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."
Inside the Atlantis Space Shuttle, astronauts waited for liftoff. At T-minus six seconds, the main engines ignited, rattling the capsule “like a skyscraper in an earthquake,” according to astronaut Tom Jones, describing the 1988 launch in Air & Space Magazine. Liftoff came with what felt like “a massive kick in the back,” he recalled, along with more shaking. As the rocket accelerated to three times the force of gravity on Earth, “It felt as if two of my friends were standing on my chest and wouldn’t get off!” Finally, at 25 times the speed of sound, Atlantis reached orbit. The main engines cut off, and the astronauts were weightless.
Since 1961, NASA has sent hundreds of astronauts into space while working to making their voyages safer and smoother. Yet, challenges remain. Weightlessness may look amusing when watched from Earth, but it has myriad effects on cognition, movement and other functions. When missions to space stretch to six months or longer, microgravity can harm astronauts’ health and performance, making it more difficult to operate their spacecraft.
Yesterday, NASA astronaut Frank Rubio returned to Earth after over one year, the longest single spaceflight for a U.S. astronaut. But this is just the start; longer and more complex missions into deep space loom ahead, from returning to the moon in 2025 to eventually sending humans to Mars. Understanding how spaceflight affects the body is vital to success. By studying these impacts, NASA aims to help astronauts perform in space as well as they do on Earth.
The dangers of microgravity are real
A NASA report published in 2016 details a long list of incidents and near-misses caused – at least partly – by space-induced changes in astronauts’ vision and coordination. These issues make it harder to move with precision and to judge distance and velocity.
According to the report, in 1997, a resupply ship collided with the Mir space station, possibly because a crew member bumped into the commander during the final docking maneuver. This mishap caused significant damage to the space station.
Returns to Earth suffered from problems, too. The same report notes that touchdown speeds during the first 100 space shuttle landings were “outside acceptable limits. The fastest landing on record – 224 knots (258 miles) per hour – was linked to the commander’s momentary spatial disorientation.” Earlier, each of the six Apollo crews that landed on the moon had difficulty recognizing moon landmarks and estimating distances. For example, Apollo 15 landed in an unplanned area, ultimately straddling the rim of a five-foot deep crater on the moon, harming one of its engines.
Spaceflight causes unique stresses on astronauts’ brains and central nervous systems. NASA is working to reduce these harmful effects.
Space messes up your brain
In space, astronauts face the challenges of microgravity, ionizing radiation, social isolation, high workloads, altered circadian rhythms, monotony, confined living quarters and a high-risk environment. Among these issues, microgravity is one of the most consequential in terms of physiological changes. It changes the brain’s structure and its functioning, which can hurt astronauts’ performance.
The brain shifts upwards within the skull, displacing the cerebrospinal fluid, which reduces the brain’s cushioning. Essentially, the brain becomes crowded inside the skull like a pair of too-tight shoes.
That’s partly because of how being in space alters blood flow. On Earth, gravity pulls our blood and other internal fluids toward our feet, but our circulatory valves ensure that the fluids are evenly distributed throughout the body. In space, there’s not enough gravity to pull the fluids down, and they shift up, says Rachael D. Seidler, a physiologist specializing in spaceflight at the University of Florida and principal investigator on many space-related studies. The head swells and legs appear thinner, causing what astronauts call “puffy face chicken legs.”
“The brain changes at the structural and functional level,” says Steven Jillings, equilibrium and aerospace researcher at the University of Antwerp in Belgium. “The brain shifts upwards within the skull,” displacing the cerebrospinal fluid, which reduces the brain’s cushioning. Essentially, the brain becomes crowded inside the skull like a pair of too-tight shoes. Some of the displaced cerebrospinal fluid goes into cavities within the brain, called ventricles, enlarging them. “The remaining fluids pool near the chest and heart,” explains Jillings. After 12 consecutive months in space, one astronaut had a ventricle that was 25 percent larger than before the mission.
Some changes reverse themselves while others persist for a while. An example of a longer-lasting problem is spaceflight-induced neuro-ocular syndrome, which results in near-sightedness and pressure inside the skull. A study of approximately 300 astronauts shows near-sightedness affects about 60 percent of astronauts after long missions on the International Space Station (ISS) and more than 25 percent after spaceflights of only a few weeks.
Another long-term change could be the decreased ability of cerebrospinal fluid to clear waste products from the brain, Seidler says. That’s because compressing the brain also compresses its waste-removing glymphatic pathways, resulting in inflammation, vulnerability to injuries and worsening its overall health.
The effects of long space missions were best demonstrated on astronaut twins Scott and Mark Kelly. This NASA Twins Study showed multiple, perhaps permanent, changes in Scott after his 340-day mission aboard the ISS, compared to Mark, who remained on Earth. The differences included declines in Scott’s speed, accuracy and cognitive abilities that persisted longer than six months after returning to Earth in March 2016.
By the end of 2020, Scott’s cognitive abilities improved, but structural and physiological changes to his eyes still remained, he said in a BBC interview.
“It seems clear that the upward shift of the brain and compression of the surrounding tissues with ventricular expansion might not be a good thing,” Seidler says. “But, at this point, the long-term consequences to brain health and human performance are not really known.”
NASA astronaut Kate Rubins conducts a session for the Neuromapping investigation.
Staying sharp in space
To investigate how prolonged space travel affects the brain, NASA launched a new initiative called the Complement of Integrated Protocols for Human Exploration Research (CIPHER). “CIPHER investigates how long-duration spaceflight affects both brain structure and function,” says neurobehavioral scientist Mathias Basner at the University of Pennsylvania, a principal investigator for several NASA studies. “Through it, we can find out how the brain adapts to the spaceflight environment and how certain brain regions (behave) differently after – relative to before – the mission.”
