Blog Post

One diabetic’s take on Google’s Smart Contact Lenses

One in 19 people on this planet have diabetes. I am one.

About four to six times a day I am reminded of many things — my genetics, my disgusting smoking habit (which I have kicked) and my (now modified) terrible eating habits that have led me to where I am today. I prick the skin on one of my fingers, take a droplet of blood, put it on a tiny strip that is attached to my AccuCheck Glucose meter, and make a note in a little black book. At some point during the day I have to take insulin shots so I can get through my day; doing the things that 18 of 19 people get to do without that extra help. I have to watch everything I eat: one extra macaroon or not taking a walk before going to bed can cause mood swings and lead to other complications. But like I said, it is a mess of my own making.

Just to give you a size and scope of the diabetes problem, I quote from a recent report from  Credit Suisse Research Institute which estimates that “close to 400 million people worldwide are affected by type II diabetes alone – a number that is quickly rising” and the “associated costs for the global healthcare system are estimated at a staggering $376 billion every year, representing over 10 percent of all healthcare costs.” By 2020 this could be a disease that impacts 500 million people and the costs “could rise to a whopping $700 billion,” the study said.

So when I read about Google’s “smart contact lens project,” which allows these lenses to measure blood sugar levels, for a very brief instant I was excited. It seemed like that finally we are getting to a place where needles, blood-soaked alcohol swabs and cotton balls are going to be history.

Over the years, many scientists have investigated various body fluids—such as tears—in the hopes of finding an easier way for people to track their glucose levels. But as you can imagine, tears are hard to collect and study. At Google[x], we wondered if miniaturized electronics—think: chips and sensors so small they look like bits of glitter, and an antenna thinner than a human hair—might be a way to crack the mystery of tear glucose and measure it with greater accuracy.

We’re now testing a smart contact lens that’s built to measure glucose levels in tears using a tiny wireless chip and miniaturized glucose sensor that are embedded between two layers of soft contact lens material. We’re testing prototypes that can generate a reading once per second. We’re also investigating the potential for this to serve as an early warning for the wearer, so we’re exploring integrating tiny LED lights that could light up to indicate that glucose levels have crossed above or below certain thresholds. It’s still early days for this technology, but we’ve completed multiple clinical research studies which are helping to refine our prototype. We hope this could someday lead to a new way for people with diabetes to manage their disease.

Google glucose testing contact lens

Reality Check

But after the initial excitement was over, cold reality set in. It also prompted me to ask the question: why is it that a company with such good intentions fails to ask itself very basic of questions, something a normal human being would ponder before embarking on a scientific quest?

For example,  why would they ignore the fact that as a diabetes patient, it is generally recommended that I not wear contact lenses. Yes, I understand that there are many different opinions about this, but it is generally thought of as smart to not wear contact lenses, as they always carry the risk of increased complications for diabetics. And on top of that if you have say, astigmatism (like I do), then contacts are less of an option.


Never mind the big fact that most of the people who do suffer from diabetes (Type II) tend to get it because of poor diet, most often because of lack of better diet options due to increased economic and financial stratification of our society. Diabetes is a growing problem in countries in South Asia and parts of Asia and Latin America, especially among those who fit in the lower income category; you know, the kind of people who might find contact lenses an expensive luxury. The less financially fortunate among us are very same group who are much more likely to not monitor their blood sugar levels due to work conditions and financial limitations.

I emailed Google’s press relations department — about 12 hours ago — asking them to elaborate on why the researchers took this specific approach. Yes, as a diabetic, I appreciate the efforts of Google in general and specifically Brian Otis and Babak Parviz, the project smart contact lens co-founders, just as I appreciate every effort made by every scientist and doctor around the world trying to tackle one of the worst diseases to afflict the post-industrial society.

Tone deaf?

And yet, I cannot get over what seems to me a tone-deaf approach by Google’s scientists. It also highlights Google’s fundamental challenge: it fails to think about people as people, instead it treats them as an academic or an engineering problem. Instead of trying to understand the needs of actual people, they emerge with an elegant technological solution.

It is not just this one time. Google+, their social network, is a fail because it fundamentally isn’t social or about people — it is an effort to solve Google’s need for social data for better advertising using machines. Similarly, Google Glasses are a cringe-worthy assault to the social interactions of normals, but because a certain subset of Googlers — including co-founders Sergey Brin and Larry Page — have a cyborg fetish, it is okay to make that design. It is frustrating for me to keep repeating this, because Google is a company with huge resources and those resources could be deployed more effectively and have a much more positive impact, more quickly. And to do that, the company needs to learn to be human and develop compassion for human condition.

