without humans, but humans cannot live without insects and the many services they provide to us,” says Philipp Lehmann, a researcher in the Department of Zoology at Stockholm University in Sweden. Their decline is not just bad, Lehmann adds. “It’s devastating news for humans.
”Insects and other invertebrates are the most diverse organisms on the planet. They fill most niches in terrestrial and aquatic environments and drive ecosystem functions. Many insects are also economically vital because they pollinate crops that humans depend on for food, including cereals, vegetables, fruits, and nuts. A paper published in PNAS notes that insects alone are worth more than $70 billion a year to the U.S. economy. In places where pollinators like honeybees are in decline, farmers now buy them from rearing facilities at steep prices rather than relying on “Mother Nature.”
And because many insects serve as food for other species—bats, birds and freshwater fish—they’re an integral part of the ecosystem’s food chain. “If you like to eat good food, you should thank an insect,” says Scott Hoffman Black, an ecologist and executive director of the Xerces Society for Invertebrate Conservation in Portland, Oregon. “And if you like birds in your trees and fish in your streams, you should be concerned with insect conservation.”
Deforestation, urbanization, and agricultural spread have eaten away at large swaths of insect habitat. The increasingly poorly controlled use of insecticides, which harms unintended species, and the proliferation of invasive insect species that disrupt native ecosystems compound the problem.
“There is not a single reason why insects are in decline,” says Jessica L. Ware, associate curator in the Division of Invertebrate Zoology at the American Museum of Natural History in New York, and president of the Entomological Society of America. “There are over one million described insect species, occupying different niches and responding to environmental stressors in different ways.”
Jessica Ware, an entomologist at the American Museum of Natural History, is using DNA methods to monitor insects.
In addition to habitat loss fueling the decline in insect populations, the other “major drivers” Ware identified are invasive species, climate change, pollution, and fluctuating levels of nitrogen, which play a major role in the lifecycle of plants, some of which serve as insect habitants and others as their food. “The causes of world insect population declines are, unfortunately, very easy to link to human activities,” Lehmann says.
Climate change will undoubtedly make the problem worse. “As temperatures start to rise, it can essentially make it too hot for some insects to survive,” says Emily McDermott, an assistant professor in the Department of Entomology and Plant Pathology at the University of Arkansas. “Conversely in other areas, it could potentially also allow other insects to expand their ranges.”
Without Pollinators Humans Will Starve
We may not think much of our planet’s getting warmer by only one degree Celsius, but it can spell catastrophe for many insects, plants, and animals, because it’s often accompanied by less rainfall. “Changes in precipitation patterns will have cascading consequences across the tree of life,” says David Wagner, a professor of ecology and evolutionary biology at the University of Connecticut. Insects, in particular, are “very vulnerable” because “they’re small and susceptible to drying.”
For instance, droughts have put the monarch butterfly at risk of being unable to find nectar to “recharge its engine” as it migrates from Canada and New England to Mexico for winter, where it enters a hibernation state until it journeys back in the spring. “The monarch is an iconic and a much-loved insect,” whose migration “is imperiled by climate change,” Wagner says.
Warming and drying trends in the Western United States are perhaps having an even more severe impact on insects than in the eastern region. As a result, “we are seeing fewer individual butterflies per year,” says Matt Forister, a professor of insect ecology at the University of Nevada, Reno.
There are hundreds of butterfly species in the United States and thousands in the world. They are pollinators and can serve as good indicators of other species’ health. “Although butterflies are only one group among many important pollinators, in general we assume that what’s bad for butterflies is probably bad for other insects,” says Forister, whose research focuses on butterflies. Climate change and habitat destruction are wreaking havoc on butterflies as well as plants, leading to a further indirect effect on caterpillars and butterflies.
Different insect species have different levels of sensitivity to environmental changes. For example, one-half of the bumblebee species in the United States are showing declines, whereas the other half are not, says Christina Grozinger, a professor of entomology at the Pennsylvania State University. Some species of bumble bees are even increasing in their range, seemingly resilient to environmental changes. But other pollinators are dwindling to the point that farmers have to buy from the rearing facilities, which is the case for the California almond industry. “This is a massive cost to the farmer, which could be provided for free, in case the local habitats supported these pollinators,” Lehmann says.
