Monday, November 26, 2012

Close your eyes and tap your heels

How can you tell an extroverted engineer? When he talks to you, he looks at your shoes instead of his own. And now there is a good reason to do it - as new "No place like home" GPS shoes will be pointing to where the person is going. And displaying the progress bar - marking the beginning of the journey with one red light and indicating successful arrival to the place of destination with a green light on the top of the right toes. A red light on the other shoe will  display the correct direction to walk, illuminating on the circle of LEDs like an arrow of the compass. How will the shoes know where to go? By consulting the map uploaded via USB and its own GPS receivers, wirelessly communicating with each other. For future models, you could probably set up WiFi to let your shoes download more information, talk with other people's shoes and modify your route on the go.
So your footware might need its own network access, like agent Maxwell Smart's left shoe with a mobile subscription plan.
The "No place like home" shoes are built around two microcontrollers called Arduinos: A magnet in the right shoe and sensor in the left shoe communicate with each other and with the GPS antenna in the red tag at the back. Clicking the heels starts the GPS. So all you need to do is to close your eyes and tap your heels together. And there will be no need to follow the yellow brick road or say the magic words.

The smart shoes - designed by artist Dominic Wilcox and custom-made by Stamp Shoes might be a bit costly: £1,100 (about $1,750). A bit less sophisticated Aetrex Navistar GPS shoes developed for sufferers of Altzheimer's disease and dementia cost $299.99, and come with two monthly subscription plans - a basic 30 minute tracking plan, which reports every 30 minutes ($34.99) and for an additional $5 per month a premier 10 minute tracking plan. Nike was offering their own GPS footware too, for fitness enthusiasts, but decided that it's cheaper to use iPhone's location sensor to figure distance and serve as a pedometer.

Yet, sensors in high-tech shoes could be helpful. For example, they could detect if their owner is tired or exhausted. Fatigue Monitoring System (FAMOS, recently developed and tested in patients with multiple sclerosis (MS) and healthy individuals) continuously measures motions of feet, in addition to electrocardiogram, body-skin temperature and electromyogram. And the system can reliably distinguish the symptoms of fatigue. The shoe sensors could provide a wealth of information about motion and assess such things as the risk of falling. And this information can be combined with data collected through other channels.  Aurametrix, for example, can determine how food, air quality, the weather and various activities affect energy levels and generate suggestions on what to do - at the right time and right place. Systems like Aurametrix could eventually integrate our observations with data coming from smart objects such as shoes and heart monitors, to speed up not only walking but also the understanding of the human body, for a healthier world.


Yu F, Bilberg A, Stenager E, Rabotti C, Zhang B, & Mischi M (2012). A wireless body measurement system to study fatigue in multiple sclerosis. Physiological measurement, 33 (12), 2033-2048 PMID: 23151461

Marschollek, M., Rehwald, A., Wolf, K., Gietzelt, M., Nemitz, G., zu Schwabedissen, H., & Schulze, M. (2011). Sensors vs. experts - A performance comparison of sensor-based fall risk assessment vs. conventional assessment in a sample of geriatric patients BMC Medical Informatics and Decision Making, 11 (1) DOI: 10.1186/1472-6947-11-48

Tuesday, September 25, 2012

Mirror, Mirror on the wall, Am I healthy after all?

Health gadgets continue to evolve in many forms and shapes - from something that fits in your pocket to something that is wearable or walkable. Everyday objects are turning into "Smart objects", building the foundation for the next version of the Internet. And it's not all smoke and mirrors. So let's talk about mirrors.

Fairy tales and science fiction stories often pave the way to real world technology. Magic mirrors have been used in Snow White and Harry Potter's world. Now you can get one, too - manufactured by a Hong Kong company James Law Cybertecture International.

Cybertecture mirror can tell you about the weather or your last weight readings reported by the scales. It can show you a TV channel, let you browse Facebook or twitter and help you to exercise. Impressive, yet so much more is yet to come.

Could a mirror tell us how healthy we are? For example, could it measure our heart rate at a distance? Sure, it could. And it has already been demonstrated as a concept prototype (Cardiocam, MIT media labs, Poh et al. 2010), although the designer is now focusing on mobile devices (check his company Cardiio).

What other health metrics could be performed by the mirror during your regular morning hygiene routine? If a camera can measure minute changes in the color of your face to determine your heart rate, it could also measure your facial expressions and emotions or perform observational analysis  - the first of four methods of diagnosis performed by traditional Chinese medicine.

