Proceedings of the Standing Senate Committee on
Social Affairs, Science and Technology
Issue No. 14 - Evidence - February 2, 2017
OTTAWA, Thursday, February 2, 2017
The Standing Senate Committee on Social Affairs, Science and Technology met today at 10:30, to continue its study on the role of robotics, 3D printing and artificial intelligence in the healthcare system.
Senator Kelvin Kenneth Ogilvie (Chair) in the chair.
[Translation]
The Chair: Welcome to the Standing Senate Committee on Social Affairs, Science and Technology.
[English]
I'm Kelvin Ogilvie, a senator from Nova Scotia and chair of the committee. I would like to start the meeting by asking my colleagues to introduce themselves.
Senator Mégie: My name is Marie-Françoise Mégie. I'm from Montreal.
[Translation]
I was a family physician until December 31, 2016. I retired recently and am now a senator in the Senate of Canada.
[English]
Senator Stewart Olsen: Carolyn Stewart Olsen, New Brunswick.
Senator Seidman: Judith Seidman from Montreal, Quebec.
Senator Raine: Nancy Greene Raine from B.C.
[Translation]
Senator Petitclerc: Chantal Petitclerc from Montreal, Quebec.
Senator Hartling: Hello, I am Nancy Hartling from New Brunswick.
[English]
Senator Dean: Tony Dean from Toronto, representing Ontario.
[Translation]
Senator Cormier: René Cormier, senator from New Brunswick.
[English]
Senator Meredith: Senator Don Meredith, Ontario.
Senator Merchant: Pana Merchant, Saskatchewan.
Senator Eggleton: Art Eggleton, senator from Toronto and deputy chair of the committee.
The Chair: Thank you, colleagues. Today we are continuing our study, which we began yesterday, of the role of robotics, 3-D printing and artificial intelligence in the health care system.
Today the committee will hear from two individuals, experts in new technologies, to speak as futurists on the potential of these technological advances in robotics, artificial intelligence, 3-D printing and their application to health and health care.
Colleagues, because we have one of our guests on video conference, it will be critical when you ask your question to identify the witness that you want to ask first. Don't simply say, "I'm throwing this out to either of you.'' That doesn't work in this circumstance. If you want both to speak, address it to one of them first and then we will go to the second one. And you will also know that you need to deal with a slight delay in the response. In this case, we've checked it and it's working quite well so there is not much of a delay, but give him a slight chance to respond.
Because technology is not totally without the possibility of difficulties, we will invite our guest, who is appearing as an individual by video conference in the first instance, and so Dr. Bertalan Mesko, who is a medical futurist, I now turn to you to present to us.
Bertalan Mesko, Medical Futurist, as an individual: Thank you, Mr. Chair. It is an honour to get a chance as the medical futurist to share my research on how automation can change health care. I firmly believe that automation, especially artificial intelligence or AI, robotics and 3-D printing are essential in making health care sustainable, because health care today is not sustainable. Costs are rising, patients get diagnosed at a late stage, and the World Health Organization estimates that there is a shortage of about 4 million doctors worldwide.
For 2,000 years, since Hippocrates, all the medical information, studies and technologies have only been accessible within the so-called ivory tower of medicine, but that's not the case anymore. Instead of spending time with the art of improving and saving lives, physicians have dedicated a lot of time to administration and tasks that could be automated. I think they are unable now to dedicate their time to challenges that require the judgment and creativity of the trained human mind.
Technology and the term "digital health'' have been changing the status quo. Now the doctor/patient relationship is an equal level collaboration. Patients want to get empowered, and physicians can access all the information they need, including technologies. All of these are within patients' and physicians' reach.
I think the benefits are clear. This new collaboration between doctors and their patients, powered by disruptive technologies, can change the overall status quo. This can create a system in which it is always cheaper to prevent than treat a disease. And I think automation in this vision is simply inevitable. I will give first a few examples about artificial intelligence.
IBM Watson created an algorithm in the cloud that can find the most evidence-based information and studies for the doctor's use when they check a patient's case. It takes a minute for the algorithm, while it could take decades for a physician to get through that much information.
A second example about AI is making pharmaceuticals faster. Making pharmaceuticals is a very long and very expensive process. It sometimes takes more than a decade and costs billions of dollars. But a company called Atomwise came up with the idea that they could find two drugs using AI algorithms in the cloud to make the Ebola infectivity much smaller. It took them one day, while usually it takes months or years of research to come up with such a result.
A second group of examples is about robotics. The company InTouch Health has been creating tele-medical robots that can provide consultation for patients who are living in remote, underdeveloped areas. In those areas, even automated medical drones could deliver equipment and drugs to locations that are unreachable by local ambulances.
Regarding 3-D printing, a company called Organovo can print out liver tissue that can function like the liver for more than 40 days. According to experts, within one decade, we might be able to print out solid organs — from kidneys, livers and even skin. Imagine the impact this technology could have on the lives of patients who are currently waiting for transplantation. There are hundreds of thousands of them worldwide.
Finally, about 3-D printing, last year the FDA approved the first drug printed by a 3-D printer. It prints out the drug layer by layer to make it dissolve much faster than average pills. It's used in emergency situations. I think it creates the era of personalized medicine where 3-D printed drugs and food could be printed out, tailored to our specific molecular traits and even health conditions.
However, these are only individual trends within technologies. These do not have an impact on patients' lives if the role of governments does not change. I think technology is not the final solution for health care's problems. No technology, no available health record or medical robot or AI algorithm can change our lifestyles. We can change our lifestyle with help from really good, regulated technologies, and smart regulations could facilitate the adoption of such technologies.
One example is from the United Kingdom, where this year the National Health Service, the NHS, will give out free applications and smart devices to millions of patients because they want to help them to get empowered and help them manage chronic conditions much better to save costs on treatments and rehabilitation.
To sum up, I think with automation, we could finally create an ecosystem in health care in which, first, we could utilize the brain power of physicians instead of pushing them towards repetitive tasks that could be automated. We train physicians to use digital health technologies in a way that the human touch and making really hard decisions could remain the core essence of their job. They could prevent, predict and treat diseases like never before — an ecosystem in which first we empower patients to take proactive care about their health and disease; second, with the swarm of available health sensors, we have momentum now that we can finally put their data into their own hands to empower them further; and third, an ecosystem in which we can allow start-ups developing innovative technologies to go through the regulations quickly in order to reach patients. Otherwise, patients will find their own solutions in the jungle of digital health care solutions.
I firmly believe that a government that welcomes innovation in a way that it empowers patients further could change the status quo. Including automation and practice in a way that the most precious part of a doctor/patient relationship — the human touch, the creativity, the human judgment, the empathy — remains the essential part can create a real new health care system.
I think this way, all the relative tasks, big data analysis, epidemiology and the collection of medical information could be taken care of by regulated, efficient and affordable automated solutions by patients and physicians who finally have time to pay attention to each other.
Thank you so much for this opportunity.
The Chair: Thank you very much. Now I'll come back to our in-room witness. It's a pleasure to welcome Abishur Prakash, Geopolitical Futurist with the Center for Innovating the Future. Please proceed.
Abishur Prakash, Geopolitical Futurist, Center for Innovating the Future: Good morning and thank you for inviting me. My name is Abishur Prakash. I work as a geopolitical futurist at a strategy innovation lab located in Toronto. For those of you who are not familiar with this role, I work at the intersection of geopolitics, technology and future. I have also recently published a book called Next Geopolitics.
As a geopolitical futurist, one of my key responsibilities is to look at how countries are adopting new technologies such as robotics, artificial intelligence and 3-D printing. For example, in China, scientists have 3-D printed blood vessels and implanted them in a monkey. This is one of the first steps toward 3-D printing organs and is hope for the 1.8 billion people around the world with heart problems.
In South Korea, researchers have developed a smart contact lens and eyeglass to help people with diabetes. The smart contact lens and eyeglass work together by monitoring diabetes in a patient through their eye and then dispensing drugs. The researchers say that a person could wear this for as long as a month.
In the U.S., a hospital has begun using dozens of TUGs, the name of an autonomous robot. These robots can deliver medicine, change sheets, bring meals and carry away dirty dishes. As of 2015, these robots represented the largest fleet of medical robots in operation around the world.
In Finland, municipalities and hospitals are looking at applying artificial intelligence to health care. In Finland, all health care and patient-related data has been digitized for the past 30 years. One hospital district, which spans 24 municipalities, wants to use artificial intelligence to predict health issues.
In the U.K., a start-up has developed an AI-chatbot to help people with health questions and diagnosis. It uses data, which they call a "medical map,'' that allows a person to chat with the bot, share their symptoms and receive diagnosis and further questions. The bot itself is constantly learning from its interactions.
