Researchers at Rochester Institute of Technology and physicians at Rochester Regional Health are advancing thermal imaging techniques as a potentially safer and less invasive diagnostic tool for the detection of early-stage breast cancer.
A National Science Foundation grant of $99,620 is supporting the two-year project and the multidisciplinary team that represents the strategic collaboration of the RIT & Rochester Regional Health Alliance. Satish Kandlikar, the Gleason Professor of Mechanical Engineering in RIT's Kate Gleason College of Engineering, is leading the NSF-funded study to establish modern thermal, or infrared, imaging as a viable alternative to other technologies, such as ultrasound and MRI, used in addition to conventional mammograms.
"Modern infrared imaging has the potential to significantly increase the accuracy of screening for breast cancer and could have broad implications for preventive medicine," Kandlikar said.
According to the American Cancer Society, breast cancer will affect one in eight women in the United States, and higher breast density further increases the risk.
"About 50 percent of women have dense breast tissue, and mammography alone may not be as effective as we need it to be for screening," said Dr. Lori Medeiros, medical director, Rochester Regional Health Breast Center. "Advances in thermal imaging technology would provide a relatively inexpensive, non-invasive approach that doesn't use ionizing radiation and causes minimal discomfort to the patient."
In dense breast tissue, layers of glandular and connective tissue appear white on a mammogram and can make tumors--which also look white--harder to find. Expensive diagnostic procedures may be required to rule out abnormalities.
Kandlikar said existing thermal imaging techniques are not very accurate, are cumbersome and uncomfortable to the patient, requiring cooling of the breast tissue by 3 to 5 degrees Celsius with a blast of air or metal plates, and are not widely used.
New thermal imaging techniques can detect when cancer alters the metabolic behavior of tissues and radiates specific thermal signatures, or localized "hot spots." Kandlikar's method combines thermal imaging with an artificial intelligence system that predicts the location and size of tumors on a thermal map. The algorithms will simulate scenarios based on numerical models of thermal-signature patterns derived from mammography and MRI conducted at Rochester General Hospital.
"We will develop models and a large set of benchmark data sets, which will be validated with clinical data obtained with our Rochester Regional Health partners," Kandlikar said.
"The team isn't interested in replicating the current methods of thermography; we want to improve the technology enough that thermal imaging will be a fully-reliable complement to mammography in breast cancer detection," said Medeiros. "We're looking forward to the development of this new technique as part of our commitment to the community in the screening and treatment of breast cancer."
Kandlikar's research team includes Medeiros and RGH medical personnel Dr. Donnette Dabydeen, diagnostic radiologist, and Dr. Prad Phatak, medical director of the Lipson Cancer Institute and interim chief of medicine at Rochester General Hospital.
RIT engineering Ph.D. students Jose Luis Hernandez and Abigail Clark, respectively, will conduct the numerical simulation and develop the infrared imaging system for obtaining patient data. Clark will work in close supervision with the medical personnel at the hospital.
Three early-phase clinical studies show promising initial data for patients with lymphoma, gastric cancers