On a quiet Wednesday night in April, an unusual group has assembled in a garage turned hacker studio nestled in a student-dominated neighborhood outside Boston. Those gathered here—mostly in their 20s or 30s and mostly male—are united by a deep interest in themselves. They have come to share the results of their latest self-experiments: monthlong tests of the Zeo, a consumer device designed to analyze sleep.
The group is part of a rapidly growing movement of fitness buffs, techno-geeks, and patients with chronic conditions who obsessively monitor various personal metrics. At the center of the movement is a loosely organized group known as the Quantified Self, whose members are driven by the idea that collecting detailed data can help them make better choices about their health and behavior. In meetings held all over the world, self-trackers discuss how they use a combination of traditional spreadsheets, an expanding selection of smart-phone apps, and various consumer and custom-built devices to monitor patterns of food intake, sleep, fatigue, mood, and heart rate.
Of course, self-tracking is not new. Many athletes have been meticulously monitoring personal metrics for decades. And some people with chronic conditions such as migraines, diabetes, and allergies have done the same in an effort to shed light on how daily habits may influence their symptoms. But new consumer tools have made self-tracking both simpler and more rigorous, generating reams of data that can be scrutinized for patterns and clues. The new devices, along with the increasing ease of sharing data with other users through social-networking sites, mean that more and more people are finding it useful to quantify their lives. The Zeo, a $199 device based on technology that until recently required the services of a trained technician, makes it easy for users to track their sleep cycles. The device consists of a soft headband with a fabric sensor that wirelessly transmits EEG data to a bedside monitor. A programmable alarm clock wakes the wearer at the optimal phase of sleep. And each night’s data can be uploaded to a computer, where users can study how their sleep is affected by environmental factors such as weather, light, and more.
Sanjiv Shah, a longtime insomniac who participates in the Boston group, believes that wearing orange-tinted glasses for several hours before bed makes it easier for him to fall asleep. (The theory is that the orange tint blocks blue light, which has been shown in both human and animal studies to influence circadian rhythms.) To quantify the effects, he used not only the Zeo but also a thumb-size device called the Fitbit, which incorporates an accelerometer that measures movement, and a camera trained on his bed to record his sleep for a month. His results: without the glasses, he took an average of 28 minutes to fall asleep, but with them he took only four.
See a gallery of self-tracking devices.
View a day in the life of editor Emily Singer as tracked by devices.
Kyle Machulis liberates data from self-tracking devices.
Dave Marvit demonstrates a new method for monitoring stress.
The experiment has an obvious flaw: Shah knows when he is wearing the glasses, and he believes they work, so the placebo effect could be responsible for their success. Matt Bianchi, a neurologist at Massachusetts General Hospital who spoke at the get-together, says no large-scale studies have shown that orange glasses improve sleep. (By the end of the evening, plans for a group experiment to test the technique were under way.) But self-trackers say the idea of reproducing the results in scientific tests misses the point. The glasses clearly work for Shah. And an $8 pair of plastic glasses is certainly preferable to sleep drugs as a way to gain that benefit.
As Gary Wolf, a journalist and cofounder of the Quantified Self, puts it, “It’s a trial that begins with one very important person: yourself.”
Over Memorial Day weekend, approximately 400 hackers, programmers, designers, engineers, and health-care professionals gathered at the Computer History Museum, in the tech mecca of Mountain View, California, for the first annual Quantified Self conference. Attendees showed off fitness monitors, apps to gather and display data, and even a set of sticker sensors with embedded accelerometers to detect movement, which are designed to be stuck on toothbrushes, water bottles, or a dog’s leash.
Standing out in the crowd was Alex Gilman, a researcher at Fujitsu Laboratories of America, who ambled down the main hall with a bag slung over his shoulder. A tangle of wires sprouting from it led to monitors on different parts of his body: a white plastic ear clip, which measured his blood oxygen levels; a blood pressure cuff around his arm; and a combination heart rate monitor, EKG, temperature gauge, and accelerometer strapped to his chest. The bag itself held a prototype device designed to gather and synchronize the data from all those sensors and analyze it with the help of new algorithms.
