TED日本語 - ダニエル・クラフト: 医学の未来ですか？アプリがありますよ

A couple of years ago when I was attending the TED Conference in Long Beach, I met Harriet. We'd actually met online before -- not the way you're thinking. We were actually introduced because we both knew Linda Avey,one of the founders of the first online personal genomic companies. And because we shared our genetic information with Linda, she could see that Harriet and I shared a very rare type of mitochondrial DNA -- Haplotype K1a1b1a -- which meant that we were distantly related. We actually share the same genealogy with Ozzie the iceman. So Ozzie, Harriet and me. And being the current day, of course, we started our own Facebook group. You're all welcome to join. And when I met Harriet in person the next year at the TED Conference, she'd gone online and ordered our own happy Haplotype T-shirts. (Laughter)

Now why am I telling you this story, and what does this have to do with the future of health? Well the way I met Harriet is actually an example of how leveraging cross-disciplinary, exponentially-growing technologies is affecting our future of health and wellness -- from low-cost gene analysis to the ability to do powerful bio-informatics to the connection of the Internet and social networking. What I'd like to talk about today is understanding these exponential technologies. We often think linearly. But if you think about it, if you have a lily pad and it just divided every single day -- two,four,eight,16 -- in 15 days you have 32,000. What do you think you have in a month? We're at a billion. So if we start to think exponentially, we can see how this is starting to affect all the technologies around us.

And many of these technologies -- speaking as a physician and innovator -- we can really start to leverage to impact the future of our own health and of health care, and to address many of the major challenges that we have in health care today, ranging from the really exponential costs to the aging population, the way we really don't use information very well today, the fragmentation of care and often the very difficult course of adoption of innovation. And one of the major things we can do we've talked a bit about here today is moving the curve to the left. We spend most of our money on the last 20 percent of life. What if we could spend and incentivize positions in the health care system and our own self to move the curve to the left and improve our health, leveraging technology as well? Now my favorite technology, example of exponential technology, we all have in our pocket. So if you think about it, these are really dramatically improving. I mean this is the iPhone 4. Imagine what the iPhone 8 will be able to do.

Now, I've gained some insight into this. I've been the track share for the medicine portion of a new institution called Singularity University based in Silicon Valley. And we bring together every summer about 100 very talented students from around the world. And we look at these exponential technologies from medicine, biotech, artificial intelligence, robotics, nanotechnology, space, and address how can we cross-train and leverage these to impact major unmet goals. We also have seven-day executive programs. And coming up next month is actually Future Med, a program to help cross-train and leverage technologies into medicine.

Now I mentioned the phone. These mobile phones have over 20,000 different mobile apps available -- to the point where there's one out of the U.K. where you can pee on a little chip connected to your iPhone and check yourself for an STD. I don't know if I'd try that yet, but that's available. There are all other sorts of applications, merging your phone and diagnostics, for example -- measuring your blood glucose on your iPhone and sending that, potentially, to your physician so they can better understand and you can better understand your blood sugars as a diabetic. So let's see now how exponential technologies are taking health care. Let's start with faster. Well it's no secret that computers, through Moore's law, are speeding up faster and faster.

We have the ability to do more powerful things with them. They're really approaching, in many cases surpassing, the ability of the human mind. But where I think computational speed is most applicable is in that of imaging. The ability now to look inside the body in real time with very high resolution is really becoming incredible. And we're layering multiple technologies -- PET scans, CT scans and molecular diagnostics -- to find and to seek things at different levels. Here you're going to see the very highest resolution MRI scan done today, reconstructed of Marc Hodosh, the curator of TEDMED. And now we can see inside of the brain with a resolution and ability that was never before available, and essentially learn how to reconstruct, and maybe even re-engineer, or backwards engineer, the brain so we can better understand pathology, disease and therapy. We can look inside with real time fMRI -- in the brain at real time. And by understanding these sorts of processes and these sorts of connections, we're going to understand the effects of medication or meditation and better personalize and make effective, for example, psychoactive drugs.

The scanners for these are getting small, less expensive and more portable. And this sort of data explosion available from these is really almost becoming a challenge. The scan of today takes up about 800 books, or 20 gigabytes. The scan in a couple of years will be one terabyte, or 800,000 books. How do you leverage that information? Let's get personal. I won't ask who here's had a colonoscopy, but if you're over age 50, it's time for your screening colonoscopy. How would you like to avoid the pointy end of the stick? Well now there's essentially a virtual colonoscopy. Compare those two pictures, and now as a radiologist, you can essentially fly through your patient's colon and, augmenting that with artificial intelligence, identify potentially, as you see here, a lesion. Oh, we might have missed it, but using A.I. on top of radiology, we can find lesions that were missed before. And maybe this will encourage people to get colonoscopies that wouldn't have otherwise.

