The Robots Are Coming!

by Andres Oppenheimer

  Unlike most businesspeople who brag about their products, Shoham seemed to play down his accomplishments. When I asked him about his more than thirty international patents—which is more than what most of Israel’s neighbors and many other countries have registered in recent years—he tried to change the conversation. After I insisted, he talked about his innovations as if they were collective projects in which he had played only a small role. When we finally got onto the topic I was there for—the micro-robot that could unclog human blood vessels, which he was developing along with another company called Microbot Medical—Shoham started reciting a long list of names of people who had worked with him on the project, and looked somewhat annoyed when he noticed I wasn’t jotting down the names. I tried to explain that I wouldn’t have enough space in this book to get into so much detail, but Shoham didn’t accept that answer. He told me, in a polite yet firm tone, that he was morally obliged to give me all the names of his colleagues who had worked with him to develop this micro-robot. He then repeated their names and waited for me to write them down. And so I did, realizing that it would be the quickest way to move forward with the interview.

  THE MICRO-ROBOT THAT CLEANS ARTERIES

  Shoham explained that his Microbot Medical micro-robot would be among the first to be used to clean and drain the urethra and even vascular tubes in the brain without the need for invasive operations. It’s essentially a titanium robot one millimeter in diameter and up to four millimeters long that is inserted into the body and operated externally by remote control. But in the very near future, he explained, it will also be used to remove plaque from coronary arteries, reducing the risk of heart attacks. Asked how long before it will be used on patients, Shoham replied that they were just then finishing up the animal testing phase and that the product could be available to the public as early as 2020 or 2021.

  The micro-robot, known as ViRob, will do much more than cleaning. It will perform biopsies without the need for doctors to open up the body to take samples of potentially malignant tissue, he said. For instance, ViRob will travel—perhaps “swim” is a better word—through the body to reach the hidden cavities that are difficult for even the most skilled surgeons to reach, and perform the biopsy. And perhaps even more important, ViRob will also be used to detect and discharge localized treatments for diseases such as cancer. It will be able to release cancer-fighting medication directly into the tissue without the need for invasive treatments like chemotherapy, Shoham said.

  Like many other inventions, the idea of a micro-robot to clean out the body’s internal plumbing came to Shoham almost by chance. A few years after developing his robot-surgeon for spinal operations, Shoham and his team of Ph.D. students at the Technion had begun working on a robot to clean cities’ sewage and water pipes and to fix leaks in their plumbing systems. Their original theory was that cities could save millions of dollars a year by using a robot that could clean up the filth that clogs pipes or locate leaks in the pipe system. But after Shoham made a speech about his new project at an interdisciplinary conference, he was approached by Menashe Zaaroor, a fellow professor of neurosurgery at the Technion and director of the department of neurosurgery at Rambam Medical Center in Haifa. Why not create a miniaturized version of that robot for medical uses? Zaaroor asked. If a robot can be used to clean up clogged city pipes, why not create one that does the same with the human body’s plumbing system? Zaaroor had explained to Shoham that such a mini-robot could be extremely useful in cases of hydrocephalus, a condition in which there is an accumulation of cerebrospinal fluid in the brain. Until now, doctors had not been able to solve the problem of blocked drainage tubes in these cases. Patients had to undergo multiple operations to clear out these tubes. So, the neurosurgeon asked, why not use micro-robots to clean them instead? Soon enough, Shoham and Zaaroor founded Microbot Medical.

  Shoham laughed when I asked him if his microbot would be replacing human surgeons in the near future. He said the first time he’d been asked that question was seventeen years earlier, at a robotics conference. Since then, journalists inevitably turned out articles about his research with provocative titles like “Robots to Replace Surgeons,” he commented. Responding to my question, he said that “robots will not replace surgeons. What they will do is change the sort of work that surgeons do. Surgeons will have to program the robot, and they will need a lot of experience in programming and engineering.” In other words, surgeons—at least for now, until artificial intelligence evolves further—will be directing the surgical robots.

  Shoham showed me on his laptop an illustration of what a typical operation will look like in the future. It showed a patient laid out on a stretcher while a robot was arched over him performing the surgery, with the doctor directing—or rather, controlling—the procedure from an insulated room next door to protect him from radiation. Or in the case of microbots, the doctor in the adjacent room will be using a joystick to steer the micro-robot toward a tumor and will press a key to shoot the medicine directly into the malignant tissue. Like with other nano-robots that are being developed in other countries, the device will disintegrate and dissolve into the bloodstream once it has fulfilled its mission, leaving no need for it to be removed from the body. “Micro-robots have several advantages,” Shoham explained. “They give the surgeon more accessibility, because they can reach nearly inaccessible parts of the body without the need for invasive surgery, and they also have much more precision, because a well-programmed robot makes fewer errors than humans,” he said.