The devices are a taste of the not-so-distant future, when the monitoring tools now typical of a hospital’s intensive-care unit will be transformed into wearable gadgets that are unobtrusive and effortless to use. Gilman’s chest strap is from a company called Zephyr, which has traditionally made equipment to track the physiology of military personnel and emergency workers during stressful situations. Zephyr is developing simplified consumer versions of its products; the latest one tracks motion, heart rate, and respiration and includes software to assess the user’s fitness level. The blood pressure cuff and the clip to measure blood oxygen, which come from different manufacturers, are still too bulky to incorporate into consumer devices. The data, however, can be integrated into a single online dashboard with the help of Zephyr software.
The new generation of devices rely on inexpensive, low-power wireless transceivers that can automatically send data to the wearer’s cell phone or computer. Compared with the limited snapshot of health that is captured during an annual visit to the doctor’s office, these tools and techniques could reveal the measures of someone’s health “in context, and with a much richer resolution,” says Paul Tarini, a senior program officer at the Robert Wood Johnson Foundation, which donated $64,000 to help the Quantified Self group create a guide to self-tracking tools.
Wearable sensors that measure vital signs such as blood pressure and heart rhythm around the clock could lead to applications we haven’t thought of yet, says cardiologist Eric Topol, director of the Scripps Institute for Translational Medicine. Perhaps they could help people get a handle on health concerns such as headaches or fatigue, which don’t qualify as diseases but can have a huge effect on quality of life. “People often get light-headed in daily activities,” Topol says. “Is that symptom linked to an abnormal heart rhythm? Are headaches linked to abnormally high blood pressure?”
At the Quantified Self conference, the museum’s walls were lined with posters describing personalized dashboards and other apps for collecting and aggregating data. But tools for analyzing the data are much harder to come by. That’s why Gilman and collaborators at Fujitsu built the device in his shoulder bag. One application they’ve developed is a way to use time-stamped raw data from wearable blood pressure monitors to make sure readings aren’t taken when the user is active, which can yield misleading results. The new software tells the device to calculate blood pressure only when another monitor reveals that the wearer has been sitting still, as indicated by a steady heart rate.
The Fujitsu researchers are especially excited about using information collected instantaneously from the EKG to calculate heart rate variability, a well-validated indicator of stress. Taking a reading with previous instruments requires the subject to stand or sit still for several minutes, says Dave Marvit, vice president of the Connected Information Innovation Center at Fujitsu Laboratories of America. That makes it difficult to monitor stress as people go about their daily lives. Recently, Marvit tested the algorithm while moving naturally during an online game of speed chess. A graph charting his stress level in real time showed a spike as he contemplated a move to throw off his opponent’s strategy, and a drop as he relaxed with the satisfaction of winning the game. “Seeing the physiological consequences of the mental state makes it much more real,” he says. “It’s much more interesting to measure stress while you’re living your life than when you’re standing still.”Better Medicine
Perhaps the most interesting consequences of the self-tracking movement will come when its adherents merge their findings into databases. The Zeo, for example, gives its users the option of making anonymized data available for research; the result is a database orders of magnitude larger than any other repository of information on sleep stages. Given that the vast majority of our knowledge about sleep—including the idea that eight hours is optimal—comes from highly controlled studies, this type of database could help to redefine healthy sleep behavior. Sleep patterns may be much more variable than is currently thought. Zeo researchers have already found that women get less REM sleep than men and are now analyzing whether the effect of aging on sleep differs by sex. The database is obviously biased, given that it is limited to people who bought the Zeo; those people are mostly men, with ample income and presumably some sleep-related concerns. But the sample is still probably at least as diverse as the population of the typical sleep study. Bianchi, who studies a number of sleep disorders and is developing his own home sleep-tracking tool, says an individualized approach to the study of sleep may help shed light on its complexities. “I have become skeptical of sleep science and clinical trials, so I am very interested in what individuals have to say,” he says.
There are plenty of reasons to believe that people sharing data about themselves can produce powerful medical insights. Patient groups formed around specific diseases have been among the first to recognize the benefits to be derived from aggregating such information and sharing it.
In 2004, Alexandra Carmichael, a longtime migraine sufferer, identified dairy and gluten as the triggers for her headaches after extensively tracking her pain and correlating it with diet and other factors. Hoping to help others find relief from chronic pain, she founded CureTogether, a social-networking site where patients can list their symptoms, the treatments they have tried, and the results they’ve observed. Aggregating and analyzing the information has begun to reveal broader trends. For example, Carmichael and other members of CureTogether found evidence that people who experience vertigo with their migraines are four times more likely to see their pain increase than decrease if they take Imitrex, a migraine medication that constricts blood vessels. In the near term, new members to the site can use this information to help decide which treatments to try first. In the longer term, scientists studying migraines could explore this link more formally.
