Quantifying B-mode images of in vivo rat mammary tumors by the frequency dependence of backscatter

Objective. To evaluate the frequency dependence of ultrasonic backscatter for its ability to differentiate between neoplastic and healthy tissue. Methods. Standard B-mode images were created of 5 rats with spontaneous mammary tumors, and regions of interest in the lesion and surrounding tissue were parameterized by the slope of the backscatter amplitude versus frequency. Results. In 4 of the 5 rats, the averaged backscatter slope of the regions of interest in the tumor was significantly (P < .05) different from that of the surrounding tissue, and the fifth case had a moderate difference (P = .20). The consistency of the averaged slope values (1.2-1.8 dB/MHz) across all but 1 of the mammary tumors was encouraging for the prospect of identifying a tissue type by its backscatter slope. Conclusions. This work suggests that characterization and diagnosis of tissue types may be possible by using ultrasonographic images quantified by the frequency dependence of backscatter.