And this is an example of this paradigm shift. We're moving to this integration of biomedicine, information technology, wireless and, I would say, mobile now -- this era of digital medicine. So even my stethoscope is now digital. And of course, there's an app for that. We're moving, obviously, to the era of the tricorder. So the handheld ultrasound is basically surpassing and supplanting the stethoscope. These are now at a price point of -- what used to be 100,000 euros or a couple of hundred-thousand dollars -- for about 5,000 dollars, I can have the power of a very powerful diagnostic device in my hand. And merging this now with the advent of electronic medical records -- in the United States, we're still less than 20 percent electronic. Here in the Netherlands, I think it's more than 80 percent.

But now that we're switching to merging medical data, making it available electronically, we can crowd source that information, and now as a physician, I can access my patients' data from wherever I am just through my mobile device. And now, of course, we're in the era of the iPad, even the iPad 2. And just last month the first FDA-approved application was approved to allow radiologists to do actual reading on these sorts of devices. So certainly, the physicians of today, including myself, are completely reliable on these devices. And as you saw just about a month ago, Watson from IBM beat the two champions in Jeopardy. So I want you to imagine when in a couple of years, when we've started to apply this cloud-based information, when we really have the A.I. physician and leverage our brains to connectivity to make decisions and diagnostics at a level never done. Already today, you don't need to go to your physician in many cases. Only for about 20 percent of actual visits do you have to lay hands on the patient. We're now in the era of virtual visits -- from sort of the Skype-type visits you can do with American Well, to Cisco that's developed a very complex health presence system.

The ability to interact with your health care provider is different. And these are being augmented even by our devices again today. Here my friend Jessica sent me a picture of her head laceration so I can save her a trip to the emergency room -- I can do some diagnostics that way. Or might we be able to leverage today's gaming technology, like the Microsoft Kinect, and hack that to enable diagnostics, for example, in diagnosing stroke, using simple motion detection, using hundred-dollar devices. We can actually now visit our patients robotically -- this is the RP7; if I'm a hematologist, visit another clinic, visit a hospital. These will be augmented by a whole suite of tools actually in the home now. So imagine we already have wireless scales. You can step on the scale. You can Tweet your weight to your friends, and they can keep you in line.

We have wireless blood pressure cuffs. A whole gamut of these technologies are being put together. So instead of wearing these kludgy devices, we can put on a simple patch. This was developed by colleagues at Stanford, called the iRhythm -- completely supplants the prior technology at a much lower price point with much more effectivity. Now we're also in the era, today, of quantified self. Consumers now can buy basically hundred-dollar devices, like this little FitBit. I can measure my steps, my caloric outtake. I can get insight into that on a daily basis. I can share that with my friends, with my physician. There's watches coming out that will measure your heart rate, the Zeo sleep monitors, a whole suite of tools that can enable you to leverage and have insight into your own health.

And as we start to integrate this information, we're going to know better what to do with it and how better to have insight into our own pathologies, health and wellness. There's even mirrors today that can pick up your pulse rate. And I would argue, in the future, we'll have wearable devices in our clothes, monitoring ourselves 24/7. And just like we have the OnStar system in cars, your red light might go on -- it won't say "check engine" though. It's going to be "check your body" light, and go in and get it taken care of. Probably in a few years, you'll check into your mirror and it's going to be diagnosing you. (Laughter) For those of you with kiddos at home, how would you like to have the wireless diaper that supports your ... too much information, I think, than you might need. But it's going to be here.

Now we've heard a lot today about new technology and connection. And I think some of these technologies will enable us to be more connected with our patients, and take more time and actually do the important human touch elements of medicine, as augmented by these sorts of technologies. Now we've talked about augmenting the patient, to some degree. How about augmenting the physician? We're now in the era of super-enabling the surgeon who can now go inside the body and do things with robotic surgery, which is here today, at a level that was not really possible even five years ago. Now this is being augmented with further layers of technology like augmented reality. So the surgeon can see inside the patient, through their lens, where the tumor is, where the blood vessels are. This can be integrated with decisions support. A surgeon in New York can be helping a surgeon in Amsterdam, for example. And we're entering an era of really, truly scarless surgery called NOTES, where the robotic endoscope can come out the stomach and pull out that gallbladder all in a scarless way and robotically. And this is called NOTES, and this is coming -- basically scarless surgery, as mediated by robotic surgery.

