New Study Predict How Much A Person Will Eat From Smartphone Data

Increasingly smartphones are being used be people to log various aspects of their life. Recently my group published a paper in which we analyzed data from individuals’ smartphones, to see whether we could model their dietary patterns. Rather than compare across people, we focused on the patterns that exist for specific individuals. We examined whether the portion size for each person’s meal tended to followed a routine based on time of day, whether the portion size was related to how much physical activity they engaged in before the meal, or whether the their “food environment” — the quantity, diversity, and types of food establishments around them — influenced the portion size. We also looked at the person’s mood to see it influenced their meals. All of these data were collected from a smartphone app we developed (voice-annotated video recording of meals, accelerometry for physical activity, GPS for food environments, and occasional prompts to ask about mood).

What we found was fascinating. In some cases, certain individuals exhibited fairly routine eating patterns, while for other individuals, exercise before the meal was an important factor in determining their portion sizes. However, we were surprised to find that across all of the subjects, the food environment tended to be an important factor for explaining portion sizes.

This study was based on data from a fairly small group of test subjects who were only monitored over a short period of time, but we now have hundreds of gigabytes of similar data for which we will be doing the same sort of analyses.

The work provides an early glimpse of our ability to integrate various different sorts of sensor, self-report data, contextual data to model specific individuals behavioral patterns.  Moving forward, we are interested in evaluating how well these models work for prediction, and eventual use for tailoring behavioral interventions to specific individuals to help them live healthier lives.

This study was recently published in the journal PLoS ONE: http://dx.doi.org/10.1371/journal.pone.0153085

 

My NIEHS Webinar: Environmental Sensors, Citizen Science, and Quantified Self


On April 5, I gave a webinar “Sensor Technologies for Improving Environmental Health: Juxtaposing the Citizen Science and Quantified Self Movements”.  Thanks to all who attended, and especially to those who provided questions/feedback on my group’s work.  Here’s the abstract:

In recent years, numerous sensor technologies have been developed that offer the ability to collect detailed data on environmental conditions and their impact on human health. These technologies will likely change how communities and individuals access environmental health information, and the amount of data that are available for improved decision-making.

An example of the potential impact of emerging sensor technology can be seen through the development of low-cost direct-reading air pollution monitors many of which are now commercially available.  While researchers continue to conduct studies to answer the fundamental question of “how well do these new devices perform?,” perhaps the more intriguing question is “if useful air pollution data could be obtained from a device that many people could afford, how would this change our understanding of air pollution-related health?”

This webinar will discuss multiple answers to this question, including how community groups, Citizen Science, epidemiologic researchers and individuals may benefit from these new sensors.

One answer to the above question is that community-based environmental groups may have better access to monitoring technologies to document environmental injustices. In some respects, this is not entirely new, as community groups have for many years documented their local knowledge of air pollution levels in much more detail than what was possible through government monitoring efforts. The difference now, is that monitors are more readily available for these groups to collect their own objective measurements. Because of the low cost of each monitor, it is not implausible to imagine entire communities blanketed with a high density of air pollution sensors. And, in fact, NIEHS-funded research is demonstrating that such community-engaged monitoring networks are possible.

Slightly different from the community-initiated and led research described above, new air pollution technologies are enabling a new form of environmental research within our communities, called Citizen Science. While there are different models for conducting Citizen Science, the more intriguing examples are those that are organized over the Internet, involve many individuals who work together to crowdsource data, and result in massive amounts of data that are shared openly. In some cases, technology-savvy Citizen Science leaders are developing and sharing their designs for new monitors, providing proof of concept for how measurements can be made with low-cost sensors.

Another possible answer to the above question is that an increasing number of environmental epidemiology studies may use sensors to conduct personal exposure assessments. Sensors are not only getting cheaper, but also battery-powered and are getting smaller in size, making them increasingly practical for use in a variety of cohort studies. An exciting example of NIH support for this is the new Pediatric Research Using Integrated Sensor Monitoring Systems (PRISMS) program, within which various research groups are developing new wearable sensors that can measure environmental exposures that can be related to health symptoms for future children’s asthma studies. An exciting aspect of the PRISMS program is the recognition that future sensors will likely need to be network-enabled, which would provide more immediate data from research subjects, as well as also enable more immediate feedback to research subjects.

While NIH programs like PRISMS are fostering future sensor technologies for epidemiologic research, the private sector is also commercializing air pollution monitors for the consumer health and wellness market. Smart technologies (e.g., smartphone apps, smart watches, fitness trackers, GPS loggers, etc.) used to be primarily marketed to the Quantified Self movement – individuals who use devices and data to track and optimize various aspects of their life – air pollution monitors being just one of latest devices that such an individual may want to use. But, there is a large group of individuals who have pre-existing health conditions and may be susceptible to air pollution exposures, which may be interested in understanding their air pollution exposures by either having a household or wearable monitor. While these individuals may be the greatest market for these new air pollution monitors, it remains unclear how people will respond to personalized air pollution exposure data (e.g., what are the best ways to communicate individual-based air pollution exposures and risks?), or whether people have practical ways to manage their exposures.

In summary, the recent developments in low-cost air pollution monitoring devices illustrate various opportunities for improving environmental health through sensor technologies. The benefits to traditional community-based and epidemiologic research studies are somewhat clear, with new monitoring devices potentially providing data for more persons, places, and times than previously possible. Less clear, but no less exciting because of the reach and numbers of people potentially involved, are the novel ways in which new air pollution monitors are being adopted into Citizen Science and consumer health and wellness applications.

More, and an eventual link to Youtube can be found here:
https://tools.niehs.nih.gov/conference/exposome_webinar/