Although land mines and breast cancer would seem to have little in common, Ben Ellwood ’05 (Orwigsburg, Pa.) used technology that could dramatically alter the method for detecting the two.
Ellwood, an electrical and computer engineering major, worked on a new detection process by using a computer program called a support vector machine.
“The support vector machine is a computer program that is able to learn,” says Ellwood, who collaborated with Ismail Jouny, professor and head of electrical and computer engineering. “When you give it information, it forms information in its own way, called support vectors.”
Ellwood and Jouny partnered in the research through Lafayette’s distinctive EXCEL Scholars program, in which students conduct research with faculty while earning a stipend. The program has helped to make Lafayette a national leader in undergraduate research. Many of the more than 160 students who participate each year share their work through articles in academic journals and/or conference presentations.
The support vector method is more accurate than other techniques for detecting cancers and land mines, which can often be misleading, Ellwood says.
For example, using a metal detector is often futile because many mines have just trace amounts of metal. Obviously, in life-or-death situations such as developing cancer or stepping on a land mine, the more accurate the test to find a lethal object, the better.
“It’s a method that looks at existing data and draws the line between what would be considered the lethal object and the benign object — it finds those invisible lines in space that help us to distinguish between those signatures,” says Jouny.
“If I feed it 10 images where a mine was present and 10 where it was not, it might pick out features from the 10 that would best describe when the mine is present,” Ellwood says.
It not only saves time, but lives also, he adds.
Earlier this year, the international publication for members of the Institute of Electrical and Electronics Engineers cited Jouny’s research on landmine detection in an article about developments in the field.
The project had numerous major implications for research and provided Ellwood with a bird’s-eye view of the life of a researcher.
“It helps him in many ways,” Jouny says. “It helps him get a glimpse of what goes on in graduate school and to decide whether he wants to pursue that or not — it could encourage him. It also shows him what goes on in research situations. For instance, much of his work [was] exactly what goes on in the defense industry. It also gives him a work ethic and patience for sitting for eight hours, and as far as dealing with computers, this is an experience that benefits him in any job.”
What the student determined this summer will be recorded as a contribution to the body of literature on mammography and land mine detection, he adds.
In the short term, just being part of a project that has immediate impact can be motivating.
“Students are excited about it because they can relate to it — it’s not some abstract concept that they may relate to later,” Jouny says. “In this one, they see the concept immediately.”
Not only did Ellwood get the thrill of instant gratification, he used technology not many before him had touched.
“He got to see state-of-the-art signal processing and pattern detection technologies, he used advanced software, and he got to work on a real application,” adds Jouny.
While Ellwood says that knowing more about signal detection would have made his summer project easier, he proved to himself that his previous coursework had real value.
“It allowed me to take what I learned last semester and build upon it,” he notes.
Along the way, he picked up a life-long lesson.
“Research is not all glory,” he says. “It’s a lot of hard work and a lot of hours in lab. But as long as you concentrate and can find some meaningful results, you can help other people in the world.”
A graduate of Blue Mountain High School, Ellwood is a member of the Delta Epsilon fraternity and plays intramural sports.
As a national leader in undergraduate research, Lafayette sends one of the largest contingents to the National Conference on Undergraduate Research each year. Forty-two students were accepted to present their work at the last annual conference in April.