That's a look at the present state of the union when it comes to health care in robotics, artificial intelligence and 3D printing, but there are also trends emerging. The first trend is called smartphone doctors. While many of us already use apps and wearables to track our health and fitness, robotics and AI will take this to the next level.
In the future, our smartphones will help us diagnose diseases. This means we will not have to depend on hospitals and clinics to find out what is going on. For example, using a smartphone, people will be able to take pictures of a rash or skin marking, and AI will tell them if it's a disease or not. This will impact health card usage across Canada. Technology is moving toward a point where smartphones can talk with biosensors attached to a patient and monitor them. Remote monitoring will allow more people, for less money, to be monitored at home.
Can this go one step further? In the future, if multiple people are sneezing or coughing in a neighbourhood, AI on phones can identify this as an emerging epidemic, and then this system can inform local hospitals, clinics and pharmacies.
The second trend is called predictive health care. As most of you already know, heart attacks are the second leading cause of death in Canada. Nearly 600,000 Canadians live with some form of heart disease. It is estimated that 50,000 new cases are diagnosed each year. I don't have to go into any particular detail about the impact of this on ambulances, surgeries, medication, lifestyle changes and more. So what can robotics do here?
Researchers in the United Kingdom have developed an artificial intelligence that can predict when a person with heart failure will die. It has an 80 per cent accuracy compared with a human doctor's accuracy of 60 per cent. If AI can predict when a person with heart failure will die, preventative health care becomes extremely important. It can also lead to new ways to prioritize wait lists for organ donors, heart donors, and it could lead to faster development of 3-D printed hearts.
AI will also be able to predict other health conditions. A technology company is saying that AI will be able to predict mental conditions like depression, Alzheimer's and more within five years.
The third trend is called automated hospital. Around the world, robots are being deployed in hospitals. In Toronto, a robot is mixing chemotherapy medicines. In the United Arab Emirates, a hospital will begin using the world's first robot pharmacist, capable of dispensing 12 medications in under one minute. In one surgery, a robot called STAR out- performed human surgeons when it operated on a pig. In the U.S, an algorithm has been developed to diagnose skin cancer, while a hospital is using robots to fight germs. An automated hospital means existing hospitals can become more efficient, newer hospitals can be created at much less cost and hospitals can be created and designed around certain types of demographics or diseases.
The fourth and final trend is called automation blowback. The Canadian health care system needs to begin preparing for blowback from automation. Two studies are showing that 42 per cent of Canadian jobs are at risk of being automated, or 7.5 million jobs. As that happens, it will change the design and delivery of the health care system. Will people become more depressed? Will they stop working out and eating healthy? Will they drink more alcohol, consume illicit drugs or eat fattier foods? Will they engage in unprotected sex? And, where will the revenue for health care come from if people have no job and are not paying taxes?
Self-driving cars present an equally important challenge. Self-driving cars are expected to dramatically reduce accidents, since at least 90 per cent of accidents involve human error. However, one-in-five organ donations come from victims of car accidents. If car accidents decline, then organ donations also decline. This is where 3-D printing will play a huge role. It will force people, if they don't have access to organs, to stay in hospitals longer, try alternative treatments or stay on wait lists.
Those are my initial opening remarks on the present state of the union and what will take place in the future. I look forward to discussing this more over the next hour. Thank you.
The Chair: Thank you both very much. I want to remind my colleagues to identify the initial responder to your question, given we have a distant connection. I will also remind you that I will monitor time because I want to make sure every senator has an opportunity to get involved in the discussion. They have certainly given us lots to discuss and move forward on. I will start with Senator Eggleton.
Senator Eggleton: In accordance with your request, I will start with Mr. Mesko, who is coming to us from Hungary today. I would like follow-up comments, though, because the question could apply to both of them.
This is a massive array. It's hard to digest all of the things you are both talking about as possibilities, but we have to determine how they get into the system and how government can assist getting them into the system. You have to start by perhaps overcoming some skepticism on the part of some health care professionals. When you bring on something new, you have to convince them that this is something that really is worthwhile doing, but hand in hand with that is the economics of it all, the cost-benefit of many of these technologies that you are talking about.
How would you suggest we make that kind of transition to see these ideas tested in the marketplace, so to speak — in other words, within the health care system and hospitals and other components of the health care system?
Mr. Mesko: Thank you for raising this issue. This is a struggle. I've been teaching medical students for more than a decade now. When you come up with such ideas about how technologies could remove jobs from health care or even change the role of medical professionals, physicians become quite resistant. But when you show them how these things work in practice, and when they understand that technology is not here to replace them but, rather, to inspire and assist them in many ways, then they understand and can say, "I just need to learn how to work with technologies, but they won't replace me. My job will be to step on the shoulder of technology to look further away.''
When physicians become resistant when I talk to them about these new solutions, I ask them: Do you enjoy what you are doing right now, every single day? Are you really doing why you became a physician for many years ago? Because most of the time you spend is on administration and making phone calls. You don't have time for your patient. On average, the time we have for a patient is about three and half minutes. That's nonsense. It's not enough for anything, even just measuring the blood pressure and nothing else. I believe the time spent on repetitive tasks could be automated, and then the rest of the time, the precious time, is time we could spend with patients.
To answer your question, we need to train physicians for digital health because they have never been trained to deal with any kind of technology. By the time they get out of medical school, the technological landscape could change entirely. In five or seven years, smartphones came out from nowhere, virtual reality became a possibility and social media became huge, with billions of users. We cannot tell them which technology they will have to use, but based on my experience, by teaching this to medical students, we can give them skills that they will use later to make their own assumptions about which technology might be useful for their own practices and which should not be used at all.
I always wear two caps, one being a medical doctor. I truly and firmly believe in the power of the human touch and the relationship with the patient based on empathy. But being a futurist, I also firmly believe that we need to include technologies. The real challenge we face here is not whether one replaces the other or whether one is better than the other, but how to find the balance by preparing medical professionals and patients for this technological world, and also how to hold back the weird, sometimes crazy ideas technologies tell about the future, to find the balance so that we can still keep the human touch, because I do believe that's the essence of providing care.
Mr. Prakash: Senator, your question had two components. The first was essentially how do we regulate and integrate these technologies into the Canadian economy. The second component was how we afford it.
To answer the first question, I believe that Canada needs to have an official robotic strategy. Japan has one; it's called the New Robot Strategy. China has one; it's called Made in China 2025. But these are more geared toward manufacturing, not health care.
Nonetheless, Canada does not have an official robotic strategy. Right now, it has grants. Even at the provincial level, there are no provincial robotic strategies. So the first way to integrate robots into the Canadian economy — in terms of not only do they profit and help companies, not only do they help patients, but are they safe? — is to have some kind of regulatory framework.
The answer to your second component is that in terms of cost, right now health care robots do cost quite a bit. However, there are changes taking place within the robotics industry. One of the main changes is called robotics as a service''
Let's say we are all clinics or hospitals and we want to own a robot. We have to spend $250,000 or $300,000 on a robot or robots upfront. Robotics as a service adopts a Netflix-type model. Instead of owning a robot, you pay for it every month. There is a company out of China that has just launched a robot. For the U.S. market, they will be adopting this robotics as a service model, and one of the applications for their robot is health care.
Yes, there are upfront costs. As of now, 99 per cent of robotic applications in the health care service are not robotics as a service; they are standard "pay up front and you will get your robot.'' But there are changes, and I believe we will see more robotics as a service in the coming years.
Senator Stewart Olsen: Mr. Prakash, thank you very much for your presentation. When I told you I didn't have a clue here, you were very good about explaining. I really don't understand this, and I suspect most of the people who may be watching don't understand. Can you tell me how 3-D printing actually creates a vein or a hip that can be implanted? I don't understand that.
Mr. Prakash: The best person to ask would probably be an engineer. I'm not an engineer, and I don't want to give the wrong answer in terms of the technicalities. However, when it comes to 3D-printed organs, I do know that there are several challenges.
The first is that people say that we can 3-D print livers, kidneys or lungs, and it all sounds great. But one of the hardest parts about 3-D printing an organ is 3-D printing the actual cells of that organ and having them stay alive after it has been printed. In most cases, when you 3-D print an organ, the organ dies immediately or shortly after because the cells cannot stay alive. We are not able to reverse-engineer that. In South Korea, researchers have 3-D printed liver cells from mice, and they were able to keep it alive for 30 days. So that's the first issue.
The second issue is really its cost. Right now, if you are 3-D printing a pencil, you can buy it for your home. But when it comes to health care 3D printing, the cost is really high, especially in the academic world. That's why a lot of 3D-printed innovations from the University of North Carolina and from other universities are all academic based, because the schools are funding it, or private institutions are doing it, but individuals are not able to do this in their garage, and hospitals are unable to do that without serious government funding. I would say it's the cells part and the cost part that are the limitations of 3-D printing.