For bees and other insects, climate change can harm the plants they depend on for survival or have a negative impact on the insects directly. Overly rainy and hot conditions may limit flowering in plants or reduce the ability of a pollinator to forage and feed, which then decreases their reproductive success, resulting in dwindling populations, Grozinger explains.
“Nutritional deprivation can also make pollinators more sensitive to viruses and parasites and therefore cause disease spread,” she says. “There are many ways that climate change can reduce our pollinator populations and make it more difficult to grow the many fruit, vegetable and nut crops that depend on pollinators.”
Disease-Causing Insects Can Bring More Outbreaks
While some much-needed insects are declining, certain disease-causing species may be spreading and proliferating, which is another reason for human concern. Many mosquito types spread malaria, Zika virus, West Nile virus, and a brain infection called equine encephalitis, along with other diseases as well as heartworms in dogs, says Michael Sabourin, president of the Vermont Entomological Society. An animal health specialist for the state, Sabourin conducts vector surveys that identify ticks and mosquitoes.
Scientists refer to disease-carrying insects as vector species and, while there’s a limited number of them, many of these infections can be deadly. Fleas were a well-known vector for the bubonic plague, while kissing bugs are a vector for Chagas disease, a potentially life-threatening parasitic illness in humans, dogs, and other mammals, Sabourin says.
As the planet heats up, some of the creepy crawlers are able to survive milder winters or move up north. Warmer temperatures and a shorter snow season have spawned an increasing abundance of ticks in Maine, including the blacklegged tick (Ixodes scapularis), known to transmit Lyme disease, says Sean Birkel, an assistant professor in the Climate Change Institute and Cooperative Extension at the University of Maine.
Coupled with more frequent and heavier precipitation, rising temperatures bring a longer warm season that can also lead to a longer period of mosquito activity. “While other factors may be at play, climate change affects important underlying conditions that can, in turn, facilitate the spread of vector-borne disease,” Birkel says.
For example, if mosquitoes are finding fewer of their preferred food sources, they may bite humans more. Both male and female mosquitoes feed on sugar as part of their normal behavior, but if they aren’t eating their fill, they may become more bloodthirsty. One recent paper found that sugar-deprived Anopheles gambiae females go for larger blood meals to stay in good health and lay eggs. “More blood meals equals more chances to pick up and transmit a pathogen,” McDermott says, He adds that climate change could reduce the number of available plants to feed on. And while most mosquitoes are “generalist sugar-feeders” meaning that they will likely find alternatives, losing their favorite plants can make them hungrier for blood
Similar to the effect of losing plants, mosquitoes may get turned onto people if they lose their favorite animal species. For example, some studies found that Culex pipiens mosquitoes that transmit the West Nile virus feed primarily on birds in summer. But that changes in the fall, at least in some places. Because there are fewer birds around, C. pipiens switch to mammals, including humans. And if some disease-carrying insect species proliferate or increase their ranges, that increases chances for human infection, says McDermott. “A larger concern is that climate change could increase vector population sizes, making it more likely that people or animals would be bitten by an infected insect.”
Science Can Help Bring Back the Buzz
To help friendly insects thrive and keep the foes in check, scientists need better ways of trapping, counting, and monitoring insects. It’s not an easy job, but artificial intelligence and molecular methods can help. Ware’s lab uses various environmental DNA methods to monitor freshwater habitats. Molecular technologies hold much promise. The so-called DNA barcodes, in which species are identified using a short string of their genes, can now be used to identify birds, bees, moths and other creatures, and should be used on a larger scale, says Wagner, the University of Connecticut professor. “One day, something akin to Star Trek’s tricorder will soon be on sale down at the local science store.”
Scientists are also deploying artificial intelligence, or AI, to identify insects in agricultural systems and north latitudes where there are fewer bugs, Wagner says. For instance, some automated traps already use the wingbeat frequencies of mosquitoes to distinguish the harmless ones from the disease-carriers. But new technology and software are needed to further expand detection based on vision, sound, and odors.
“Because of their ubiquity, enormity of numbers, and seemingly boundless diversity, we desperately need to develop molecular and AI technologies that will allow us to automate sampling and identification,” says Wagner. “That would accelerate our ability to track insect populations, alert us to the presence of new disease vectors, exotic pest introductions, and unexpected declines.”