Now how about controlling other elements? For those who have disabilities -- the paraplegic -- there's the era of brain-computer interface, or BCI, where chips have been put on the motor cortex of completely quadriplegic patients and they can control a curser or a wheelchair or, eventually, a robotic arm. And these devices are getting smaller and going into more and more of these patients. Still in clinical trials, but imagine when we can connect these, for example, to the amazing bionic limb, such as the DEKA Arm built by Dean Kamen and colleagues, which has 17 degrees of motion and freedom and can allow the person who's lost a limb to have much higher levels of dexterity or control than they've had in the past.

So we're really entering the era of wearable robotics actually. If you haven't lost a limb -- you've had a stroke, for example -- you can wear these augmented limbs. Or if you're a paraplegic -- like I've visited the folks at Berkley Bionics -- they've developed eLEGS. I took this video last week. Here's a paraplegic patient actually walking by strapping on these exoskeletons. He's otherwise completely wheelchair-bound. And now this is the early era of wearable robotics. And I think by leveraging these sorts of technologies, we're going to change the definition of disability to in some cases be superability, or super-enabling. This is Aimee Mullins, who lost her lower limbs as a young child, and Hugh Herr, who's a professor at MIT who lost his limbs in a climbing accident. And now both of these can climb better, move faster, swim differently with their prosthetics than us normal-abled persons.

Now how about other exponentials? Clearly the obesity trend is exponentially going in the wrong direction, including with huge costs. But the trend in medicine actually is to get exponentially smaller. So a few examples: we're now in the era of "Fantastic Voyage," the iPill. You can swallow this completely integrated device. It can take pictures of your GI system, help diagnose and treat as it moves through your GI tract. We get into even smaller micro-robots that will eventually autonomously move through your system again and be able to do things that surgeons can't do in a much less invasive manner. Sometimes these might self-assemble in your GI system and be augmented in that reality.

On the cardiac side, pacemakers are getting smaller and much easier to place so you don't need to train an interventional cardiologist to place them. And they're going to be wirelessly telemetered again to your mobile devices so you can go places and be monitored remotely. These are shrinking even further. Here's one that's in prototyping by Medtronic that's smaller than a penny. Artificial retinas, the ability to put these arrays on the back of the eyeball and allow the blind to see. Again, in early trials, but moving into the future. These are going to be game changing. Or for those of us who are sighted, how about having the assisted-living contact lens? BlueTooth, WiFi available -- beams back images to your eye. Now if you have trouble maintaining your diet, it might help to have some extra imagery to remind you how many calories are going to be coming at you.

How about enabling the pathologist to use their cell phone again to see at a microscopic level and to lumber that data back to the cloud and make better diagnostics? In fact, the whole era of laboratory medicine is completely changing. We can now leverage microfluidics, like this chip made by Steve Quake at Stanford. Microfluidics can replace an entire lab of technicians. Put it on a chip, enable thousands of tests to be done at the point of care, anywhere in the world. And this is really going to leverage technology to the rural and the under-served and enable what used to be thousand-dollar tests to be done at pennies and at the point of care. If we go down the small pathway a little bit farther, we're entering the era of nanomedicine, the ability to make devices super small to the point where we can design red blood cells or microrobots that will monitor our blood system or immune system, or even those that might clear out the clots from our arteries.

Now how about exponentially cheaper? Not something we usually think about in the era of medicine, but hard disks used to be 3,400 dollars for 10 megabytes -- exponentially cheaper. In genomics now, the genome cost about a billion dollars about 10 years ago when the first one came out. We're now approaching essentially a thousand-dollar genome -- probably next year to two years, probably a hundred-dollar genome. What are we going to do with hundred-dollar genomes? And soon we'll have millions of these tests available. And that's when it gets interesting, when we start to crowdsource that information. And we enter the era of true personalized medicine -- the right drug for the right person at the right time -- instead of what we're doing today, which is the same drug for everybody -- sort of blockbuster drug medications, medications which don't work for you, the individual. And many, many different companies are working on leveraging these approaches.