As a diabetic, the only solution I am looking for is non-invasive and one that keeps me in a state of constant alertness about my blood sugar levels while matching that data with advice about what I should do.

Instead of trying to develop a contact lens that will still be quite expensive for many of the world’s poor, diabetic patients, why not focus all monetary and intellectual energies towards developing a more simpler solution that can be built at scale, very cheaply? Why not take the open source ethos and develop a product that actually be given away to anyone — say, emerging world pharmaceutical companies — who wants to manufacture it cheaply? The licensing terms could/should include the Tom’s Shoes Principle: buy one, give one away for free to someone who cannot afford them.

Instead of contact lenses, how about Google’s mighty X-machine focus its microscopic chip approach that Otis and Parviz are taking on a patch that can be applied to the skin — akin to a nicotine patch. The patch could carry the chip and send data to nearest phone — be it a smartphone or a cheap feature phone — and alert diabetics when there are major spikes or declines in blood sugar levels. These would be easier to use, cheaper to product and much more easily distributed.  It is a moonshot, and totally worth it.

Sure, it would lack the whiz-bang nature of the smart contact lenses, but it would be something I would welcome with a big grin and a credit card while thanking my stars and Google for having developed that. It’s not that I am not unappreciative of the impact that these contact lens breakthroughs might have in a few years. But another few years would mean another 100 million fellow humans suffering from something I live with everyday.


172 Responses to “One diabetic’s take on Google’s Smart Contact Lenses”

  1. adamold

    As a type-1 diabetic who has worn contact lenses without issue for the better part of 20 years, I think the fear about them is largely misplaced. Diabetics are cautioned to be careful with contact lenses precisely because many do not actively monitor their blood glucose levels, and as a consequence suffer side effects include dry eyes and increased ocular infections.

    Having a continuous glucose monitor is a game changer in terms of how easily blood sugar is regulated and how well the patient understands what the food they are eating will do to their blood sugar levels. I have been diabetic 26 years and was pretty well controlled with some testing but now I am on a Continuous Glucose Monitor and I am learning new things about common foods that weren’t obvious to me before.

    So…a continuous monitor has the capability to completely eliminate the diabetic side effects that make contact lenses riskier for diabetics.

  2. Tobias Jost

    Interesting take. But I see several problems with your proposed approach.

    The reason there isn’t a non-invasive blood sugar test on the market is that it’s extremely hard to do without some form of bodily fluid. Usually capillary blood, because that is the actual relevant medium and therefor can be used to take readings about what’s happening *now* instead of what happened hours or days ago (like with urine tests, which where the preferred way before the blood tests became available).

    Most research in the non-invasive field currently focuses on testing blood *through* a thin body part, mostly the earlobe since it has many capillary blood vessels, to measure light wave spectrum interference by shining different light frequencies from one side and measuring how much reaches the other side. Several companies have claimed success with this approach over the last decade, but so far none have actually reached the large scale clinical testing stage, never mind a working product for sale.

    As far as I recall, only a single company has ever claimed any success with the approach you are proposing, measuring through the skin from one side only. They claimed to use light wave reflection interference to do it. I don’t remember their name right now, but afaik, they are now defunct and their lead researcher is now working for Apple (or maybe they bought the company?)

    Both approaches, even if they managed to actually work, wouldn’t be a simple skin patch, because of the power requirements for the intense light required for the measurements and the processing power to compute the multitudes of interferences by the many elements of the skin. All prototypes so far have been handheld devices with large external battery packs and/or direct connections to a computer.

    Sure, they will shrink *over time*, but not without releasing several generations of ungainly and probably expensive devices first. And that leads me to my second problem: When has there ever been a new and complicated technology that focused on cheap mass production from the start? I mean, sure, new uses and refinement of existing tech can be made that way, but completely new tech? It’s always going to be for better off first, because that’s how you recoup the development costs. A cheap skin patch for every diabetic out there sound awesome, but since the tech currently simply doesn’t exist, it’s not gonna happen anytime soon and no company with an eye on profits would be even thinking seriously about it right now.