Such studies obviously lack the rigor of clinical trials, but they have their own advantages. Clinical trials usually impose stringent criteria, excluding people who have conditions or take medications other than the one being studied. But self-tracking studies often include such people, so their pool of participants may better reflect the actual patient population.
PatientsLikeMe, a social-networking site that provides users with tools to track their health status and communicate with other patients, has gathered a wealth of data on its 105,000 members. (The site makes money by anonymizing the data and selling it to pharmaceutical companies and other customers.) In 2008, after a small Italian study published in the Proceedings of the National Academy of Sciences suggested that lithium could delay the progression of ALS, or Lou Gehrig’s disease, a small group of the ALS patients on PatientsLikeMe began taking the drug, and the company rolled out a number of tools to help them track their symptoms, their respiratory capacity, their dosage and blood levels of lithium, and any side effects they observed. Because the patients had collected so much data on themselves before starting the drug, researchers could analyze how their symptoms changed in the 12 months before they began taking it as well studying any changes that came after—something that’s not possible in the typical clinical trial. The company published a study based on its data in April. The drug, unfortunately, was found to have no effect.
The growing availability of new monitoring devices and the increasing sophistication of social networks promise to make self-tracking much more powerful than it was when patients who wanted to monitor their own conditions were limited to standard spreadsheets and daily logs. “We see the potential to change the power dynamics in health care,” says the Robert Wood Johnson Foundation’s Tarini. People could take far more responsibility for monitoring their own health. The concept of personalized medicine could change as well; rather than relying on pharmaceutical companies that have little incentive to individualize treatments, patients could simply try different interventions and record how their physiological signs and symptoms change in response.
Of course, it remains to be seen whether a movement rooted in individual experimentation can scale up in ways that will affect public health. Even if it has the potential to do so, incorporating findings of this type into the health-care system is likely to be an enormous challenge. When you start with information from a study of one person, says Tarini, “the system doesn’t have a way of determining what should be explored further.” And because many of the new tools for tracking are aimed at consumers rather than the medical market, they have not undergone the rigorous testing required of medical devices. Still, Tarini is optimistic. “We have the opportunity to explore a whole new set of information,” he says. “That has the potential to teach us a lot about health care.”
The Big Payoff
The early adopters of self-tracking are often odd. In one breakout session at the May conference, a group earnestly discussed the results of their experiments. Standing on one leg for eight minutes a day helped one person sleep. Eating butter helped another think better. One had logged every line of computer code he’d written for a decade. But there is a far more pragmatic side to the movement, too. Across the building from the butter eater, another group, made up mostly of entrepreneurs, discussed business models for selling self-tracking apps and devices.
The favored strategy of the moment is to weave together self-tracking tools with social networks and gaming, using the lessons of behavioral economics to keep users motivated enough to meet any health goals they’ve set for themselves. “We want to create an engaging device that makes people want to make better health choices,” says Julie Wilner, product director at Basis, a startup developing a new watch laden with sensors. “We do that by tracking data and showing it on the Web and on mobile devices, and by sharing it with friends.”
Withings, a French company that makes wireless scales and blood pressure monitors, gives users the option of tweeting their weight, with the goal of adding social pressure to make people stick to a diet. (Only a small percentage of users employ that feature, and the vast majority of them are men. The company is also experimenting with delaying readings from the scale. That way, the user may be less likely to get discouraged on a bad day and stop weighing herself.) And Green Goose, the startup developing the sensor-equipped stickers for household objects, plans to create a game based on personal health goals, awarding points whenever the user walks the dog or takes vitamins.
Yet even as startups plot how to profit from the trend, the people behind the self-tracking movement have a very different mind-set—and very different goals. “I find that the most interesting tools are those that give us the chance to reflect on who we are,” says Wolf, the Quantified Self founder. The problems self-tracking tries to solve, he says, are important to everyone’s life: “How to eat, how to sleep, how to learn, how to work, how to be happy.”
Emily Singer is Technology Review ’s senior editor for biomedicine.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Driving companywide efficiencies with AI
Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.