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."
Today’s podcast guest is Rosalind Picard, a researcher, inventor named on over 100 patents, entrepreneur, author, professor and engineer. When it comes to the science related to endowing computer software with emotional intelligence, she wrote the book. It’s published by MIT Press and called Affective Computing.
Dr. Picard is founder and director of the MIT Media Lab’s Affective Computing Research Group. Her research and engineering contributions have been recognized internationally. For example, she received the 2022 International Lombardy Prize for Computer Science Research, considered by many to be the Nobel prize in computer science.
Through her research and companies, Dr. Picard has developed wearable sensors, algorithms and systems for sensing, recognizing and responding to information about human emotion. Her products are focused on using fitness trackers to advance clinical quality treatments for a range of conditions.
Meanwhile, in just the past few years, numerous fitness tracking companies have released products with their own stress sensors and systems. You may have heard about Fitbit’s Stress Management Score, or Whoop’s Stress Monitor – these features and apps measure things like your heart rhythm and a certain type of invisible sweat to identify stress. They’re designed to raise awareness about forms of stress such as anxieties and anger, and suggest strategies like meditation to relax in real time when stress occurs.
But how well do these off-the-shelf gadgets work? There’s no one more knowledgeable and experienced than Rosalind Picard to explain the science behind these stress features, what they do exactly, how they might be able to help us, and their current shortcomings.
Dr. Picard is a member of the National Academy of Engineering and a Fellow of the National Academy of Inventors, and a popular speaker who’s given over a hundred invited keynote talks and a TED talk with over 2 million views. She holds a Bachelors in Electrical Engineering from Georgia Tech, and Masters and Doctorate degrees in Electrical Engineering and Computer Science from MIT. She lives in Newton, Massachusetts with her husband, where they’ve raised three sons.
In our conversation, we discuss stress scores on fitness trackers to improve well-being. She describes the difference between commercial products that might help people become more mindful of their health and products that are FDA approved and really capable of advancing the science. We also talk about several fascinating findings and concepts discovered in Dr. Picard’s lab including the multiple arousal theory, a phenomenon you’ll want to hear about. And we explore the complexity of stress, one reason it’s so tough to measure. For example, many forms of stress are actually good for us. Can fitness trackers tell the difference between stress that’s healthy and unhealthy?
- Dr. Picard’s book, Affective Computing
- Dr. Picard’s bio
- Dr. Picard on Twitter
- Dr. Picard’s company, Empatica - https://www.empatica.com/ - The FDA-cleared Empatica Health Monitoring Platform provides accurate, continuous health insights for researchers and clinicians, collected in the real world
- Empatica Twitter
- Dr. Picard and her team have published hundreds of peer-reviewed articles across AI, Machine Learning, Affective Computing, Digital Health, and Human-computer interaction.
- Dr. Picard’s TED talk
If you look back on the last century of scientific achievements, you might notice that most of the scientists we celebrate are overwhelmingly white, while scientists of color take a backseat. Since the Nobel Prize was introduced in 1901, for example, no black scientists have landed this prestigious award.
The work of black women scientists has gone unrecognized in particular. Their work uncredited and often stolen, black women have nevertheless contributed to some of the most important advancements of the last 100 years, from the polio vaccine to GPS.
Here are five black women who have changed science forever.
Dr. May Edward Chinn
Dr. May Edward Chinn practicing medicine in Harlem
George B. Davis, PhD.
Chinn was born to poor parents in New York City just before the start of the 20th century. Although she showed great promise as a pianist, playing with the legendary musician Paul Robeson throughout the 1920s, she decided to study medicine instead. Chinn, like other black doctors of the time, were barred from studying or practicing in New York hospitals. So Chinn formed a private practice and made house calls, sometimes operating in patients’ living rooms, using an ironing board as a makeshift operating table.