    The same can be said for Google’s contact lenses, btw. Sure, the first generation will probably be expensive and have problems like you mentioned. But the cost will come down over time and those problems may be mitigated. For example, since they are not primary meant as a seeing aid, they could be made smaller than regular lenses, be made to let liquid pass through, etc. Since they have a very small mass, the material costs are low, so they could become very, very cheap with mass production. How is that any different to what would happen with any other approach to the problem?

    My problem with them is a completely different one: The fact that they don’t use blood. I can see how using tear fluid is a much easier approach than the other non-invasive ideas, but since it’s a fluid that’s quite a bit removed from the thing we want to measure, blood, I don’t see how this can ever deliver accurate and most important timely results. How long does it take for the tear glans to use the material it already has at hand to produce the tears? How long does it take for it to travel along the tear ducts to reach the eye? How long to mix sufficiently with the fluid already there to be measured? What about the difference in tear fluid production between people in very humid and very dry environments?

    At best, measurements will always be a certain amount of time removed from what’s happening in the blood, making them less useful for anyone having to use insulin to manage their sugar levels.

  3. Vassil Mladjov

    As a husband of a Type I person for 22+ years, I love what Google is trying to do. I have talked to many manufactures of glucose monitors, I even tried to create an online secure social network for tracking sugars of my loved one. I know what it is to deal with a person who has a low or is going into shock and you need to use the glucogon or even worst, get several fireman to insert glucose IV. Any technology that can help a person with diabetes (type I) is a good thing. Is this new Google thing going to be perfect, no is it the answer, no, but is pushing the market to innovate.
    Now, I do think that many of the medical glucose vendors do not want to see this succeed at all, since this is how they make the big bucks! Just imagine that you don’t need to use 10 strips a day any more or that your CGM does not need to be changed every few days, who will NOT benefit from this? Well, you can figure this one out. I love that Google is doing here and I hope this forces many of the big vendors to move their asses and offer more innovative solutions.

  4. Gus S. Calabrese

    If tears have blood glucose level information , why can’t that tear be monitored at the eye duct or a similar place ?

    I did some searching and there is some possibility that glucose levels can be detected in all four serums…. blood, saliva, tears and urine.
    Certainly saliva and urine would be easier to obtain than tears and blood.

    The more free & open source sharing #FOSS there is… the faster the development will move… that includes getting the FDA and other regulators out of the way. Those who want to live in a regulated state can choose to do so and they can let the rest of us alone…. Best ……. 99guspuppet

  5. TheDarwin

    > Instead of contact lenses, how about Google’s mighty X-machine focus its microscopic
    > chip approach that Otis and Parviz are taking on a patch that can be applied to the
    > skin — akin to a nicotine patch.

    Om, speaking of “reality check”, did you actually look to see if that had been tried and failed?

    Or to see whether somebody (Georgia Tech perhaps in conjunction with partners) holds a huge pool of patents (on remarkably unfinished-looking technology, full of forward looking statements) that make it well-nigh impossible for anyone else to build a glucose-monitoring skin patch?


    These guys are at least doing something positive, not kicking back trying to tell others how to do their jobs. Like aarondfrancis says, “It has to start somewhere “

  6. David Knowles

    Open source is all very good but who going to fund the clinical trials they will need to prove their device works and get permission to sell them as a medical product in the West

    An there a big problem with your last suggestion of using a patch to monitor blood glucose levels through the skin, no one has developed a sensor capable of doing this accurately yet. But it would be a great Google X project and for all we know they have a team right now playing around with near infrared spectronomy, ultrasound, and many others types of sensors trying to improve upon the many failures in the past to make this technology work, accurately and consistently accurately. Google X projects are only announce when Google can no longer hide the project, so there could be many other other wearable projects at Google that we have yet to see or hear about.

  7. While I think your article is well written and well documented i think it misses the point..the point being that once the possibilities are shown by a company other companies often take that idea and run in a different direction with it, making a fortune and helping millions with their own twist. I think Google is both right and wrong in this contact lens thing, I would wear them, I am a type II, and even go thru the hassle of putting them in and taking them out if I didn’t have to test as often! If the lenses were able to be in for say 16 hours per day then I could put them in when I wake up, and test, and then wear thru my dinner and pre bed time testings before taking them off for the night. ie no more finger pricks for me!! Thus they WOULD make my life better, which is the initial point and they would then inspire other companies/researchers to say ‘what if’ and this could lead to some skin patch that we wear that does the same thing except over 3 days or even a week, sending the info to our smartphones for logging.

