Women’s chances for early detection of breast cancer might increase thanks to Iowa State chemistry professor David Hoffman, who has contributed to research that could give radiologists a clearer view of mammograms.

Along with researchers at the University of Houston, Hoffman recently wrote a paper and helped develop software that shows promise in feature enhancement of digital images, particular those of mammograms.

“This is a very important area for people to be working in,” he said. “I’d be very pleased if this work led to something that was really useful for a lot to people.”

Statistics from the American Cancer Society in 1997 showed a five-year survival rate for breast cancer patients decreases from about 96 percent for early stage detection to just 20 percent for cases detected in the later stages.

And even though regular mammography combined with clinical breast examination offers a good chance of early detection, only 60 percent of breast cancer victims are diagnosed at this stage.

Some of the 40 percent that go undetected might be caused by inefficient exams. As many as 10 percent of all cases go undetected by mammography, said Dr. John Shierholz, radiologist at the McFarland Clinic, 1215 Duff Ave.

The reason for the inefficiency is because breast cancer is very difficult to detect, Hoffman said.

“The distinction between healthy tissue and diseased tissue is very small, and you need experienced radiologists just to interpret a mammogram,” he said.

Current mammograms, for the most part, still are being developed with film for analysis. However, the field of digital mammography is growing rapidly. It offers doctors a medium that can be enhanced, enlarged and easily sent electronically for further analysis.

“Digital mammography appears to have some real possibilities,” Shierholz said.

Along with longtime colleague Don Kouri, professor of chemistry and physics at the University of Houston, Hoffman was able to provide new methods in solving advanced equations in quantum mechanics. It was these equations that opened the door for researchers to develop a software program to enhance digital images.

The result is computer-aided diagnosis (CAD) software called “Sparkle.” The new program was designed by a team of scientists from the University of Houston and employs the methods developed by Hoffman and Kouri.

“When we showed the results of our work to some people in the [digital mammography] industry, they were quite impressed,” Kouri said.

Hoffman compared his groundbreaking work to listening to a radio in a car while driving away from the source of a broadcast. The farther the car is away from the source, the more the signal weakens and increases in static and other background noise.

The same can be said of blurry or grainy images, except the real image is obscured by visual noise instead of audio noise.

“What we were able to do was remove a good deal of visual noise that was obscuring the mammogram,” he said.

Hoffman said although breakthroughs have been made in digital mammography, it will be some time before it is the industry standard. For now, the papers he and others have written are up for publication, and the software still is being developed.