Nearly all of General Electric Healthcare's software is developed at one of two sites: its Milwaukee headquarters or its Bangalore, India development center.

Now for the first time GE has signed a landmark agreement to develop products with hospital researchers at St. Joseph's Hospital and Medical Center in Phoenix. Leading the team of enthusiastic physicians and two GE employees is a St. Joseph's senior staff scientist, Dr. James Pipe, who will seek to improve imaging technology for treating and diagnosing patients.

Pipe and St. Joseph's have a long history with GE. Pipe's predecessor at the hospital, Paul Keller, worked extensively with the company, and Pipe, after graduating from University of Michigan with a bachelor's in electrical engineering, did two summer internships with GE.

What is unique about this latest hospital-industry alliance is that, for first time, GE has entered into a financial relationship with St. Joseph's in which the researchers will collaborate with computer technicians and then get paid royalties for the work they do.

"St. Joe's will get a portion of the sales from products sold by GE with our technology," Pipe said.

For Pipe, a soft-spoken bioengineer who once considered entering the field of nuclear power, this agreement marks both the capstone and the beginning of an exciting time that began in 1999 when St. Joseph's recruited him from Wayne State University.

While at Wayne State, Pipe developed a technology called PROPELLER (Periodically Rotated Overlapping ParallEL Lines with Enhanced Reconstruction) that helped vastly improve magnetic resonance imaging by lessening the requirement that patients hold perfectly still during scans.

Magnetic resonance imaging scans require that people who are scanned remain motionless for as long as 45 minutes. For some people this is not a problem, but for other--such as children, or people with Parkinson's--lying still is nearly impossible.

Pipe published the technology, but the work went nowhere until he arrived at St. Joseph's in 1999 at the age of 34, and a flurry of activity began to take place. The physicians there were excited about the prospects of clearer images regardless of human motion, and they wanted to work with him to make the technology a reality.

Just seven years later, PROPELLER has been incorporated into all GE scanners, and with the new agreement between the hospital and the company, Pipe hopes it will get even better.

"I really want to see this grow and become successful, and I don't mean successful financially," Pipe said. "I really want us to have a positive impact on how people are scanned in MR."

To learn more about Pipe's work in MR and his vision for the future of imaging, we sat down with him in his office at St. Joseph's. Flanked by plans for a new St. Joseph's research center and a whiteboard filled with math equations, diagrams, and to-do lists, Pipe described his collaborative work with GE and the St. Joseph's physicians, who together share the goal of advancing imaging technology for the benefit of patients all over the world.

Recently General Electric Healthcare and St. Joseph's Hospital signed what has been touted as a "landmark agreement." Can you explain the significance of this agreement?

From our point of view at Barrow, we think that having GE on as partners will help us get things done much faster and more efficiently. A lot of the time we spend developing new ideas requires us to find technical details about the scanners that GE could help us with in a much more efficient way. I think from GE's perspective, this is the very first time they are going to partner with any of their sites in the way they are working with us now. They have scientists at other sites who are working on research but the people who are coming here are much more oriented towards what's going to go out as product now rather than a long time from now. We will be developing technology they could be bringing to the patient in two years time. For GE, this is a way to partner with someone who knows the clinical side and can give them the clinical feedback in order to get their ideas into the marketplace faster. This is an experiment, and we're both going to see how it works.

Financially, what is GE providing St. Joseph's?

Financially the big thing is going to be the royalties paid. When technologies are developed here and sold on GE scanners, we will get a fraction of the sales price. In the long run if this is successful, it's going to be really good for everyone. And I think St. Joe's and GE see the money from the profits going back to the center as a way of reinvesting in this relationship to keep it going. You know, we talked about how to fund this. And I wanted to fund it through royalties because it's a way to tie the rewards to the success. If it's successful, GE's going to pay a lot more, but it's going to be worth it.

How does this model compare to other financial arrangements between companies and researchers?