  8. Perhaps the problem here is really that institutions like universities are beholden to the drug companies etc for research grants and there is very little blue skies and pure research going on. If not Google then who? is a more pertinent question to ask. Diabetes research institutions, as with other disease related organisations, get a small percentage of the available world wide funds to spend on research; most of it is spent by drug companies etc. who expect a return on their investment (just like Google).

  9. Steve Silansky

    May I suggest that diabetics here do research on the benefits of coconut oil?

    The whole saturated-fat / coconut bad press was a very successful smear campaign by vegetable oil producers – because in fact coconut oil is high in medium-chain triglycerides, which are beneficial to the body.

    Coconut oil has such a wide range of benefits, including diabetes, that I think y’all should look further into it.



  10. David Fischer

    A quick point: contacts can correct some astigmatism.

    I found your introduction compelling: that contacts are known to be bad for diabetics and pursuing that shows Google simply isn’t paying attention to basic parameters. (I don’t know anything diabetes, so I’ll trust that’s correct.)

    I don’t understand the longer point about costs. You argue contacts are too expensive, but argue in favor of other micro-electronic devices that connect wirelessly to phones. Is this really less expensive? You assert it, but it’s not obvious, and not really supported in your essay.

    As a contacts wearer, I see the appeal of Google’s project. If they could report on my health, that would be a free benefit; no additional effort on my part. That brings it back around to your starting point: fundamentally wrong choice to report on diabetes, since a diabetic won’t be wearing contacts.

  11. Jim Hassinger

    Seriously, why the attachment to the eye? I don’t pretend to know what diabetics need, but why not develop a patch with bluetooth that you can leave on for a week and gives you constant feedback, and alarms, from a device in your pocket? It wouldn’t have the COOL factor? With a patch, you’d have to change locations, and you’d have to make the adhesive material non-allergenic, but this is much less of a problem than something you stick on your eye. I don’t have diabetes, so I don’t have any insights there, but what exactly is the attraction to the cyborg that Google has?

    • Jim:
      Current CGM technology with a 6mm electrode works just like this. I wear the site for 7-10 days and don’t notice it at all. It is invasive, but minimally, especially compared to earlier generations of the tech I’ve used. The handheld receiver (44x98x9mm) gives graphs, sounds alarms, etc. I agree that putting something in your eye better be pretty darn amazing compared to current techs for it to be worth it. But as many have said, it may be more about the cyborg experience for Google.

  12. CFGurney

    OM, I appreciate you writing about this from a personal perspective and there are a couple of issues that your post raises that I think point to a larger question in terms of healthcare. Specifically, your inference that contacts are bad for someone with diabetes and therefore this particular initiate is bad is to me is problematic.

    I’m writing from the perspective of someone who has Type 1 (for those of you not familiar with the confusing nomenclature – Type 1 is insulin dependent and is caused by genetics. Without daily insulin injections those with Type 1 will become quite ill and even die) and who has 3 extended family members also with Type 1. None of our primary care nor endocrinologists (including pediatric endocrinologists) have ever raised the concern you shared about corrective lens use.

    Does this mean your statements are wrong? Absolutely not! It does point out the extreme variance in standard of care and advice a patient will receive. So I’m concerned about a conclusion drawn from one’s own experience. And this case demonstrates that clearly: Om has one set of advice and my cohort of 4 has a different set and we can both draw completely different conclusions based on that.

    As for the Google initiative itself, I have the following opinions. First, sucessful noninvasive glucose measuring is a critical step towards the ultimate goal of a self-regulating insulin infusion system. Essentially an “artificial pancreas”. So it is important that different entities are working on different monitoring approaches. Second, we know from the history of technological development that often times innovations that seem to be a dead end actually provide the foundation for another innovation to become reality at some future point.

    Om is absolutely correct to point out that these high-tech, high-cost solutions do nothing for the majority of the world – simply because there is a massive barrier to adoption and the necessary support systems simply do not exist. The most promising solution in the long-term may be insulins that are encased in a microscopic structure that responds to chemical signals in the body itself and are released as needed.

    My opinion is that the Google initiative (while surprising) is welcome. Is it a solution in and of itself? I say no, but it is clearly making a contribution to the overall science.