Engineering for Human Technology Interface

Engineering is about solving problems. Whether it's creating a computer language with children's building blocks, helping teachers develop hands-on engineering curricula, or managing a new product line for a biotech company, Tufts graduate students are training the problem solvers of tomorrow and getting the real-world skills to be the engineering leaders of today. School of Engineering researchers are working to make computers smarter, too—better able to learn from experience. This could help scientists and doctors make the most of their data or provide users in general with an interface that automatically adjusts to suit their needs.

Center for Engineering Education and Outreach

In a technology-saturated society confronted by technical challenges such as an addiction to fossil fuels, engineering literacy is increasingly critical. Enter the Tufts Center for Engineering Education and Outreach (CEEO), where pedagogic research, instructional product design, classroom outreach, and workshops improve the teaching of engineering and spark more interest in science and math. Research at the CEEO is driven by students earning doctorates in math, science, technology, and engineering (MSTE) education.

Recent graduate Morgan Hynes developed a middle-school engineering curriculum using a robotics lab developed by LEGO, Tufts, and National Instruments. In another dissertation project, doctoral student Brian Gravel studied how having kids make stop-action movies reinforces their understanding of science and engineering. "Our mission is to motivate people of all ages to understand math, science, and engineering through hands-on, open-ended engineering projects," says CEEO director and mechanical engineering professor Chris Rogers.

One of the core programs at CEEO is the Student Teacher Outreach Mentorship Program (STOMP), in which undergraduate and graduate students work with teachers over several years to incorporate the math and science already in the teachers' curricula into interactive engineering projects, such as using static electricity to create mini-lightning bolts and making ice cream without an ice-cream maker.

Machine Learning

The Machine Learning Group, led by Professor Roni Khardon and Visiting Assistant Professor Liping Liu, teams up on interdisciplinary projects with scientists, engineers, and doctors from various fields who need help sifting, sorting, and mining information. They develop algorithms that allow a computer to learn from experience by recognizing complex patterns and "rules of thumb" in massive data sets.

In one application, the team worked with astronomers to detect anomalies in star-light measurements captured over time by huge telescopes to aid in the discovery of unusual astronomical phenomena. In another project, they worked with the evidence-based medicine group at the Tufts School of Medicine to automate the process of sorting through thousands of journal abstracts for relevant research.

Human-Computer Interaction

As a graduate student, Mike Horn was in an elementary school on a fellowship, watching a math teacher whose curriculum included some computer programming to help the kids learn geometry. But the teacher just skipped over it and went on to the next lesson, which wasn't surprising to Horn. "He had about 25 kids and he had a handful of aging desktop computers," Horn says. "It made me think, Is there a way for the kids to learn programming without being tethered to a computer?"

So Horn, who earned his Ph.D. in computer science at Tufts in 2009, developed a tangible programming language while working in Tufts' Human Computer Interaction lab. Instead of grappling with complicated coding syntax, kids fit together wooden blocks containing simple commands for a robot, such as "forward" or "spin around." The computer then snaps a picture of the assembled blocks, which have barcodelike symbols on top that feed the robot its program.

The Human Computer Interaction Lab is led by computer science professor Robert Jacob and the belief that if our digital devices could learn a little bit about us, then they could subtly alter their interfaces to be more efficient and user-friendly.

In recent research, Jacob and his graduate students are teaming up with biomedical engineer Sergio Fantini to give computer interfaces the ability to read our minds—if only a little bit. They're using non-invasive techniques to measure mental workload and emotional activation in the brains of computer users. At the same time, they're designing interfaces that can adjust by, say, highlighting only important details on a computer screen or gradually fading extraneous windows when signaled.

Cyber-Enabled Chemical Models

Pressing problems in biotechnology and biomedicine can be solved using computational, mathematical, and statistical methods. Chemical engineering professor Kyongbum Lee works with computer science professor Soha Hassoun to create metabolic models that could predict potentially harmful side effects of chemicals and discover targets for drug development. Chemical engineering professor Hyunmin Yi teams up with computer science professor Diane Souvaine to develop models for functional nano-scale architecture manufacturing.