  THE INVENTIONS THAT ARE CHANGING MODERN MEDICINE

  Shoham’s inventions are just a few of many examples of how medicine will be transformed thanks to robots, micro-robots, sensors inserted in watches or clothes, chips implanted under our skin, smartphones with self-diagnosis apps, telemedicine, preventive medicine, predictive data analysis, 3-D printers that will produce tailor-made organs, and virtual reality treatments.

  The most elementary and revolutionary medical innovation is one that most of us are already using in everyday life: the Google search engine, and virtual assistants like Siri or Alexa. In coming years, we will see more specialized versions of these all-purpose assistants that will be able to answer any medical questions, much as if we were talking with our personal doctor. Plus, they will send us daily reminders about exercising and taking our medicine. Alexa can already give precise instructions on what to do if you have a heart attack or how to perform CPR on someone in an emergency. Increasingly, we will be doing our medical consultations with our virtual assistants, at home or wherever we are.

  Medical virtual assistants, digital sensors, and many other similar innovations are already being used by millions of people, but in separate and disorganized ways. That will change, however. Just as smartphones existed for several years before the iPhone set the standard and became a global phenomenon, it’s only a matter of time—perhaps even as you are reading these lines—until someone comes up with a device that integrates most existing apps for medical consultations, and that will allow us to diagnose and treat many of our illnesses.

  Most gurus of medical technology say that it’s time for medicine to stop being a practice or an art, and become more of a hard science. And for that to happen, human physicians will have to turn over much of their work to computers and robots, they say. Indeed, today’s doctors are still using ancient tools, as if they had been frozen in time. Does it make sense for physicians to still use pretty much the same stethoscopes they have been using for a hundred years to listen to our heartbeats? Or to measure our blood pressure with a sleeve that is inflated with a rubber ball, much as they did a century ago? Today, digital sensors have made many of these devices look prehistoric, technologists say.

  EVERY DOCTOR CAN BE UPDATED

  The number of medical discoveries is growing so fast, with tens of thousands of new scientific studies being published every year around the world, that it makes no sense for doctors to
continue working based on the knowledge they acquired in medical school many years ago, or on their own intuition. No human being is capable of either capturing or retaining the amount of new medical information that is coming to light every day. While in the past one of medicine’s primary problems was the lack of information, today’s problem is that there is too much information.

  When it comes to diagnosing diseases, doctors just can’t beat computers. A doctor can make a diagnosis based on his experience with the few thousand patients whom he has seen, but a computer can make its judgment based on data obtained from hundreds of millions of cases. And if Watson, the IBM computer that is already being used by a number of hospitals across the United States, determines that 300 million patients did better with one medication than with another, its diagnosis is much safer than one made by a human doctor, the medical futurists say.

  According to IBM, Watson can digest information and make recommendations by processing up to 60 million pages of text per second, including all the handwritten notes doctors might have added to medical records on file, academic articles in specialized publications, online figures from public health departments around the world, and other periodicals. And since Watson isn’t just a computer but an artificially intelligent machine, not only does it instantly read all available information about a specific medical issue and make recommendations, but it can also follow a patient’s treatment step by step and learn even more from that case’s results. Watson is constantly acquiring and analyzing data, just as it did when it beat the world champions in chess and Go. And when it diagnoses a disease and recommends a treatment, it can offer a number of different possibilities and their projected success rates.

  TECHNOLOGY WILL REPLACE 80 PERCENT OF WHAT DOCTORS DO

  Vinod Khosla, the Silicon Valley medical technology tycoon and cofounder of Sun Microsystems, has predicted that technology will eventually replace 80 percent of what doctors currently do. But this doesn’t mean that 80 percent of doctors will lose their jobs, he says. Rather, it means that many of the routine tasks they perform—such as checkups, tests, diagnoses, prescription of drugs, behavior modification programs, and data centralization—will be performed much more efficiently by intelligent machines.

  “Most doctors couldn’t possibly read and digest all of the latest 5,000 research articles on heart disease,” Khosla argues. “And, most of the average doctor’s medical knowledge is from when they were in medical school, while cognitive limitations prevent them from remembering the 10,000+ diseases humans can get.”

  To back up his argument, Khosla points out that more people die today due to errors by doctors and nurses than they do from many diseases. According to a study by Johns Hopkins, as many as 40,500 people in the United States die every year in an ICU due to misdiagnoses, rivaling the number of deaths from breast cancer. Medicine should therefore be less intuitive and more scientific, Khosla says. We should forget about the television series House—in which the main character was incredibly rude but had a remarkable ability to solve complex medical cases—and start to let computers handle most tasks that are currently performed by doctors.