The Chair: Dr. Mesko may be able to answer your question in a little more detail. I will turn to Dr. Mesko and ask if he could take an example, perhaps a backbone. Pieces of backbone have been 3D printed. I think our committee would benefit from having a sense of the mixture that goes into the printer and how it is layered down to create a product.
Mr. Mesko: Absolutely. Actually, today we can only talk about 3-D printing tissues, but not organs. We are still quite far away from that.
With regard to how the actual 3-D printing of tissue works, we don't print out cells, obviously. That would be far away in the future. What these companies do is use the cells as the ink in the printer, the living cells. However, the real challenge with creating organs is their structure. These are three-dimensional structures. Cells have connections from many directions. They communicate with each other, and creating this three-dimensional structure is a huge challenge. So what these companies do is use scaffolds, very special scaffolds, made organ by organ for different purposes. They actually include those cells into the three-dimensional scaffolds in a way that the cells can keep on growing.
The reason one of the first tissues that was successfully printed out was the liver is that it has a hexagonal structure, but in three dimensions. The company called Organovo, which I talked about before, was able to make this new scaffold system into which they could inject the living liver cells that they programmed like they do in normal molecular biology labs. The scaffold they used was able to keep the cells alive. The tissue itself, the three-dimensional structure of cells, was able to function like a human liver for 40 days. That is why the company is very sure that they could come up with an actual functioning tissue that could be transplanted into people in about five to ten years.
The FDA might approve the tissue, not the organ, just liver tissue in about two or three years.
The Chair: Dr. Mesko, I wonder if I could bring you back a little bit. I think the committee would like to get a physical sense of the actual printing of an organ. That's why I asked you if you would consider using the printing of a section of backbone. I'm aware that is something that has occurred recently, and I think the components are simple enough in this context that you might be able to give us a sense of how it's layered in order to construct. In this case, it turned out to be a flexible backbone section.
Mr. Mesko: I'm aware that tissue has been printed out, but not all together the backbone itself. Unfortunately, we are again still quite far away from that. But in the printer, they use the living cells. If we take the traditional example of an actual printer that they use every day, the ink is inside the printer. They know exactly what they want to print out, and they use the paper as a scaffold. They design that they want to print out is in the software, and then by using the ink in the printer they print it out on the paper.
In the case of organ printing, which is again just tissue printing, the ink is living cells, and the scaffold is a very special structure that the company produces. The software is the one they use to create a three-dimensional model that knows the structure of the scaffold, and the living cells can be injected into that. Again, the ink is the living cell.
The Chair: Thank you very much. I think we'll leave it at this point.
Senator Seidman: I thank you both. I think I'd like to take even one step or two steps backwards because I think all of us and our audience may be thinking this is sort of science fiction. I will ask you about a couple of concepts.
One is futurist. You both call yourselves futurists, but I think most of us don't know what a futurist is. Do you have a theoretical framework, or do you sit and dream in an armchair about what might happen over the next 50 years? What is a futurist?
Mr. Mesko: I wish I could. Thank you so much for the question. This kind of job sounds amazing.
I'm the medical futurist, which means that I constantly learn about methods, including creating scenarios for the future and how to create prediction models. Obviously, it's impossible to predict the future. Also, obviously, what you cannot do is tell someone that there are so many futures they could but this one is going to be the best one for you. These things are utterly impossible.
Where I think futurists can help is that we are constantly focused on predicting different scenarios about the future, regarding costs and economics, technological advancements and cultural changes, and we try to find those scenarios that might take many of these perspectives into consideration. By sharing those scenarios with people worldwide, including technologists and decision-makers, even patients and physicians, we can keep building those models. When we see that reality and that science fiction is becoming something like one of those scenarios, then we have an already defined framework of visions, thoughts, regulations and ideas that could be implemented into the practical reality of today right now.
I hope that helps. We don't predict science fiction; we create bridges between what might become possible tomorrow and what's happening today.
Mr. Prakash: That's a very good question. You're really putting me on the spot here. As a futurist, what do I do?
My job is to help companies become future ready. I mainly work in the private sector. I help large companies, Fortune 500 companies to start-ups, become future ready and understand how their industry is changing, what kind of variables are emerging, what their industry will look like two, five, seven years from now, so that they can actively prepare and create a strategy and implement that strategy and remain relevant.
As a geopolitical futurist, my main focus is to understand how new technologies, like robotics and AI, will transform world affairs. How will these different technologies affect the foreign policy of countries? But with that focus, I'm very aware of what's going on in robotics and AI and 3-D printing and food cloning and embryo editing and all these different things we hear about. I think that's why I was called here, to give my two cents on how robotics and AI will affect health care.
More importantly, I think in today's world, everybody is calling themselves a futurist. Personally, I don't make predictions. I believe that you can't make predictions in today's world because things are moving so fast. There are so many unknown variables. What I'm talking about today will be old six months from now. What I'm talking about today is pretty incredible when it comes to AI and robotics and health care. Within six months, this will be old data. So I don't make predictions. My job is to connect the dots. My job is to make scenarios of what could happen and help companies understand their future so that they can become future ready.
Senator Seidman: You are saying what you come up with is old within six months. But then how do you, in an ongoing way, make sure that your advice-giving is not old before you give it? I'm trying to understand what kind of theoretical framework you use to be a futurist. I do seriously mean that.
Mr. Prakash: Sure. I have a retainer where I'm constantly advising companies on what's taking place. One of the things that I tell companies is that — it's a very simple expression — "Either you can become Uber or you can get Ubered.'' You have to prepare for what's coming. If you don't prepare for what's coming, then you're going to be left on the side and watching your competition basically destroy you.
There are ongoing workshops. I provide actionable intelligence workshops. I sat down with a major Canadian bank last week and ran a session with them. What you point out is a very important question. It is one thing if I talk about this today, but how do I keep you all informed six months from now? The easy answer is: Hire me. Bring me on as a consultant. That's the easy answer.
The second answer is that in today's world, I think we all have to become future ready. That puts a bit of responsibility on each of our shoulders to start researching what's going on, start paying attention to the headlines a bit more, dig deep, ask questions and email people. I do research every day, for hours, just to stay on top of what's taking place.
Senator Seidman: Now you've answered my question. Actually, that's the question I asked, is how do you do it.
The Chair: That was the answer we were waiting for.
Senator Meredith: I was moved by both of your presentations. I'm fascinated by your brain retention in terms of it. If we could actually share a percentage of that, I would take a percentage.
The fact of the matter is that you talk about moving technology, and we see the evolution of technology and how quickly it's changing. Then you talk about China having a strategy. What would you say to our government here in Canada with respect to innovation? We know that Canada is lagging behind. Some statistics show us as being ninth in the world; others show us as being fifteenth in terms of innovation and science and technology. How do you convince government that we need to be on this path?
I will cite a case just recently in my city of a gentleman who took sick at work and was rushed to the hospital. He sat in a waiting room for four hours and eventually died in the waiting room. This was just two weeks ago. Talk about utilizing the technology you mentioned to assess this patient quickly so that doctors can do the proper treatment. We know there's a line-up of different symptoms and diagnoses and so forth, but how do we move expeditiously to avoid tragedies like those from happening? This man has left two children behind, a wife and so forth. I am moved by that.
It's about time that we say we need to stop talking about it and do something about it. How do we get there, and how do we convince government that we need to be there now?
Mr. Prakash: There are two answers to your question. The first answer is that developments are already taking place to solve the problem that this unfortunate gentleman faced in the hospital.
At the University of Pennsylvania, artificial intelligence has been created to predict when people will visit the ER room. Right now it's focused on oncology, tumours, but they want to be able to predict so that surgeons know X patient is coming on Wednesday and they have everything ready for that patient instead of being caught off guard that there's a patient here and they have 10 different things to get through. They will use all kinds of data sets, radiology reports, patient reports, et cetera.
AI is also capable. For example, there's a university that fed IBM Watson about 1,000 cases. IBM Watson, 99 per cent of the time, prescribed the same treatment as human doctors. In 30 per cent of the cases, it went beyond and it showed treatment that human doctors had missed.
Right now, AI is still in its infancy, I believe. We're going to see it become more advanced in the coming years. Even right now, the developments allow a more efficient health care system to exist, one that is cheaper, more cost effective from a taxpayer's point of view, and one that doesn't result in what this gentleman went through. Of course, that brings its own challenges, which I can get to in a separate Q&A.
Second, Japan has something called a new robot strategy. They basically want to become the world's leading robotic power. China has Making China 2025, and they want to become the world's leading robotics power. Canada doesn't have anything. This is not just a health care issue; this is regarding Canada's relevance in the world. We can't just depend on oil. We have to also depend on robotics and artificial intelligence.
