When many of us hear the term “artificial intelligence” (AI), we imagine robots doing our jobs, rendering people obsolete. And, since AI-driven computers are programmed to make decisions with little human intervention, some wonder if machines will soon make the difficult decisions we now entrust to our doctors.
According to David B. Agus, MD, a professor of medicine and engineering at the University of Southern California Keck School of Medicine and Viterbi School of Engineering, it’s important to separate fact from science fiction, because AI is already here — and it’s fundamentally changing medicine.
Rather than robotics, AI in health care mainly refers to doctors and hospitals accessing vast data sets of potentially life-saving information. This includes treatment methods and their outcomes, survival rates, and speed of care gathered across millions of patients, geographical locations, and innumerable and sometimes interconnected health conditions. New computing power can detect and analyze large and small trends from the data and even make predictions through machine learning that’s designed to identify potential health outcomes.
Machine learning uses statistical techniques to give computer systems the ability to “learn” with incoming data and to identify patterns and make decisions with minimal human direction.
Armed with such targeted analytics, doctors may be better able to assess risk, make correct diagnoses, and offer patients more effective treatments, says Agus, the author of The Lucky Years: How to Thrive in the Brave New World of Health and The End of Illness. He believes AI’s potential to improve health care is “staggering.”
“We have lots of data that we’ve been collecting over decades,” he says. “For the first time, computing power allows us to use the data in a way to benefit patients.”
The challenge, he says, is that “an individual has hundreds of thousands of health care data points, if not millions. So when you have data sets of hundreds of thousands of patients, and each patient has a million data points, the data need to be collected appropriately and correctly for the power of machine learning” to bear fruit.
He offers an example. “A study came out recently that showed that if you have ovarian cancer, and you happen to also be on a beta-blocker — a drug that [can be] used for blood pressure — you lived four-and-a-half years longer,” he says. “This is an observation we would never have come up with through biology. Big data shows us. Now [this finding] needs to go to a big trial to see if it’s real.”
From a patient’s perspective, “what’s exciting is AI allows [doctors] to personalize care, something we’ve dreamed of doing for decades,” he says.
Agus can now take an individual patient and immediately find other patients with similar symptoms. “I pull them out of a database,” he says, “and I can say, ‘Here are their reactions.’ Machine learning and AI allow me to [access] all of the information and have a very educated discussion with the patient” sitting in the exam room, “unlocking data [on health conditions] that historically we’ve made simple decisions about. AI allows us to get much deeper and look for associations the human brain isn’t able to do … but a computer can.”
There are, of course, detractors regarding the use of analytics in health care, but concerns tend to focus less on AI, machine learning, and predictive tracking and more on how big data can be used to measure, reward, or penalize an entire hospital’s — or even an individual surgeon’s — performance.
Such measurements can affect how, when, or even if a patient is treated, writes Jerry Muller, author of 2018’s The Tyranny of Metrics. “Nowhere are metrics more in vogue than in the field of medicine,” he says. And with lives on the line, he concludes, “the stakes are high.”
Muller points to the problem of human nature: People and bureaucracies, he says, have been known to “game” the numbers out of self-preservation.
He cites examples of in-demand surgeons maintaining high patient survival rates by refusing to take on riskier cases, thereby potentially eliminating nonstandard treatments — and possible deaths after any kind of medical intervention — from data trends that AI might detect. Rates of success are then artificially inflated, too.
Still, Agus believes that tapping data’s power will bring big innovation. “Algorithms and AI have been around for a while, but we’re learning how to better collect and organize the data,” he says. “This past decade was about molecular biology: We sequenced DNA and looked at its associations, and that was exciting. This is going to be the decade of data.”
With top hospitals across the nation adopting AI and metric analyses with the aim of improving and streamlining care, Agus may be right. In our increasingly wired world, data and destiny are becoming forever linked.
Some examples of tech innovations in health care include the following:
Robotic reflections: Sometimes, robots are in play. A 2017 University of Bristol study found that children with autism have difficulty distinguishing facial expressions. That same year, Dell Technologies launched Milo, a 2-foot-tall, visually expressive robot that teaches autistic kids ages 5 to 17 how to identify signs of emotion, now used in educational facilities in 27 U.S. states.
Connecting ALS patients: Eye-tracking glasses that use AI technology known as brain-computer interface (BCI) enable people who’ve lost the ability to speak or move to communicate again. Patients “type” with their eyes onto a monitor that vocalizes their thoughts through computerized decoding, plus use email, read books, and stay connected to the world.
Detecting AFib: Some types of cardiac arrhythmias, particularly atrial fibrillation, can make heart attacks or strokes more likely. Research from Stanford University shows that AI software can more accurately identify arrhythmias from an electrocardiogram (EKG) than a human expert.
On the horizon: Magnetic resonance imaging (MRI) and computerized axial tomography (CT) scans provide detailed, noninvasive views of the inner body. AI may soon replace the need for additional tissue samples with next-generation radiology tools, enabling virtual biopsies of tumors.
By the Numbers:
1 in 7,000: Number of Americans of all ages with long QT syndrome, a deadly heart disorder, who could one day be helped by Kardio Pro, an AI-powered, at-home heart monitor that detects serious and benign arrhythmias.
30%: Reduction in patient wait time before admittance, reports Johns Hopkins Hospital, after it launched a digital command center in 2016 with 22 monitors to improve patient experience, lessen risk, and streamline flow.
95.5%: Percentage of accuracy, using a special microscope, with which a deep-learning computer program identified cancer cells with precision, according to a 2016 study from UCLA published in Nature Scientific Reports.
Find more articles, browse back issues, and read the current issue of WebMD Magazine.WebMD Magazine – Feature Reviewed by Arefa Cassoobhoy, MD, MPH on November 29, 2018
David B. Agus, MD, professor of medicine and engineering, founding director and CEO, Lawrence J. Ellison Institute for Transformative Medicine.
HealthIT.gov: “Artificial Intelligence for Health and Health Care.”
Health IT Analytics: “Top 12 Ways Artificial Intelligence Will Impact Healthcare,” “Machine Learning Algorithm Outperforms Cardiologists Reading EKGs,” “AI for Imaging Analytics Intrigues Healthcare Orgs, Yet Starts Slow.”
Forbes: “Artificial Intelligence in Healthcare: Separating Reality from Hype.”
Emerj Artificial Intelligence Research: “How America’s 5 Top Hospitals are Using Machine Learning Today.”
Muller, J. The Tyranny of Metrics, Princeton University Press, 2017.
SAS: “Machine Learning.”
Robots4Autism: “Robokind Wins Big at Launch Festival,” “Meet Milo!”
Dell Technologies: “The Robot Enhancing How Children With Autism Learn.”
YouTube: “Milo The Robot helps kids with autism learn social skills,” WCPO.com; “Augie Nieto — EyeMax Eye Tracking AAC Device,” Tobii Dynavox.
NBC News: “ALS Patients Communicate for First Time in Years With New Device.”
Johns Hopkins Medicine: “Command Center to Improve Patient Flow.”
Scientific Reports: “Deep Learning in Label-free Cell Classification.”
© 2019 WebMD, LLC. All rights reserved.
Share this Post