CHICAGOIn a new study, elasticity imaging software was highly sensitive and specific when classifying malignant and benign breast lesions, Richard Barr, MD, PhD, professor of radiology, Northeastern Ohio University College of Medicine, Youngstown, reported at the 92nd Annual Meeting of the Radiological Society of North America (RSNA) (abstract LL-BR4218-B01). The software captures ultrasound frames in real time as they are being collected and characterizes tissue movement to differentiate hard from soft structures.
In the study, lesions were graded as malignant or benign based on the difference in their size on conventional B-mode ultrasound and elasticity imaging. If a lesion was smaller on elasticity imaging than on B-mode ultrasound, it was classified as benign (Figure 1); if larger, it was considered malignant (Figure 2).
Of 80 patients with 123 breast lesions, elasticity imaging correctly predicted all 17 lesions that were malignant at biopsy and 105 of 106 lesions that were benign. The technique had a sensitivity rate of 100%, specificity rate of 99%, positive predictive value of 94%, and negative predictive value of 100%, Dr. Barr said.
May Reduce Unnecessary Biopsies
These early findings suggest that the software, which can be added to conventional ultrasound systems, may significantly reduce the number of biopsies performed each year. Dr. Barr noted that approximately 1.4 million breast biopsies are performed annually, and 80% find only benign masses. "Elasticity imaging may be able to decrease the need for biopsy or additional imaging, including the patient anxiety that is associated with finding new lesions and undergoing a biopsy," he said.
Elasticity imaging also is providing some unique observations about breast lesions, Dr. Barr noted. Cysts, which are seen as thin-walled structures that are free of fluid on conventional ultrasound imaging, have a characteristic bull's eye appearance on elasticity, with white dots inside and posterior to the lesion (see Figure 3).