What I would advise to the Canadian Senate, to Prime Minister Trudeau and his cabinet, is that a comprehensive robotics strategy needs to be created that takes into account the economy and economic diversification; that takes into account the social design, the effect automation will have on our societies; that takes into account the effect on the health care system and foreign policy and all these different things. It's not enough to simply have a health care-related robotics plan while the rest of the country is still backwards.
Right now, provinces are operating on their own. The province of Saskatchewan has 11 robots in operation. They have something called doctor in a box. If you're a paramedic in Saskatchewan, they have a doctor in a box, and if a paramedic is transporting a patient, they can call a doctor. The doctor will advise the paramedics on what to do with the patient in the ambulance while the ambulance is going to the hospital.
Saskatchewan is also trialing a robot called Rosy on a reserve. First Nation people on reserve might not have access to hospitals, so Rosy sits there and they can talk to a doctor wirelessly through telepresence.
So there are two answers. I hope I've answered both of them or one of them at least.
Senator Merchant: I'm going to bring you a few steps back. I am going to be a patient. I'm from Regina, Saskatchewan. Thank you for putting a good word in for our province. Yesterday I highlighted some of the other things that we have accomplished in Saskatchewan.
Dr. Mesko, you have said several times that the outcomes on my behalf as a patient will be so much better. Will I get better health results as a result of all these new technologies and this futuristic view of health care?
Secondly, I'm trying to understand how this would become cheaper. Would it become cheaper for governments? Would it become cheaper for me? Will I be able to get my medication for less? Today I may have a good diagnosis, but I can't afford to get the medication. Some medication is beyond my reach.
From a patient point of view, how am I going to benefit? I live in Canada, not in Korea or in China or in the U.S. We have our own health system. How will I benefit from all these new technologies?
Mr. Mesko: Those are two amazing questions. I will try to be short.
First, I think if a country starts focusing on a certain technology, it will yield no results for the health care system. I do believe that what we are witnessing today is not a technological revolution but a cultural one. Technology will keep on developing at its own rate no matter what we do. What matters the most is how we adopt it or reject it. We all will become patients eventually. How do we react to technological changes? The real challenge here is preparing patients and physicians for a technological world.
You're asking me if using these technologies will make my care or my life better. The only thing I can tell you is — and I might be harsh here — all lives today depend on pure luck. Variables include if you recognize a symptom in time, if you get to the right doctor in time who is just perfectly up to date about the 27 million medical studies currently available, if you get the right diagnostic procedures, the right treatment, and you comply with treatment. There are so many variables in this that we need to state out loud that our lives depend on pure luck.
I don't want my life to depend on luck. If I have a disease, I want to catch them as soon as possible. Without data about my own health, vital signs, health parameters, and genetic sequences — which now anyone can get from their couch sitting at home, ordering things online — it's impossible to make really smart decisions. Now we can only base our decisions on a few data points that we can measure or get at the doctor's office, not even at home.
A new form of health sensors has become available in the last couple of years. These sensors now are making this possible because people started measuring things about their own health. They want to manage disease in a smarter way so they use smartphones and digital health solutions.
My point here is that if a country does not have a really correct digital health strategy that takes patients, physicians, decision-makers, researchers and developers into consideration, then no technological help will make health care more affordable, more efficient or more accessible. We need to focus on the cultural aspects of developing technologies.
Your second question is the ultimate question about digital health — how it's going to be cheaper than what we have today. First, as I mentioned in my report, it's always much cheaper to prevent a disease than to treat one. One round of chemotherapy or the procedures used for diagnosing cancers are unbelievably expensive procedures and treatments. But preventing and predicting something when there are only a few variables changing in your body is invaluable. You could do so many things to either catch the disease at an early stage or prevent that from happening. The costs are not even comparable.
To give you a practical example about how disruption can make health care cheaper, I saw an example coming out from a U.S. company. They wanted to help nurses. When the nurses take the blood samples, it's really hard to find a vein. I think we've all been in such situations. The company made a vein scanner that shows the veins on the skin of the patient in seconds. It's really mind-blowing seeing the example in practice. It costs about $5,000 or $6,000 to create such a device. Last year, two medical students created a smartphone application free for everyone that can do the same just by looking through the camera of your smartphone. You can see the veins just like you can see the veins through the vein scanner. It used to be $6,000; now it's free. Basically, that's the kind of evolution that we see in the price. If a technology is not affordable, it's not enough. That's how we are pushing companies forward.
Mr. Prakash: I don't necessarily agree that it's all luck. I believe it was luck until now. Let's not forget that the iPhone is only 10 years old. It came out in 2007. A lot has changed in 10 years.
Right now, as we speak, a Scandinavian country has passed legislation that elderly homes or elderly hospitals must use robots. There's a type of robot called a social robot, and it can hold conversations and whatnot. That country is doing that right now.
In 2012, in the U.S., doctors 3D printed a windpipe and put it in a baby to help it breathe. That's happening right now, or many years ago as we speak. It's not luck anymore. It's a question of whether countries can adapt to what is taking place. If they can't adapt, if they can't integrate, if they can't align, then they remain backward. But if they can or if they can be the ones innovating, then we're going to see some great things happening.
In terms of benefit, I think the first benefit and the main benefit comes with diagnosis. Artificial intelligence is capable of diagnosing things in a way we haven't seen before. One study showed that artificial intelligence can detect, through a person's breath, 17 different diseases. I mentioned before the idea of using a smartphone to take a picture of a rash and AI will be able to tell you if it's a disease or not. That alone will take a huge load off of the health care system. People won't be visiting doctors as often.
In terms of cost of that service, right now we're thinking in terms of the Canadian health care system, in terms of those parameters, and why are we doing that? Because until now, when we get sick, we go to a human doctor in Canada. We go to a hospital maintained by humans. We go to a human pharmacist. If there is a company in the U.S. that charges you $20 a month to diagnose you through AI, that becomes a much cheaper option for you than maybe going to a health care system that is costing you too much time and that isn't doing what you want. When you go, the doctor will say, "It might be this; it might be that. Can you take some more tests?'' et cetera. In terms of the actual cost of medicine, it's a great point. The truth is there is no answer right now.
The Japanese want to use artificial intelligence to look at a patient who has cancer, take in all the different data points of that patient — how advanced the cancer is, what type of cancer it is — and have AI tell them what kind of treatment to prescribe.
It's about diagnosis, and in that sense I think that will relieve a huge burden off of the health care system.
The Chair: I want to clarify that I think that Dr. Mesko was talking about the reality of an individual in today's world, and I would say he's right in that regard. I think what you're saying is there is the potential to remove the luck, if in fact the devices become available to those who may need them at a given time.
Mr. Prakash: True.
The Chair: We won't go down that any further. I think we've understood and you've illustrated the various kinds of things that will occur.
Senator Mégie: I have two questions.
[Translation]
Do all those engineers have an ethics committee to advise them? To prevent Alzheimer's disease, for instance, Motorola came up with a little device 20 years ago that could show, with a single drop of blood, whether a person was at risk of developing Alzheimer's based on the fat level in their blood. Genetically, there are three categories. The ethics committee asked it to stop because finding out at the age of 40 that you have a 75 per cent risk of developing Alzheimer's could ruin a person's life. What happened? Progress is still being made, but Motorola did not distribute its device.
On all of these fascinating topics, do you have an ethics committee that advises you on just how far you can go?
[English]
Mr. Prakash: Do you mean as a futurist, do I have my own ethics committee?
Senator Mégie: Not you but your company. Do you have an ethics committee?
The Chair: We know what your situation is and I know where you're going with that. I think what she'd like to know is if there is an ethical environment around the application and access to these new technologies.
Mr. Prakash: That's a very important question, and the answer is no. Ethics is only emerging now. I was at CES earlier this month sitting in a robotics panel and was asked a question by my colleague. This has nothing to do with health care, but it will give you an understanding of ethics. In terms of programming a robot, it will have inherent biases. It will have inherent biases as to all kinds of different factors, and that means that the burden of responsibility lies with engineers and programmers, but who gives them the right?
When it comes to ethics of a robot or ethics of artificial intelligence, it's a very important question, and there's a huge void in the industry right now because nobody has the right answer. More importantly, every country will have a different approach.
I hope that kind of answers your question. I personally try to remain as neutral as I can in my work, but everyone has biases.
The Chair: Before we go to Dr. Mesko, I want to give you some information with regard to our plan. We have, obviously, searched for ethicists in this particular area. There are none identified, just as Mr. Prakash has indicated, but there are international persons operating in frameworks within which that is an issue, and we hope that they may be able to bring some more direct answer to your question. Now would you like to go to Dr. Mesko?
Senator Mégie: Yes, please.
[Translation]
I would like to know if there is a research ethics committee or a clinical medicine ethics committee that advises the team of futurists.
[English]
Mr. Mesko: Absolutely there is. There are many situations. The Future of Humanity Institute has bioethicists working hard to help developers, decision-makers and futurists in making the right decisions.
