New computer-aided breast cancer screening methods can distinguish tumors from dense tissue
Orange County, CA - July 15th 2016 - Women with dense breast tissue are four to five times more likely to develop cancer than women with low breast density. This staggering statistic is compounded by the fact that dense tissue appears white on a mammogram—as tumors do—making accurate diagnosis increasingly difficult for radiologists.
With these facts in mind, researchers and policy makers have made strides to educate and equip the public with knowledge and tools in recognizing the potential dangers this type of breast tissue poses. 26 states have already passed legislation requiring physicians to notify and inform women with dense tissue of the recommended protocol.
Upon diagnosis of dense breast tissue following traditional mammography; additional screening methods are recommended to rule out breast cancer. In response to the commonality of this condition, methods such as automated sonography and 3D tomosynthesis have emerged.
Findings published in American Journal of Roentgenology show that the combination of mammography and automated breast ultrasound (ABUS) improves the likelihood of detection compared to just mammography for those with dense breast tissue.
Studies have found that a majority of diagnoses from sonographies are small, node-negative, and invasive cancers that have better prognosis if caught early. A handheld ultrasound exam takes an average of 20 minutes, whereas the ABUS method can complete an exam in two minutes or less. Furthermore, ABUS advantageously allows for standardized diagnoses and a reduction in operator dependence.
Another study, conducted at St. Joseph’s Health Care found that using sonography for breast cancer detection is useful between mammography exams in exposing lumps. After reviewing 618 cases, researchers, using sonography, were able to identify changes in 311 cases (50.3 percent), while mammography could only ascertain change in 80 cases (12.9 percent). Meanwhile 234 cases were divulged through ultrasound.
“Targeted ultrasound yielded more statistically significantly diagnostic information than repeat mammography, [which] agrees with other studies [that] found a significant number of palpable abnormalities detected by ultrasound were mammographically occult,” wrote the authors.
But sonography is not the only breast cancer detection method with integrated automated detection. Last month, a category in the International Symposium on Biomedical Imaging sought to find an artificial intelligence system to facilitate automated lymph node metastasis diagnosis. When coupled with a pathologist, the winning AI system, PathAI, could detect cancerous cells in a biopsy sample with 99.5 percent accuracy.
In addition to these automated detection systems, a study from Duke found 3D digital breast tomosynthesis (DBT) produce more accurate images than 2D full field digital mammography (FFDM). Per the FDA’s approval in 2011, 3D DBT cannot be utilized without a FFDM. Since this 3D DBT technology has only been in use for five years, it hasn’t been adopted by enough institutions to become standard. However, as health care providers continue to adopt the variety of supplementary breast cancer screening exams steadily becoming available, missed diagnoses due to dense breast tissue can be assuaged.