  PATIENTS CAN LIE TO THE DOCTOR, BUT NOT TO THE ROBOT

  According to Khosla, besides making better decisions, computers and sensor equipment are much more accurate than doctors when it comes to collecting data on the medical history, the symptoms, and even the mood of patients. Today a doctor will ask us how we’re feeling, and he or she will work with a pen or a tablet and write down whatever we say. But this is an archaic way of doing things, because a patient can easily forget hidden symptoms or even lie if, for example, he’s ashamed to admit that he’s not getting enough sleep because he’s having an extramarital affair. But when patients are wearing a smartwatch with sensors that can track their vital signs all day and night, recording among other things how many hours they slept, this data can be transmitted directly to the doctor’s office via the Internet. Those data will be much more accurate. Patients will always hold back from or even lie to their doctors, but that will be much harder to do when they have a robot connected to their wrist.

  And even when patients don’t lie, they often are not paying close enough attention to their own symptoms. “Today most heart disease is identified only after patients have heart attacks. But imagine having preventative cardiac care, enabled by machine-learning software that identifies abnormalities and predicts episodes,” Khosla wrote back in 2012, before smartwatches with built-in sensors became popular. “We could discover most heart disease before a heart attack or stroke.”

  Finally, artificial intelligence is much more effective because it can consolidate a patient’s entire medical history and solve one of modern medicine’s primary problems: the fact that we go see different specialists, and none of them ever speak to one another. An average person over the age of seventy has at least seven different conditions, and usually goes to a specialist for each of them. As Khosla asks, wouldn’t it be much more logical for a single AI computer to gather all that information and advise us whether the various treatments are compatible or may work against one another? Instead of skipping from one specialist to another, patients in the future will have just one doctor who will help them understand the computer’s diagnosis and recommendation for treatment, he says.

  But what sort of effect will all of this have on a doctor’s work? According to Khosla, computers will free them up from routine checkup diagnostic tests and treatment plans, allowing them instead to focus much more on their role of supporting their patients psychologically and helping them interpret the computer’s data. As Khosla puts it, instead of Dr. House, we’ll have “Dr. Algorithm.” “Eventually, we’ll need fewer doctors,” he says, “and every patient will receive the best care.”

  FROM INTERMITTENT AND REACTIVE MEDICINE TO CONSTANT AND PROACTIVE MEDICINE

  When I interviewed Daniel Kraft, the futurist doctor and chair of Singularity University’s Exponential Medicine program, he did the exact same thing he had done four years earlier when I visited him in Silicon Valley: he showed me all the smartwatches, rings, and clothes with embedded sensors he wears, and—like a magician—started taking out of a box the latest gadgets that he thinks will radically change modern medicine. He explained that these and other devices will allow each of us to make our own diagnoses in the comfort of our own homes without having to go to a doctor or to a hospital.

  Kraft is an authority on the medicine of the future. He studied traditional medicine at Stanford before working in internal medicine and pediatrics at Harvard. But soon enough he launched a parallel career as an inventor of medical products, including a system for growing bone marrow in a faster way, and as a promoter of the latest advances in medical technology.

  While he has long been donning the latest wearables, or external sensors, on the market, Kraft is doing something new these days: he is having all of them linked to his personal doctor’s office at Stanford, where nurses can follow his vital signs twenty-four hours a day and can alert him in advance if they notice anything unusual. It won’t be long before most of us will be wearing sensors that are permanently connected to our primary doctors, he says.

  “We’re moving from an intermittent and reactive medicine to a constant and proactive medicine,” Kraft told me. He added that we already have in our smartphones most of the apps and sensors that are needed to monitor our bodies 24/7. Instead of getting a blood test done every six months or going to the doctor when we’re already feeling sick, medicine will be focused on prevention and early detection of diseases through external sensors and cell phone apps that will be checking on us constantly, he said.

  “The future of medicine will be much more about being preventative and proactive, using your genetic information and your behavioral information to help guide prevention,” he said. “If you have a higher risk of getting a type of cancer or diabetes or Alzheimer’s, then maybe with both exercise and diet,
maybe drugs and other interventions, you can do things early to prevent those diseases from occurring. So instead of spending all our dollars on sick care, I think we’re going to see more and more use of technologies that will be much more on the preventative side.” Not only is this in the best interest of the patient, but it will also help society save money, he added. Countries are going to be emphasizing preventive medicine much more in the future because it’s much cheaper than the current ways of fighting diseases, he explained.

  GOING FROM SICK CARE TO HEALTH CARE

  Kraft then showed me his Fitbit bracelet, which tracks his daily steps; his Apple watch, which measures his pulse; and a ring on his finger that he described as a “sleep lab” for monitoring his sleep at night. All these sensors represent what he calls the “Internet of the body.”