Also, every medical or clinical research that is done today needs to go through very serious ethical considerations and committees. Ethics plays a really important part today in making decisions about the future. Moreover, Elon Musk and similar technologic visionaries have opened up an OpenAI institute because they want to make sure that while we are developing artificial intelligence, we won't fall into the issue or fall into the mistake of creating something much bigger than what we could even understand. They want to make sure that we are developing AI in a box and that we are keeping control of what is inside that box.
Senator Mégie: Thank you.
Senator Dean: Thank you to our presenters. I will follow up on Senator Mégie's very important question on ethics, but I'll ask a question from a different perspective.
You describe a future where there will be a vastly more distributed approach to the provision of health care, one that inevitably and obviously will foresee a much broader role for the private sector and a private sector that increasingly crosses international boundaries.
I'm interested in the implications of that world for equity and access to health care. On the one hand, we might imagine that a personalized approach to robotics and home care would be more democratic than ever before. It might equally be the case that these new technologies and availability of print-to-order organs is available to those who can afford it.
As futurists, as you think about the future role of governments and government regulation, beyond advancing commercialization, what is the role of governments in protecting access and recognizing and protecting health equity? Do you have any thoughts on those? You talked about design and social impact, Mr. Prakash. Perhaps we could start with you.
Mr. Prakash: One of the trends identified is called cross-border health care. When the Saskatchewan robots were being used as a trial, one the challenges raised was that right now the provinces handles health care and the federal government handles health care, so it's compartmentalized. When happens when you get rid of the borders or you go cross-border? How will the government maintain charging for health care, finding revenue sources, et cetera?
Like I said before, I think we will see in the future some kind of service that allows people to tap AI to diagnose various issues. Obviously not everything can be diagnosed with AI. Right now IBM is leading a lot of these developments, and right now, as an individual, IBM Watson is free for you to use. You can access their supercomputer for no charge, with certain applications.
In the future, if IBM or Medtronic or a different company wants to a launch a ten-dollar or twenty-dollar solution, or if you feel because of robotics that spending $500 for an operation in China, where a robot will be the surgeon, is far more effective than going to a Canadian hospital, then that changes the role of government.
What do I think the role of government will be? In a country like Canada, it will be maintaining an equilibrium between the public health care system and the private sector advances and finding a way to integrate these advances into Canada's health care system, because this is a new world. We're not only looking at humans anymore. When you have AI being offered from another country, when you have robots as a service being offered, meaning surgeries that are a lot cheaper in other countries, then people may opt in for going to other countries versus accessing the Canadian health care system.
Once again, when it comes to ethics, you raise an important point, just as your colleague did. There is no real way to answer that because there is a void. There may be groups working on it, and there are groups, as my colleague pointed out. However, every country will have a different approach and every province may have a different approach, and so we have to take that into account and realize we're going to have conflict and tension. It could be geopolitical, but let's keep it limited to health care right now.
Mr. Mesko: I think the role of a government in improving health care today is creating the regulatory framework that welcomes innovation but still keeps services safe. I'm saying that because it's possible that I can send out a tissue sample to a start-up in Belgium and they make the DNA of the tissue sample available and I use an algorithm in the U.S. to find the best treatments possible in the world based on my tissue sample, and a start-up company in Spain can find the clinical trials currently being open to that kind of DNA mutation in my sample. First, the world has overcome something without real boundaries, and I haven't even met my health care system and still I found the best potential treatment for my individual problem.
Second, when we use disruptive technologies in health care, then it becomes quite evident that patients will not wait for you to create regulatory frameworks. One practical example: About five years ago, it became clear that for patients with diabetes, it's possible to create an artificial pancreas system, which is not an artificial organ obviously but a sensor on the skin that can measure the blood glucose levels, and a second sensor can add insulin when it's needed, and an algorithm can make decisions about that. It's an artificial pancreatic system.
However, the FDA in the U.S. didn't approve the first such device until last summer. So for the last five years, patients had been, in a do-it-yourself way, creating their own artificial pancreatic systems at home, and that can create chaos. But of course they knew there was a technological solution for their health problem, and either they wait for the country to create the framework in which this might be approved or they just use a smartphone and social media to find a solution. Nine out of ten people would choose the latter one and it's understandable.
Our responsibility today is to welcome innovation but still keep products safe.
Mr. Prakash: I want to add a few more points, if I could. This might be a scary scenario, but these are disruptive technologies, and we may have to radically rethink the role of government too. In relation to 3-D printing of medicines at home, I was reading about an awesome robot that attaches to your ceiling and it becomes your chef. It has arms and it can move food and clean and whatnot. What if tomorrow it can also do surgery through an update? When I say tomorrow, I don't mean 12 months from now, but you get where I'm going.
We may need to radically redefine the role of government in a way we haven't done before, and that might mean, as my colleague said, the only role is regulations. If you have an automated hospital where the only human beings are the patients, how do you oversee that and maintain that? Public policy becomes the only route.
Senator Raine: Thank you very much. This is absolutely fascinating.
I have two thoughts going through my mind. One is that public policy cannot get ahead of public opinion. We're privileged today to have you two educating us, and through CPAC hopefully we have some viewers. However, I'm wondering if we should be encouraging our national broadcaster to have a one-hour special on these kinds of future developments and the questions raised. We need to educate not just the doctors but also the patients and everybody. I would like your thoughts on that, both of you. Do you think it would be a valuable exercise? I'm thinking we have stuff on our television that's just junk filling the air, and yet we have the ability with a public broadcaster to put something out there that's valuable in educating people. I'd like your comments on that.
Secondly, I want to come back to reality today. Today you have a patient who goes in and has a biopsy, and the biopsy is now looked at by a pathologist. Are there hospitals in our country where that pathologist has access to IBM Watson or to artificial intelligence to see the best outcome for the pathology?
Mr. Prakash: For your first question, I think absolutely. There definitely needs to be a bigger push by the government to educate people about the future and what it will mean for them. I would go further than a TV program. I would say we need to integrate this into the K to 12 curriculums.
My colleague's focus is co-existence with robots. As human beings, as kids and teenagers graduate from high school or university, unlike us, we all walked in, we're all human beings. Ten or fifteen years from now, there may be robots here. We don't know. So they need to be aware of what's taking place in society and how it will transform society. And it gives them a competitive advantage at the end of the day in terms of finding a job.
Could you repeat your second question?
Senator Raine: In today's scenario with a biopsy seen by a pathologist, does he have access to AI?
Mr. Prakash: A lot of these developments that I have spoken about are still limited to the labs within universities. They are still report-based; they are still experimental-based. However, there is a report from Harvard that says the work of pathologists and I believe radiologists could be automated within 15 to 20 years. It's coming. There are no hospitals right now that I know of that are actively using artificial intelligence instead of human beings, but if they are, humans are still overlooking it and still agreeing that things can be done this way. I hope that answers your question.
Mr. Mesko: Thank you so much for the first question. That's a wonderful question. No technology can change health care if, on the level of the society, we don't have an idea about what might happen next. I think a tsunami of technological change is coming toward us, and if that wave hits us now and washes away the medical system we have been building for thousands of years, it will leave us with an emotionless business with no human interaction. We are social beings, so we need interaction. The more we prepare today, the better we can use technologies properly to make it a real cultural revolution, not just a technological one.
Regarding your second question, they don't have access to that. IBM Watson is currently being rolled out to oncology centres only in the U.S. As it is still very expensive to be purchased by one particular institution, it is only available on the level of one country such as in the case of the U.S.
Moreover, even IBM Watson is not artificial intelligence; it is narrow artificial intelligence. We are still years or maybe a decade away from true artificial intelligence, just to make it clear.
The Chair: Thank you very much.
Senator Neufeld: There are questions around adaptation to the things that you folks are talking about with government. Our health care system in Canada is very much centred on government. From the federal perspective, for provinces, you can't step out of line or you are going to get into trouble with your funding.
My sense throughout my life is that governments are some of the slowest ones to adapt to anything new, especially things that you are talking about. I think places that don't have a government health care system will advance much quicker than we may in Canada simply because of that. I'd like to get your feeling on that a bit.
Saskatchewan is obviously doing some things. Are other provinces doing anything, or have you gone to the medical folks with the federal government or the provinces and talked about the things that you talked about to us today? I'm sure you have, but I'm not familiar with it. Are governments actually looking at this? Are they listening to you? Are they even interested, or is this new to them, as it is obviously new to me — maybe not some of the people around the table here.
Mr. Prakash: I think this builds on your colleagues' point, which is what role is government going to play. Right now, the health care system revolves around the government, but we really have to think about why is that? It's because if you need to do something, right now you have to go within the health care parameters to access that thing, be it a doctor, a pharmacist, a surgery, whatever it is — that is, if you don't want to pay. If you want to pay somewhere else, you can.
