Software developers and cancer researchers are developing new technology utilizing microwave tomography imaging (MTI) to detect cancerous tumors in women with dense breast tissue at a lower cost that that of X-ray mammography.
The project is an international collaboration between medical professionals from Seoul National University, research scientists from the Electronics and Telecommunications Research Institute (ETRI) in South Korea, and faculty from South Dakota State University.
“The ETRI in South Korea holds the patent on the microwave tomography machine. The research team led by Soon-Ik Jeon, Principal Researcher and Director of Human-Life Radio Technology at ETRI, is currently developing its capabilities as a cancer-screening tool,” explains Sung Y. Shin, Professor and Graduate Coordinator of Computer Science at South Dakota State University. “Human testing of this experimental procedure was approved last summer and is being done in South Korea while the software is being developed here at South Dakota State University.”
According to Shin, Seoul National University Hospital’s radiologists conducted imaging on volunteer patients using microwave tomography and compared results with those of both traditional X-ray mammography and MRI. This trial was conducted to test the technology’s feasibility and proved 80 percent accurate, Shin says.
Cutting Edge and Cutting Costs
Shin notes that X-ray mammography, while most commonly used in the United States, is not the most accurate evaluation technique available when it comes to detecting early-stage breast tumors.
“Tomography imaging offers women an increased comfort level by eliminating compression of the breast, as well as reducing exposure to radiation to less than that of a cell phone,” Shin says. “Because tomography images are three-dimensional, only one image is required, unlike the multiple angles necessary with mammography. The number of steps involved in the screening procedure is thereby reduced.”
While many image-processing techniques and algorithms have been developed — including object extraction, classification, object correlation, image similarity and image retrieval from a large image archive — Shin says his software project, titled Intelligent Breast Cancer Decision Support System, was the first to identify tumors using MTI and then compare those images to a database of more than 100,000 MRI images. The software program extracts similar MRI images along with associated case files noting which treatments were used and each treatment’s level of success in combating cancer.
By March 2014, the ETRI plans to begin human testing using an upgraded experimental machine with a wave frequency of 6 GHz — rather than the current 3 GHz — as higher frequencies provide better images.
Shin looks to perfect the software and imaging techniques within the next two years and hopefully bring the machine to the American market by 2017. Today, a tomography machine costs less than $100,000, compared with $300,000 for a mammography unit and as much as $4 million for an MRI machine.