Academic Innovators: Tanzeem Choudhury Pushing the Boundaries of Wearable Technology at Cornell

Tanzeem Choudhury
Photo courtesy of Cornell University College of Engineering

When market are moving as rapidly as the consumer and health care market for wearables are, often times market leadership and outperforming value creation depends on protectable innovation and know-how translated into mainstream products. Tanzeem Choudhury is an Associate Professor in Information Science at Cornell University, where she directs the People-Aware Computing group that develops mobile sensing systems for capturing, learning, and interpreting people’s context, activities and social networks. It is a place where this high level of innovation is undoubtedly occurring. In 2008, MIT Technology Review recognized her as one of the top 35 innovators under the age of 35. Choudhury has also been selected as a TED Fellow (2009) and is a recipient of the NSF CAREER award (2008). We had the opportunity to sit down with Professor Choudhury to have a candid discussion about her research, wearable technology, and her vision of the future in this space.

One of the recurring themes at both Inside Activity Tracking and extensively written about by Motion Technology Partners’ David Stern, is the affect new healthcare legislation is having on the potential for products and services in the wearables market that can cross over from consumer to healthcare.  This especially applies to mental health, and area that has seen very little venture funding activity and is ripe for innovation.  This is a particular sweet spot and focus area for Professor Choudhury, as she aims to change the way mental health is diagnosed and treated by creating novel wearables that continuously track mental well being and leverage existing mobile technology. “The phone sits on wherever a person keeps their phone in the home. But because it can communicate wirelessly, there could be smaller sensors in the form of a wristwatch or even in the form of jewelry. People are building sensors that could be in the form of rings, bracelets, or earrings,” she recently said in an interview with Earthsky.

One product of her research, StressSense, monitors users’ voices during phone calls and then monitors and detects stress. A pivotal feature in this product is the privacy sensitive barrier that processes the pitch and timbre of the users’ voices instead of recording the actual call. Features like this that protect users’ privacy are important for gaining popular acceptance from the masses.  “All of these technologies that reveal information about where you are, what you’re doing, have a privacy concern. The way we are trying to deal with this is to give users complete control. The phone is on them, they can choose what information they want to share, what information they want to record. If you don’t want to record anything about your social interaction, you can turn off the audio.”

In addition, she has co-authored various papers including Personal Informatics Can Be Stressful: Collecting, Reflecting, and Embedding Stress Data in Personal Informatics,” and Unobtrusive Sleep Monitoring using Smartphones.”

During our conversation, Choudhury stressed the critical importance of making data actionable. One instance making data actionable is MoodRythm. This application can provide actionable feedback to bipolar disease patients (and other types of users) instantly helping them restore biological rhythm. In the future, she plans to add a time specific MoodRythm based algorithm that learns from personalization.

Making data actionable is challenging, especially for different groups of users so Choudhury and her colleagues pioneered a way to train activity recognition process. Instead of treating all users in the same manner, they classified users into categories using crowd-sourced data labels. This approach groups similar users and can be applied to populations ranging from college students to heart patients. It allows personalized data given users’ demographics, while still maintaining feasible computational and algorithmic costs.

Professor Choudhury summarizes her work in this way: “I build computational systems that continuously capture and understand people’s lives and their behavior – what activities they’re engaged in, how they engage with others. And we’re interested in building these systems using computers we carry with us all the time, which are our cellphones. We want to make these tools available to everyone so that they have ways of keeping track and managing their health.”

In a fast moving market like wearables, it is academics like Tanzeem Choudhury that are providing the next generation of protectable innovations that will drive long term value creation in the market, especially with products and services that have the potential to cross over into the health care delivery systems.

, ,

Comments are closed.