If you have a Netflix-type service offer to consumers from China, from the U.S., from South Korea, then it's really a question of how does the government catch up to what consumers are doing? Right now, the government can limit what consumers can do. If the government were to outlaw certain medications, consumers couldn't get access legally. Tomorrow, however, if a patient has access to AI from the U.S. that can diagnose what's going on via images, a biosensor or something else, that changes the role of government completely.
Senator Neufeld: Have you presented to Health Canada? Have you presented these ideas to any other province in Canada? Is anybody in government listening to these kinds of things from your perspective? That's my question. If they are not, it worries me.
Mr. Prakash: I have not personally presented to any of the organizations you mentioned so I can't comment on their readiness.
Your question was, if governments control health care, will they be slower to integrate these technologies than government that don't. My response is that governments who do not have a strategy for robotics in general, regardless of their health care system, will be slower to integrate robotics than governments that do. In China, by 2020, the elderly population will account for 30 per cent. They are 12 per cent now. They are investing heavily in medical robots to deal with that.
Statistics Canada indicated in 2015 that there were more people over age 65 in Canada than under age 15. There are more elderly people today than there are kids in Canada. But does the government have a strategy for elderly robotics? No, it does not.
Mr. Mesko: Regarding your question about private health insurance or private health care versus government- backed health care, I think if we provide a patient with the most modern treatment options, technologies, virtual reality, artificial intelligence today without professional help, it will not solve their problems. Even with the best technological solutions, we need the expertise, the vision and the perspectives of health care professionals. If you have a good basic health care system like you do in Canada, then adding these technologies is an added benefit but it will not replace anything already in the system.
I'm not saying this because I'm sitting in front of you today, but if there is the highest chance winning the digital health race, it might be in Canada because of the good health system you have already and the chances you have to include disruption into it. If you don't do that in time, with a few brave steps forward, again your patients will find their own way. They won't be waiting for you to come up with good regulations in 15 years because they might be dead in 15 years. They will be constantly looking for options about their health online and will crowd-source information through social media. They will crowd-fund the financial amounts needed for their treatment to become a reality. They will keep on looking for solutions until they find one because today, with digital health, there are no physical boundaries.
The Chair: Senator, I think you have put your finger on something that notes the value of the study we are doing. In fact, we are futurists in this sense, in our own way.
Senator Petitclerc: Thank you for being here. I have a question that is going to take us in a different direction. I have been amazed by everything you have been saying and all the images that you put in our heads for the future — amazed and a little scared, I have to admit.
To me, the performance benefits that we can get from that are very clear. I see that, especially with the examples you gave. Maybe Mr. Prakash could answer first. You had an example of one hospital with only robots and no humans and the kind of interaction for the elderly and things like that. Are we getting ready? Are there study groups, psychological studies or evaluations on what will be the social and psychological impact or backlash of those kinds of technologies?
Mr. Prakash: That's a great question. I'm sure there are groups out there doing this, but I have not come across any myself.
You raised an extremely important point, namely, when it comes to automation of the workforce, there will be backlash. And when it comes to automation of doctors or nurses? In Belgium, a hospital has begun using Pepper, a Japanese robot from a company called SoftBank Robotics. It's very intelligent. They've started using Pepper as a receptionist. You walk into the hospital and see a robot there, and it can tell you where you need to go. It can also lead you to where you need to go. It can identify you based on your age. In terms of these automated hospitals, it's slowly coming. A fully automated hospital is not close. It's not in five or ten years but still fifteen or twenty years away.
I have not come across studies that look at the social impact, the economic impact or the regulatory requirements. I have a data point here that over the last 14 years there have been 144 deaths and over 1,000 injuries because of robotic surgeries in the U.S., so there needs to be safety regulations, too, when people are being operated on, if they are burnt or stabbed by accident, or whatever happens. I hope that answers your question.
Mr. Mesko: I think the question might be even more futuristic than I like to be. With respect, I think the real challenge you face is not about the consequences of having hospitals just with robots. If that might be the case, it's far away in the future. The real challenge you face as a government today is that many of your patients are dying decades earlier because you can't access data about their health. You don't help them with digital data to predict diseases, again decades earlier. You don't empower them to take proactive care about their own health.
I am happy to talk about a science fictional future — 30, 40, 50 years from now — but we are still struggling with purely getting our own medical records from our health care institution, not even talking about robots and artificial intelligence, just owning my own data, to put that into my hands so that I can make decisions about my lifestyle and position. From there, when we have digital health, we can start thinking about the far-off future with robots and algorithms and amazing science fiction like paperless hospitals.
The Chair: Before going to the second round, I would like to get in on this a little bit. First of all, with regard to the last comment, Senator Dean raised a question yesterday that I cut off and shifted towards AI, which deals with electronic medical records and electronic health records. As a result of this, I would like to put on the record that, in this committee's opinion, Canada has done a desperately poor job of spending billions of dollars to attempt to develop electronic health records. The electronic health record is the total health record that we really want to see. The electronic medical record is the record in the local doctor's office. As I said, we have spent billions of dollars and we have pounded the government agency that's responsible for this to try to move this critical issue forward, because to me it is absolutely essential today, let alone in the future, to have these available.
I have delved into it a little bit outside of the committee, and it seems to me that the biggest issue is this hang-up on protection of data. In Canada, we have an absolute fascination in the protection of data, yet we all provide our income tax records to a system that is known to be — and there is more data on there than we'd get from a health record.
Since Senator Dean raised this aspect and it has come up today, I wanted to put that into perspective in terms of where we are. We have argued that that is a critical issue, and we have made serious recommendations through some of our reports in that particular area.
I want to come back to our issue today. One of the things that has come up during this session — and it comes up in discussions on the future impacts of AI — is the potential of throwing millions of people out of work, that is, terminating existing jobs. They're no longer needed; they're replaced by robots or technology of some sort. The question then becomes that if you've placed millions more people on — let's keep the term simple — unemployment rolls and you have eliminated tax-paying jobs, how do governments finance the consequences? That is, providing the services based on an artificial intelligence world, on the one hand, and providing a new lifestyle, because we are going to change lifestyles dramatically for those who are socially now unemployed, at least initially. It will be much more complex than that, but those are concepts we can understand today. The question is, how will they do that?
I want to throw one direct suggestion out to you to focus the discussion, and I'll put it to Mr. Prakash first. Why can't we look into the future and see the taxation of robots? Governments, we know, are horribly bad at innovation, but they will have to get innovative in this area. Why could they not develop some new innovative, futuristic taxation scheme that brings in revenue from the very technologies that are causing the social displacement?
Mr. Prakash: Senator, easy answer: Read my book. That's the easy answer. I covered all this in my book. Now we're talking my language.
To answer your first question about data protection, as I mentioned at the beginning, Finland has digitized all patient data for the past 30 years. That enables them to predict health issues — predictive health care.
Are we all familiar with the term "block chain''? A block chain is a digital ledger of sorts. It's a digital ledger that is recording transactions. If this senator takes out $5, it will record that. If this senator deposits $10, it will record that. It is a digital ledger. It's given to everybody. It's completely transparent. It's looked at as the future of transparency. In the U.S., IBM has partnered with the Food and Drug Administration to use block chain to secure patient data. That's just an example of what is taking place in the data protection realm.
Another point to think about is that if you have a humanoid robot sitting here, it will never be off, if you think about it. If its goal is to help the elderly, bring medicine or carry food, or if it has a video camera to talk with people, it will never be off; it will always be watching you. How do you secure that data, number one? And number two, what if it's a foreign government that has that data? If you have a Japanese robot, it will be a Japanese company with access to that data. So does that mean that Canada now legislates what China has done and mandates that companies store their data within China? Is that what Canada should do? I don't know; it's just a point.
As far as automation goes, 85 per cent of jobs are at risk in Ethiopia; 49 per cent in the U.S.; 50 per cent in the U.K.; 42 per cent in Canada; 48 or 49 per cent in Germany. In January, earlier this month, 34 people were laid off at an insurance company in Japan and they were replaced with IBM Watson. I'm not talking 15 or 20 years; I'm talking about a few weeks ago that happened.
So I think we will have to take steps to address automation. What should the government do? Is taxing robots is the right idea? The question really is that if a company comes from abroad and moves their operations here — let's say a German company comes and moves their operations and factories here and only employs robots, and you tax them. What stops them, in 24 months, from getting up and leaving? Or what stops the German government from taxing Canadian companies that are operating in Germany? Yes, we can talk about a robot tax and we can talk about taxing industrial robots, arms or whatnot, but there are ramifications for each choice we make.
The Chair: I can see ways around that, but I will not debate this today. I will go to Dr. Mesko for his input.
