Faculty of Engineering

Wearable devices already constitute a major development in consumer technologies, with everything from watches, wristbands, glasses and even rings being turned into a smart device. One key component of these devices is the ability to measure - be it your heart rate, the amount of sleep you got last night, or how many steps you took on a given day. KUAS’ Dr. Zilu Liang, whose research focuses on combining wearable technology with deep learning, has her sights set on something even bigger: she believes the future of wearables will be go beyond simply measuring.

As part of the “Quantified Self” movement, people have become more and more interested in improving their lifestyle based on measuring themselves. Usually, people start doing this with a goal in mind. One example would be trying to understand one’s sleep patterns better in an attempt to get a better night’s sleep. What current technology can do is provide quantifiable understanding, e.g. average hours of sleep on a weekday. However, there remains great potential in how to analyze and utilize this data.

First, another benefit of the spread of wearables is the ability to measure over a long period of time, as these devices are usually used daily. This can be then combined with deep-learning algorithms, making it possible to go beyond simply measuring and into the provision of analysis and solutions. The goal is to provide data-based, personalized ways to improve one’s lifestyle. Going by the previous example of sleep monitoring, the corresponding app could then give you based on your trends and behavior, potentially eclipsing the more generalized advice offered by professionals. However, Dr. Liang wants to go even further.

She believes in creating a future where we can provide analysis on things now considered abstract, such as stress or depression. By using the power of wearables and combining the various measurable they provide, this could potentially allow us to come up with a quantifiable definition of such terms, and thus help us understand what metrics, trends and precursors to look for, allowing for forecasting and early prevention.