Prototypes for computerized facial diagnostic systems already have been developed. One recent study, for example, (Li et al 2012) analyzes lips. The software segments lips from the rest of the face and extracts color, texture and shape features. Special supervised learning algorithms are then able to classify lips as deep-red, purple, red or pale and make inferences related to energy levels and circulation.

Health management applications will not be limited to smartphones or smart homes. All objects in our lives will gradually become "smarter." Mobile phones can already manage vacuum cleaners and thermostats. Refrigerators can tweet, check Google calendars, download recipes, play tunes and alert us about food spoilage. Mirrors can monitor our weight and exercise. There is still more emphasis these days on technological wizardry than on actual benefits, but data collected through different channels can be brought together for analysis and context. Aurametrix, for example, can find common ground to make connections between specific symptoms and weather, air quality, food, weight, heart rate, exercise, work-related and personal care activities - and generate suggestions on what could be affecting the symptoms, and in which amounts and combinations. Systems like Aurametrix could eventually integrate our observations with data coming from smart objects surrounding us and then generate valuable insights. And perhaps, one day, we won't regard the mirror on the wall nagging us about losing weight or commenting on the bags under our eyes as invasion of privacy. Let's build the future piece by piece - and they will come.


Poh MZ, McDuff DJ, & Picard RW (2010). Non-contact, automated cardiac pulse measurements using video imaging and blind source separation. Optics express, 18 (10), 10762-74 PMID: 20588929

Li F, Zhao C, Xia Z, Wang Y, Zhou X, & Li GZ (2012). Computer-assisted lip diagnosis on traditional Chinese medicine using multi-class support vector machines. BMC complementary and alternative medicine, 12 (1) PMID: 22898352

Littlewort, G., Whitehill, J., Wu, T., Fasel, I.R., Frank, M., Movellan, J.R., Bartlett, M.S. (2011) The Computer Expression Recognition Toolbox (CERT). Proceedings of the 9th IEEE Conference on Automatic Face and Gesture Recognition. 

Saturday, June 23, 2012

Cars That Care

Health technology of the future promises an easy life with no interruption in your daily activities. For example, information about your health could be collected while you're driving. A car is already viewed as a health platform and wellness coach by leading manufacturers. How would this work?

To begin with, by measuring our heart rate. The electrocardiographic (ECG) seat built by Ford is based on studies of sensors in beds for intensive care units. Unlike traditional monitoring systems, it does not require attaching electrodes to the skin and can measure signals through relatively thin cloth. Toyota's response to Ford's seat is an ECG-sensing steering wheel.

Regardless of what type of system incorporates the sensors, clever algorithmic science is needed to account for artifacts caused by lateral movements. Wartzek and colleagues showed that unobtrusive and reliable measurements of heart rate are indeed possible during driving by identifying useful intervals in heavily distorted ECG signals (which is easier on the highway than in city traffic). Moreover, data from ECG, GPS and optical devices could  be combined  with other measurements though, as Doherty and colleagues showed, significant data processing issues still remain. Companies like Aurametrix are addressing the problem of noisy environments with innovative approaches. 

So in a few years cars will start to take care of us. We need to polish up the sensor and data processing technologies and also manage the chemicals added to the interior of the car--including those contributing to the "new car" smell.  Unhealthy particles in some automobile interiors already exceed US EPA standards, especially in heavy traffic situations (although bicyclists and pedestrians have their own problems). The latest report by ranks over 200 of the most popular models based on chemical-emitting steering wheels, dashboards, armrests and seats. As the table shows, stylish and sporty models are at the bottom of the list.  

There are many reasons to believe these problems will be addressed. If so, we can look forward to a future with safely built in to the systems we use in our every day lives.


Wartzek T, Eilebrecht B, Lem J, Lindner HJ, Leonhardt S, & Walter M (2011). ECG on the road: robust and unobtrusive estimation of heart rate. IEEE transactions on bio-medical engineering, 58 (11), 3112-20 PMID: 21824839

Doherty ST, & Oh P (2012). A multi-sensor monitoring system of human physiology and daily activities. Telemedicine journal and e-health : the official journal of the American Telemedicine Association, 18 (3), 185-92 PMID: 22480300
blockquote { margin:1em 20px; background: #dfdfdf; padding: 8px 8px 8px 8px; font-style: italic; }