Mr. Mesko: I still believe that hackers are more interested in my bank account details than my blood pressure measurements on my smartphone. Obviously, privacy is the most important issue, although medical records are just the first step. People have bigger challenges. In Iceland, they sequenced the genomes of tens of thousands of patients in the country. It means they made the information on DNA available. The government used smart AI algorithms to mine the data, and the challenge they face today is that they can tell many patients that they might have breast or ovarian cancer later on, based on their DNA profile, but whether they should tell those patients or not is a real ethical issue. Privacy is obviously huge.
With regard to your second question, which is really futuristic — and I love science fiction myself; I'm addicted to science fiction — we need to keep in mind that when automation removes jobs from the market, it will also create new ones, because we will have new roles for people. A lot of patients require empathy, and those people who can provide them with empathy, even though the bad decisions have been made by automation and medical professionals, they will require a new kind of health care worker. We will have life-size strategists who can help patients use the data they measure about themselves to make smart lifestyle decisions. They will need people who can repair those robots, who can design chatbot interactions with artificial intelligence, who can design 3-D printers and organs for individuals, people who can become patient assistants who will help patients go through the jungle of health care information when they need to make so many decisions about which treatment to choose or which diagnostic procedure to start with. So you open up a new market for a new kind of job. That is the essence of disruption; it's not linear but exponential. While it moves things away, it creates new stuff as well.
Mr. Prakash: Those are great comments. I totally agree.
I want to make one initial point, and that is in Canada the projections are that 42 per cent of jobs are at risk from being automated. Another study put it at 7.5 million jobs.
Let's say they are wrong. These studies are talking about 15 or 20 years from now, and let's say they're wrong by 50 per cent. That's a big drop from any study. That's still 3 to 4 million jobs that will disappear. It's likely, like in past waves of automation, new jobs were created, and I don't know whether it's going to be 4 million new jobs created out of this. If even a fraction of these projections come true, it's still a huge number.
The Chair: I would love to discuss this further, but we'll go on to the second round.
Senator Meredith: My question is to both of you. It's about smart regulation. Dr. Mesko and Mr. Prakash, you talk about the private sector and government moving forward and not leaving any community behind with respect to embracing this innovation. Can you tell me about the collaboration you see between the technology that's being developed, the corporations that have this and how they can integrate to push government to ensure that those smart regulations are actually implemented, and the benefit for Canadians, firstly, in terms of the proper health care of individuals? The diagnosis is there, the statistics are there, and bridging that gap and using the technology and those companies.
I go into a hospital and quickly I'm diagnosed, I get treated, and things are then billed. Right away, the insurance companies are billing. What's their role? What's their responsibility?
Mr. Mesko: Thank you for the question. As we discussed before, 3-D printing and binding tissues, real tissues from organs, is a possibility today, but five years ago that was a science fiction idea. Yet the company that I mentioned in my report, Organovo, started working on printing liver tissue that can function like a liver. They knew they had no chance for FDA approval because of the new essence of the technology that they developed, so they kept on pushing the FDA to at least start doing workshops for companies focusing on 3-D printing. The FDA did a workshop in 2015 for companies focusing on 3-D printing, and they started collaborating or working closely with the FDA to understand what they expect from them to be able to create products that can get FDA approval eventually. The FDA wanted to understand what the technology is about. It was new for them. They had to understand the details and how they can develop their own framework to be ready when the technology is ready to go live as quick as possible. The announcement came out a few months ago that in three or four years, the first 3D-printed biological tissue might get FDA.
That's how it works. Smart companies push regulations forward, and smart regulators want to understand how technology works because they want to get the best treatment options for their patients.
Mr. Prakash: My colleague and I have spoken about a lot of different technologies, a lot of different applications. Let's forget everything for a second and let's just look at artificial intelligence for diagnosis. I talked about it before. AI can take pictures. Let's just look at that.
I'd like to ask you and the rest of the committee and the Canadian government a question too: Why should a company work with you? If a company adopts a business model the same way Uber has, why should they work with you? If a company is from the U.S. and offering this for 10 bucks a month or 20 bucks a month or $1,000 a month and people in Canada are willing to pay for it, how do you stop them? Do you introduce a Netflix tax, which I think was proposed recently? Why should they work with you? Do they have to work with you?
My answer to that is: No, they don't, on day one. They don't have to work with you on day one unless there are protests outside of their headquarter or there is violence brewing or something like that, What incentive should the Canadian government provide to help these start-ups or large corporations work with them when they come out with these business models?
The Chair: Thank you very much.
Senator Merchant: Mr. Prakash, you made some reference to this earlier on in your presentation. Technology moves very quickly and things change all the time. I'm thinking of the cellphone, the things that we're familiar with, and every so many months you get a new cellphone coming in that's better than the last one.
What is going to happen with these robots? Will we be replacing them on a constant basis? Will we be able to keep up with this? Maybe the hospital in Regina will not be able to replace things, and perhaps the doctor in the U.S. does not want to work with the model that we have in Regina.
What do you see in the future? How do you see this? Companies are going to want to make money. There will be all kinds of advertising: this is a better model and try this. Then I will say, "I don't want to go to the hospital in Regina because they don't have the latest.'' I should be able to go somewhere else, to Ottawa or Toronto or something. How is this going to work, as you look at the future?
Mr. Prakash: I was in Denmark last year, and I was touring a major robotics company. They sell a robotic arm, and it's for factories. They don't sell a warranty on it. There's no warranty on the arm. So your question is very valid, and the real answer is there is no answer.
If this is a business model, then they will keep coming out with new robots, when they want, how they want. But that's where I do think that robotics as a service, which I mentioned earlier where you can pay a monthly fee instead of owning a robot upfront, will begin to take off a lot more. For example, in Toronto, the Humber River Hospital has started using robots to mix chemotherapy medicines, to move supplies around, et cetera. It probably cost them millions of dollars, if not more. If they can simply spend $10,000 a month or $20,000 a month, and when the new version comes out they get the new version, then that incentivizes them a lot more than spending this money upfront.
Yes, companies are going to keep coming out with new robots. As I tell people, we're going to start seeing robotics go mainstream now, going forward. In that sense, we're going to start seeing robots increasingly being more advanced and advanced. So yes, they are going to keep coming out. I think robotics as a service will be the way to go.
The Chair: I think this is a great example of evolution of existing practices. Simply take the leasing of automobiles, which you exchange on a regular basis, and as they move faster and faster they give you promotions to change faster in your technology and so on. So I think you've given us an insight into a number of practical ways that you could deal with this.
Mr. Prakash: One more thing: I'm guessing you're assuming physical robots. We're talking about physical robots. In that sense, obviously they have to be replaced. But when it comes to artificial intelligence, your AI doctors, you may not have to spend any more. It is just like Netflix keeps updating their catalogue. They might keep updating their capabilities and algorithms.
The Chair: Your biometrics are updated on a regular basis.
[Translation]
Senator Mégie: I would like to talk about the making of 3-D organs. I know the technology is not there yet, but it will come. Have you thought about the possibility of rejection? When an organ is transplanted from one person to another, the recipient has to take expensive medications and be regularly monitored by medical staff because if the organ is rejected, it's game over.
In developing 3-D objects, has consideration been given to the body's tolerance to a product such as ink, paper or plastic?
[English]
Mr. Prakash: Yes, I've been thinking about rejection. I'll bring up another point at the end.
There's a university in the U.S. that is building what they call a hyper-realistic bone from 3-D printing. Right now, if you need a bone transplant, they take grafts from a bone somewhere else in your body. Those grafts have stem cells in them. When they move it, it can automatically grow on its own. The problem is, how do you reciprocate that in a 3-D printer? They've managed to do that with what they call a 3-D printing bone. It's actually based upon what my colleague said when it comes to scaffolding.
I'd like to raise a different issue when it comes to 3-D printing. If people can 3-D print organs, and this is far away yet, how will the government regulate that? We'll see another black market emerging. We'll see 3-D printed kidneys being traded, 3-D printed livers, hearts, you name it. It's one thing if 3-D printing reaches a point where each of us can afford it, but if there's still $500,000 for a machine or a million dollars for a machine, then I think you're going to see a very big black market emerge. Perhaps that's something the committee can discuss at a later date, but it's something to think about.
Mr. Mesko: I think companies working on 3-D bioprinting would be more than happy to focus on the rejection issue today because it means that they would already have functioning organ tissues ready for market, but they don't have it yet. It will take one or two decades to get to this point, although there's new research focusing on a nanometre-sized coating where a small layer could be put around organ tissue and can prevent it from rejection, even though it's not made from the patient's stem cell. Again, that's an issue in the far future. They are still focused on creating organ tissues that can function like human organs, and that is still a huge challenge.
The Chair: Isn't it possible that, in some of the examples you've given of approaches to functioning organs, that the cells could be harvested and multiplied from the patient and then used as the ink? If the cells and other tissues are cloned from the patient and incorporated into the process, wouldn't that go a long way to helping with the problem?
