February 1, 2010

UNT professor works to improve colon cancer screening

DENTON (UNT), Texas -- More than 130,000 people each year are diagnosed with colon cancer, resulting in some 50,000 deaths.

Colonoscopies help detect polyps, the precursor lesions of cancer, but the procedure is far from perfect.

Jung Hwan Oh, an assistant professor in computer science and engineering at the University of North Texas, is developing software to improve colonoscopies and create important quality controls for the procedure.

"Colonoscopies are crucial to detecting cancer, but the technology could be much better," Oh said. "These updates would improve health care."

During colonoscopies, physicians inspect the inner lining of the human colon and can perform therapeutic operations.

Oh and his colleagues, who work at the Mayo Clinic and Iowa State University, first developed a prototype that provides a recording of the procedure in addition to showing the live image. That way, the doctor can return to the images after the procedure to take a second look.

The prototype is being used in a handful of hospitals across the country, including the Mayo Clinic in Minnesota.

Oh and his colleagues are now working to develop quality-control measures for colonoscopies. Recent data suggests that four to 12 percent of even large polyps are missed in a routine colonoscopy.

To combat the miss-rate, Oh is working on software that would automatically detect polyps, rather than relying on the doctor to find them.

In addition, he is working on an automated system that would alert the doctor to any problems. For example, the American College of Gastroenterology suggests the doctor should spend at least six minutes inspecting the colon. If a doctor were to try to end the inspection before that point, a warning light could be activated.

"This would just give some extra assurances to the doctor and patient that the procedure was thorough," Oh said. 

Oh's research is funded by more than $450,000 in grants, including a new $184,000 grant from the National Institutes of Health.

Two UNT post-doctoral students and one graduate student work with Oh on the research.