Chinn worked among the city’s poor, and in doing this, started to notice her patients had late-stage cancers that often had gone undetected or untreated for years. To learn more about cancer and its prevention, Chinn begged information off white doctors who were willing to share with her, and even accompanied her patients to other clinic appointments in the city, claiming to be the family physician. Chinn took this information and integrated it into her own practice, creating guidelines for early cancer detection that were revolutionary at the time—for instance, checking patient health histories, checking family histories, performing routine pap smears, and screening patients for cancer even before they showed symptoms. For years, Chinn was the only black female doctor working in Harlem, and she continued to work closely with the poor and advocate for early cancer screenings until she retired at age 81.
Pictorial Press Ltd/Alamy
Alice Ball was a chemist best known for her groundbreaking work on the development of the “Ball Method,” the first successful treatment for those suffering from leprosy during the early 20th century.
In 1916, while she was an undergraduate student at the University of Hawaii, Ball studied the effects of Chaulmoogra oil in treating leprosy. This oil was a well-established therapy in Asian countries, but it had such a foul taste and led to such unpleasant side effects that many patients refused to take it.
So Ball developed a method to isolate and extract the active compounds from Chaulmoogra oil to create an injectable medicine. This marked a significant breakthrough in leprosy treatment and became the standard of care for several decades afterward.
Unfortunately, Ball died before she could publish her results, and credit for this discovery was given to another scientist. One of her colleagues, however, was able to properly credit her in a publication in 1922.
onathan Newton/The Washington Post/Getty
The person who arguably contributed the most to scientific research in the last century, surprisingly, wasn’t even a scientist. Henrietta Lacks was a tobacco farmer and mother of five children who lived in Maryland during the 1940s. In 1951, Lacks visited Johns Hopkins Hospital where doctors found a cancerous tumor on her cervix. Before treating the tumor, the doctor who examined Lacks clipped two small samples of tissue from Lacks’ cervix without her knowledge or consent—something unthinkable today thanks to informed consent practices, but commonplace back then.
As Lacks underwent treatment for her cancer, her tissue samples made their way to the desk of George Otto Gey, a cancer researcher at Johns Hopkins. He noticed that unlike the other cell cultures that came into his lab, Lacks’ cells grew and multiplied instead of dying out. Lacks’ cells were “immortal,” meaning that because of a genetic defect, they were able to reproduce indefinitely as long as certain conditions were kept stable inside the lab.
Gey started shipping Lacks’ cells to other researchers across the globe, and scientists were thrilled to have an unlimited amount of sturdy human cells with which to experiment. Long after Lacks died of cervical cancer in 1951, her cells continued to multiply and scientists continued to use them to develop cancer treatments, to learn more about HIV/AIDS, to pioneer fertility treatments like in vitro fertilization, and to develop the polio vaccine. To this day, Lacks’ cells have saved an estimated 10 million lives, and her family is beginning to get the compensation and recognition that Henrietta deserved.
Dr. Gladys West
Gladys West was a mathematician who helped invent something nearly everyone uses today. West started her career in the 1950s at the Naval Surface Warfare Center Dahlgren Division in Virginia, and took data from satellites to create a mathematical model of the Earth’s shape and gravitational field. This important work would lay the groundwork for the technology that would later become the Global Positioning System, or GPS. West’s work was not widely recognized until she was honored by the US Air Force in 2018.
Dr. Kizzmekia "Kizzy" Corbett
At just 35 years old, immunologist Kizzmekia “Kizzy” Corbett has already made history. A viral immunologist by training, Corbett studied coronaviruses at the National Institutes of Health (NIH) and researched possible vaccines for coronaviruses such as SARS (Severe Acute Respiratory Syndrome) and MERS (Middle East Respiratory Syndrome).At the start of the COVID pandemic, Corbett and her team at the NIH partnered with pharmaceutical giant Moderna to develop an mRNA-based vaccine against the virus. Corbett’s previous work with mRNA and coronaviruses was vital in developing the vaccine, which became one of the first to be authorized for emergency use in the United States. The vaccine, along with others, is responsible for saving an estimated 14 million lives.
Sarah Watts is a health and science writer based in Chicago.