Mr. Mesko: I think it's absolutely possible, although I'm not a molecular biologist. I still think that creating the three-dimensional structure in which the cells communicate with each other, just like in an organ, is a much bigger issue than doing the same originally from the stem cells from the patient. The structural challenges are bigger now.
The Chair: Absolutely. Thank you very much.
Senator Seidman: Perhaps this question is best addressed to you, Mr. Prakash. In the big pot of national R&D spending on robots, AI and 3-D printing, what would you say is the proportion of that pot that's spent in the health care area? Is there any overlap among industries or are they siloed?
Mr. Prakash: I'm going to take a global approach to this. Are we all familiar with the term robot density? Robot density is used to measure how advanced a country is when it comes to robotics. It's essentially the number of robots per 10,000 workers. That's robot density.
Right now, Japan, South Korea and Germany dominate the index. Canada is far below. In that sense, when it comes to government investments in robotics right now, South Korea just unveiled US$5.8 billion to invest in robotics and artificial intelligence. The main goal is industrial automation, factory work, automating factory work.
I have not come across a government that is investing billions simply on health care robots, although I do think — and this is my own opinion — that China is because one of their provinces has earmarked $150 billion for robotics in automation. With the elderly population rising, I do think Beijing will be investing in health care robots. But it's hard to see how China is investing. These things are still quite censored.
Mr. Mesko: If you look at the most innovative start-ups worldwide, I would say in my experience I have no studies for that. Less than 10 per cent of them are focusing on health care. Even those that are purely technological companies might start diving into health care because they see the opportunity, but they are afraid of the vast amount of regulation and the boundaries and how reluctant physicians are and how hard it is to motivate patients to change their lifestyle.
Even Google failed at this. Google launched Google Health about seven or eight years ago as the ultimate medical records solution globally. We know that they can come up with global solutions in so many areas, but they failed very fast. They realized that this is a market they just don't want to dive in to yet.
This year, 2017, might be the year when at least one technology giant starts developing something really mind- blowing in health care. I know that the mother company of Google became Alphabet last year, and Alphabet started investing in companies that focused on health care technologies but with the brain power, the resources, of Google.
For example, they just wanted to come up with a new surgical robot system. They are working with the pharma company Johnson & Johnson. That could be the first competitor of the da Vinci surgical system that has been around for so many years.
So we know tech giants are working on health care solutions, but none of them dedicate enough attention, money and human resources to that.
Mr. Prakash: There's also the indirect investment in health care. Japan's official strategy is called New Robot Strategy. One of the goals of that strategy is something they call a smart factory, which is a fully automated factory that runs 24-7 with no humans. In terms of the entirety of Japan, will some of those factories be for medicine? Will some of those factories be for something else? We don't know yet. There's also the indirect effect.
Like in Japan, IBM Watson replacing 34 employees a few weeks ago, will that be taken into account when it comes to robotics investments and automation in health care? There might be indirect effects too.
Senator Raine: I'd like to pull back a little bit closer to where we are now. The future is almost mind-boggling. We seem to have problems coming up with good electronic health record systems and systems that can be accessed not only by all doctors wherever you are in a country but also by the patients. In British Columbia we now have a program called my ehealth.
There's always a danger in saying we need one system for everyone where perhaps what we really need is a way for different systems to talk to each other. In terms of the future, is there a specialty on how to make all of these different things talk to each other?
Mr. Prakash: You're talking about interoperability between different systems, and that's an issue that's being raised. Right now, from my understanding, that's being tackled in the military sphere because you have advances such as DARPA making advances with artificial intelligence and drones and how will these different technologies work together on a battlefield. How will they communicate with each other? This is similar to the question you had, how will these different systems relate, and not just in terms of digitized health records.
For example, there is a robot that was unveiled last month or this month, and it's meant for elderly patients. You bring it home, you lay it on your desk, and its goal is to dispense medicine when you need it. On top of that, it can communicate with health officials and answer your health questions. So you have a robot that comes preprogramed with that software, but tomorrow, what if the health official it's speaking to is AI? How does it communicate with AI? The answer is there is nothing really out there to enable that, nothing that I've come across, but we need to solve that problem.
Senator Raine: If all of these advances are being made by corporations, they don't really have a motivation to be connective to other corporations.
Mr. Prakash: Yes and no. Just as my colleague brought up, Elon Musk and a few others have created something called OpenAI, which is an open source way of creating AI to enable it to go mainstream faster versus Apple creating AI and Microsoft creating AI and they all have their different silos.
We will see companies working together based on common objectives. An artificial intelligence group was just created by different companies to tackle how to regulate AI. What are the ethics in AI? They came together for that.
Will they come together for health care? Will the systems work together? More importantly, will they work with the systems the government has? If the government is still using version 1 and they're using something else, is it interoperable? I don't know at this point.
Mr. Mesko: The way the question was asked refers to the old structure of medicine and health care in which a government, an institution or an insurance company provides care and patients receive it. However, based on current trends, that's not the case. I'm pretty sure I will have all my medical records, the results from my genetic tests I had myself, on my smartphone sooner than any medical record system in the world could handle it.
Digital health refers to patients becoming empowered. I don't want my data to be somewhere else. I want to own my data that I paid for either through taxes or paying for service. I want to use my data for further analysis. I want to use physicians, second opinion and artificial intelligence to keep on analyzing my data. When I bring these results to my professional or my physician, I want to be in equal partnership. That means that the medical record system is where patients are.
Point of care is the term that we use for the place where diagnostic procedures or treatments take place. The point of care has been the hospital, the clinic or the practice of the physician for hundreds of years. Tomorrow, the point of care will be the patient themselves. Wherever the patient is, they will be able to access all that information through digital health services or smartphones, and they will be able to draw even further conclusions from that.
The Chair: Thank you both very much. Before I sum up, if there's something that hasn't come up today that you would like to leave us with as a particular message to give as an overall comment, please go ahead. I'll start with you, Dr. Mesko. Is there something that hasn't come up that you would like to speak to before you leave?
Mr. Mesko: It's a favour or a pledge. In my work, I constantly focus on looking at trends and technologies that will change health care for the better. I look at how the cultural revolution is occurring. If anyone focuses on certain technologies, nothing good will happen to people. We have to focus on how we help people understand technologies, how we help them implement technology into their lifestyle, and how we empower patients to make their own decisions with their medical professionals. No one will take care of our health or disease except us. It's our chance. We need to put data into their own hands and start becoming proactive.
If you look at what empowered patients in your country need from you, what they request from you, that's the right path to walk on. Any country that's going after their empowered patients who are having their voices heard through social media so much, they are on the right path. Anybody else, we just keep on focusing on technology.
Thank you for this chance to speak with you about these issues today.
Mr. Prakash: My only comment is to keep on doing what you are doing. I mean that to all of you. What this committee is doing is trying to decipher the future and what it means for Canada. That's a requirement. If the government doesn't do it and prepare the country for it, then who will? So keep doing what you're doing. Thank you for inviting me.
The Chair: Thank you very much. This has been a fascinating coverage of so many aspects of this critical issue.
I was struck by your last comment, Dr. Mesko, with regard to the information and having the availability of it, but we have to have a way that patients can quickly identify what source of information is reliable.
We've done a recent study on dementia in Canada, where there is no organized approach to helping Canadians deal with this disease. We've recommended that Canada move forward on a national strategy in this area. The challenge of an individual who has dementia is where to go to get information in Canada on this issue. If you googled dementia — I did so yesterday morning — you will get over 12 million hits. How does the individual patient, just recently diagnosed, have any idea how to move forward? If you look at the major hits on Google, those are largely people advertising various remedies for this, that or the other.
We, in this particular case, recommended that the Public Health Agency of Canada, which has a mandate to inform Canadians in these important areas, set up and organize a website that Canadians could go to initially to find really good reference information. In the case of Canada, that would refer them probably to a site in their own province and move forward.
Do you have a comment on the need to have people recognize the reliability of their information, Mr. Prakash?
Mr. Prakash: I think that there's a different way to look at it. In the future, information will come to us. To give you an example of what that means, right now our smartphones are dumb. They wait for us to interact with it. Huawei, a Chinese company, has recently announced that they want to make their phones a double of you through artificial intelligence. What that means is the phone will be interacting with the world around it, even if it's just in your pocket. So if you think about an AI that can predict based on pollution and weather that the area you are walking into is dangerous for your health, you will know that. It will come to you. It will be like a notification. Right now we get Twitter notifications; tomorrow we will get AI notifications for health. Information will come to you and you won't have to necessarily go out and find it.
The Chair: I think that's excellent.
With that, colleagues, I want to thank you for the questions you've raised that have enlightened us through the answers we've received. This has been an excellent start on this absolutely important study.
(The committee adjourned.)