Almost a decade ago, Matthew O'Donnell, then a General Electric (GE) scientist, started thinking about an ultrasound device to help direct the navigation of a catheter inside the tiny arteries of the heart. Soon after, O'Donnell learned that a small company called Endosonics was developing just such an imaging system. "I heard they'd built a crude system," he recalls. "The pictures were absolutely horrible, but I thought we [GE] could do it right."

But GE didn't want to pursue the project, so O'Donnell set aside his ideas until 1990, when he joined the University of Michigan College of Engineering faculty. Once in Ann Arbor, O'Donnell was free to pursue this research interest. Eventually, he joined forces with Endosonics in what has proven to be a successful collaboration that continues today.

Endosonics' ultrasound-guided catheter is used in hospitals for an artery-opening procedure called balloon angioplasty used in treating arteriosclerosis. The success of the product is due in part to improvements based on research in O'Donnell's laboratory.

"I think this is an ideal example of a university-industry collaboration," he explains. "We had imaging knowledge that they [Endosonics] didn't have, but they have the expertise with building these devices [ultrasound-guided catheters] with elegance-that are miniature, safe, and FDA approved. Even if we could build one, it would be very difficult."

O'Donnell has a personal consulting agreement with Endosonics, which holds all of the patents for the catheter technology and takes complete responsibility for constructing the device. This arrangement of shared expertise and a clear division of tasks has provided great benefit to the company and UM research, O'Donnell believes.

"After being in industry for ten years, it seems to me that for technology transfer to succeed, the exchange of knowledge needs to go two ways-from academic scientist to private company and the other way," he says.

"In many university-industry collaborations, the time window when collaboration is most profitable can be short, unless there is a basis for continuing mutual benefit. In our case, we each get what we need-they've got new products; we've got grants to do the studies that provide fundamental knowledge for continually improving their products."

When O'Donnell first took up this project in 1990, it was another story. He remembers writing a paper about an ultrasound-guided catheter system-not based on any experiments, but to explore some theoretical ideas and problems to solve.

"The Endosonics people didn't appreciate the article," he says. The company thought O'Donnell might be infringing on its proprietary interests. A meeting with company scientists, however, changed their view of O'Donnell from threat to partner. "After we started collaborating, they decided to build a whole new machine," he says.

O'Donnell is happy to see aspects of all of his research find its way into medical practice. "I wouldn't want to work on a project unless I thought it would someday be used," he says. "Working with industry helps guarantee that will happen."