To do this, he says, “Astronauts will perform NASA’s cognition test battery before, during and after six- to 12-month missions, and will also perform the same test battery in an MRI scanner before and after the mission. We have to make sure we better understand the functional consequences of spaceflight on the human brain before we can send humans safely to the moon and, especially, to Mars.”
As we go deeper into space, astronauts cognitive and physical functions will be even more important. “A trip to Mars will take about one year…and will introduce long communication delays,” Seidler says. “If you are on that mission and have a problem, it may take eight to 10 minutes for your message to reach mission control, and another eight to 10 minutes for the response to get back to you.” In an emergency situation, that may be too late for the response to matter.
“On a mission to Mars, astronauts will be exposed to stressors for unprecedented amounts of time,” Basner says. To counter them, NASA is considering the continuous use of artificial gravity during the journey, and Seidler is studying whether artificial gravity can reduce the harmful effects of microgravity. Some scientists are looking at precision brain stimulation as a way to improve memory and reduce anxiety due to prolonged exposure to radiation in space.
To boldly go where no astronauts have gone before, they must have optimal reflexes, vision and decision-making. In the era of deep space exploration, the brain—without a doubt—is the final frontier.
Additionally, NASA is scrutinizing each aspect of the mission, including astronaut exercise, nutrition and intellectual engagement. “We need to give astronauts meaningful work. We need to stimulate their sensory, cognitive and other systems appropriately,” Basner says, especially given their extreme confinement and isolation. The scientific experiments performed on the ISS – like studying how microgravity affects the ability of tissue to regenerate is a good example.
“We need to keep them engaged socially, too,” he continues. The ISS crew, for example, regularly broadcasts from space and answers prerecorded questions from students on Earth, and can engage with social media in real time. And, despite tight quarters, NASA is ensuring the crew capsule and living quarters on the moon or Mars include private space, which is critical for good mental health.
Exploring deep space builds on a foundation that began when astronauts first left the planet. With each mission, scientists learn more about spaceflight effects on astronauts’ bodies. NASA will be using these lessons to succeed with its plans to build science stations on the moon and, eventually, Mars.
“Through internally and externally led research, investigations implemented in space and in spaceflight simulations on Earth, we are striving to reduce the likelihood and potential impacts of neurostructural changes in future, extended spaceflight,” summarizes NASA scientist Alexandra Whitmire. To boldly go where no astronauts have gone before, they must have optimal reflexes, vision and decision-making. In the era of deep space exploration, the brain—without a doubt—is the final frontier.
Swiss researchers have discovered a third type of brain cell that appears to be a hybrid of the two other primary types — and it could lead to new treatments for many brain disorders.
The challenge: Most of the cells in the brain are either neurons or glial cells. While neurons use electrical and chemical signals to send messages to one another across small gaps called synapses, glial cells exist to support and protect neurons.
Astrocytes are a type of glial cell found near synapses. This close proximity to the place where brain signals are sent and received has led researchers to suspect that astrocytes might play an active role in the transmission of information inside the brain — a.k.a. “neurotransmission” — but no one has been able to prove the theory.
A new brain cell: Researchers at the Wyss Center for Bio and Neuroengineering and the University of Lausanne believe they’ve definitively proven that some astrocytes do actively participate in neurotransmission, making them a sort of hybrid of neurons and glial cells.
According to the researchers, this third type of brain cell, which they call a “glutamatergic astrocyte,” could offer a way to treat Alzheimer’s, Parkinson’s, and other disorders of the nervous system.
“Its discovery opens up immense research prospects,” said study co-director Andrea Volterra.
The study: Neurotransmission starts with a neuron releasing a chemical called a neurotransmitter, so the first thing the researchers did in their study was look at whether astrocytes can release the main neurotransmitter used by neurons: glutamate.
By analyzing astrocytes taken from the brains of mice, they discovered that certain astrocytes in the brain’s hippocampus did include the “molecular machinery” needed to excrete glutamate. They found evidence of the same machinery when they looked at datasets of human glial cells.
Finally, to demonstrate that these hybrid cells are actually playing a role in brain signaling, the researchers suppressed their ability to secrete glutamate in the brains of mice. This caused the rodents to experience memory problems.
“Our next studies will explore the potential protective role of this type of cell against memory impairment in Alzheimer’s disease, as well as its role in other regions and pathologies than those explored here,” said Andrea Volterra, University of Lausanne.
But why? The researchers aren’t sure why the brain needs glutamatergic astrocytes when it already has neurons, but Volterra suspects the hybrid brain cells may help with the distribution of signals — a single astrocyte can be in contact with thousands of synapses.
“Often, we have neuronal information that needs to spread to larger ensembles, and neurons are not very good for the coordination of this,” researcher Ludovic Telley told New Scientist.
Looking ahead: More research is needed to see how the new brain cell functions in people, but the discovery that it plays a role in memory in mice suggests it might be a worthwhile target for Alzheimer’s disease treatments.
The researchers also found evidence during their study that the cell might play a role in brain circuits linked to seizures and voluntary movements, meaning it’s also a new lead in the hunt for better epilepsy and Parkinson’s treatments.
“Our next studies will explore the potential protective role of this type of cell against memory impairment in Alzheimer’s disease, as well as its role in other regions and pathologies than those explored here,” said Volterra.