Usually companies pay researchers for intellectual property and patents. We're moving away from that model in this regard in that they are not paying us a royalty for patents. Instead, they're actually paying us for the development work and not just for the intellectual property piece. That's probably a pretty unique thing. We're trying something that's very different. And I think it'll work because GE and St. Joe's, we both want our technology to be really clinical, really useful for patients. From St. Joe's side, it will help our patients. From GE's side, people will want it and they can sell their product.
How will technology benefit patients and Arizonans?

We focus a lot of our work on making imaging technology that can work with everyone. For example, there's a lot of great imaging technology but the problem is that it requires people to hold very, very still for a very long time. Sometimes that can get really great results. But in a clinical setting where you have a lot of patients who can't hold still for a long time, we need to focus on how to make the machines less sensitive to human motion. Also, we are concerned with showing the radiologists new things that they couldn't see before with another kind of MR. That way physicians can get better information to make choices about treatment.

Can you talk a little bit about about the development that went into PROPELLER?

Much of my research has focused on something called "diffusion-weighted MRI," which is what we use to look at strokes. In diffusion-weighted MRI strokes show up really bright. Now this sequence has to be really sensitive to motion because we're need to look at how much the individual water molecules move around in the tissue. The problem is that if a patient moves, the scan picks it up in addition to the pulsations in the brain caused by the pumping heart. I was playing around with this while at Wayne State University and figured out a way to correct for the motion, and it worked.

How does it work?

Every image picture is collected in pieces, and each piece is collected in about a tenth of a second. Then we wait a little while, then we collect another picture piece in a tenth of a second. And so on. We have a blurry image every single time from each piece, but the way the pictures pieces are taken each piece also includes extra unique information about detailed features in the image. So we line all the edges up, use a mathematical algorithm to take out any movement detected in the images, and you get a nice image.

Did this development make the headlines right after you discovered it?

Well, we published it and then it never went anywhere. The scanner I was on, it was not flexible enough to do the kind of research I needed to do. Then I came here and got on a better scanner for my work. Also, it's important to note that the physicians here are great in terms of interacting with research. They were really interested in the PROPELLER technology, and we worked together to get it going clinically. It's really nice that the research that goes on here is tied very heavily to the clinically side of things. It really helps everybody.

To what extent do you collaborate with the physicians here?

There are three radiologists in this hall, and they come in a lot. They'll tell us we're having problems imaging this or that, and then we'll think about ways we might be able to fix the problem. Or if I have an idea on a great new way to see a particular anatomy or a particular function or piece of tissue, I can go to them and say, I can do this, but does this matter, is it something I want to spend my time on. And then they'll give feedback. I'm also doing a lot of work with our neurosurgeons and neuropsychologist to image the brain and find out what areas of a person's brain are controlling specific functions. This is important for presurgical planning. When you have a tumor that's growing and pushing things out of the way in the brain, people's functionalities are different. And in that case a neurosurgeon might not have a great idea of where everything is, so these images, produced mostly by our neuropsychologist, Leslie Baxter, helps a lot.

Are there any challenges to being a researcher in the clinical environment?

There's always the pressure to make what you do more clinically focused. In university you can do basic research and have more freedom. It focuses what you do a lot. Sometimes it's a challenge, but I like that a lot. You get immediate feedback abut what you're doing, and hopefully sometimes you'll see it make a difference.

What's next with PROPELLER?

There are other types of MR images in the head. We're going to try to as much as possible make all the images in the head, make them resistant to motion. Right now propeller only applies to a few things, and we want to make that broader. We also want to look at imaging in the body.

Where do you see your work taking you in fifteen to twenty years?

One of the wonderful things is I have no idea, that's the real answer. Fifteen years ago, how could have predicted we'd be where we are? I think that we're starting a really big venture here with St. Joe's and GE potentially with ASU and UA. I really want to see this grow and become successful--and I don't mean successful financially. I really want us to have a positive impact on how people are scanned in MR. The vision for a lot of people in this community is genetic, personalized medicine. Maybe we can start some sort of scanning where we detect disease earlier and earlier. Rather than manage people after something catastrophic has happened, we can find the problems earlier. I think that's interesting for a direction in which to move imaging in.