And I'll also show you a simple example, from 23andMe again. My data indicates that I've got about average risk for developing macular degeneration, a kind of blindness. But if I take that same data, upload it to deCODEme, I can look at my risk for sample Type II diabetes. I'm at almost twice the risk for Type II diabetes. I might want to watch how much dessert I have at the lunch break for example. It might change my behavior. Leveraging my knowledge of my pharmacogenomics -- how my genes modulate, what my drugs do and what doses I need are going to become increasingly important, and once in the hands of the individual and the patient, will make better drug dosing and selection available.

So again, it's not just genes, it's multiple details -- our habits, our environmental exposure. When was the last time your physician asked you where you've lived? Geomedicine: where you've lived, what you've been exposed to, can dramatically affect your health. We can capture that information. So genomics, proteomics, the environment, all this data streaming at us individually and us, as poor physicians, how do we manage it? Well we're now entering the era of systems medicine, or systems biology, where we can start to integrate all of this information. And by looking at the patterns, for example, in our blood of 10,000 biomarkers in a single test, we can start to look at these little patterns and detect disease at a much earlier stage. This has been called by Lee Hood, the father of the field, P4 medicine. We're going to be predictive; we're going to know what you're likely to have. We can be preventative; that prevention can be personalized; and more importantly, it's going to become increasingly participatory. Through websites like Patients Like Me or managing your data on Microsoft HealthVault or Google Health, leveraging this together in participatory ways is going to become increasingly important.

So I'll finish up with exponentially better. We'd like to get therapies better and more effective. Now today we treat high blood pressure mostly with pills. What if we take a new device and knock out the nerve vessels that help mediate blood pressure and in a single therapy to cure hypertension? This is a new device that is essentially doing that. It should be on the market within a year or two. How about more targeted therapies for cancer? Right, I'm an oncologist and I have to say most of what we give is actually poison. We've learned at Stanford and other places that we can discover cancer stem cells, the ones that seem to be really responsible for disease relapse. So if you think of cancer as a weed, we often can whack the weed away. It seems to shrink, but it often comes back. So we're attacking the wrong target. The cancer stem cells remain, and the tumor can return months or years later. We're now learning to identify the cancer stem cells and identify those as targets and go for the long-term cure. And we're entering the era of personalized oncology, the ability to leverage all of this data together, analyze the tumor and come up with a real, specific cocktail for the individual patient.

Now I'll close with regenerative medicine. So I've studied a lot about stem cells -- embryonic stem cells are particularly powerful. We also have adult stem cells throughout our body. We use those in my field of bone marrow transplantation. Geron, just last year, started the first trial using human embryonic stem cells to treat spinal cord injuries. Still a Phase I trial, but evolving. We've been actually using adult stem cells now in clinical trials for about 15 years to approach a whole range of topics, particularly in cardiovascular disease. We take our own bone marrow cells and treat a patient with a heart attack, we can see much improved heart function and actually better survival using our own bone marrow drive cells after a heart attack.

I invented a device called the MarrowMiner, a much less invasive way for harvesting bone marrow. It's now been FDA approved, and it'll hopefully be on the market in the next year or so. Hopefully you can appreciate the device there curving through the patient's body and removing the patient's bone marrow, instead of with 200 punctures, with just a single puncture under local anesthesia.

But where is stem cell therapy really going? If you think about it, every cell in your body has the same DNA as you had when you were an embryo. We can now reprogram your skin cells to actually act like a pluripotent embryonic stem cell and to utilize those potentially to treat multiple organs in that same patient -- making your own personalized stem cell lines. And I think they'll be a new era of your own stem cell banking to have in the freezer your own cardiac cells, myocytes and neural cells to use them in the future, should you need them. And we're integrating this now with a whole era of cellular engineering, and integrating exponential technologies for essentially 3D organ printing -- replacing the ink with cells and essentially building and reconstructing a 3D organ.

That's where things are going to head -- still very early days. But I think, as integration of exponential technologies, this is the example. So in close, as you think about technology trends and how to impact health and medicine, we're entering an era of miniaturization, decentralization and personalization. And I think by pulling these things together, if we can start to think about how to understand and leverage these, we're going to empower the patient, enable the doctor, enhance wellness and begin to cure the well before they get sick. Because I know as a doctor, if someone comes to me with Stage I disease, I'm thrilled -- we can often cure them. But often it's too late and it's Stage III or IV cancer, for example. So by leveraging these technologies together, I think we'll enter a new era that I like to call Stage 0 medicine. And as a cancer doctor, I'm looking forward to being out of a job.

Thanks very much.

Host: Thank you. Thank you.

(Applause)

Take a bow. Take a bow.