Abstract

Abstract Measurement of the strain induced by compression can be useful in the diagnosis of breast tumors. Microcalcification (MC) detected by mammography could be the only symptom of early breast cancers. The purpose of this study was to develop a quantitative method for a differential diagnosis of MC lesions based on the strain ratio (SR) obtained by applying different breast compression pressure during mammography. Methods: This prospective study was conducted with institutional review board approval and the written informed consent was obtained. Twenty-five mastectomy specimens (24 for treatment of breast cancer and 1 from contralateral prophylactic mastectomy) with MCs were examined by digital mammography. There were 24 malignant MCs (mean size of 18.8 mm) detected in the 24 separate specimens, including 14 lesions of invasive carcinoma and 10 lesions of carcinoma in situ. Five benign MCs (mean size of 21.6 mm) were detected, including four lesions which coexisted with malignant MCs in the four separate specimens, and one was discovered in the prophylactic mastectomy specimen. Each specimen experienced four different magnitudes of compression force and the corresponding images were acquired for further analysis. Images in Digital Imaging and Communications in Medicine format (DICOM) were exported into custom software, written in Matlab, to contour the region of interest. Stress was defined as the force divided by the whole specimen area, and the square root of MC area was calculated to represent strain deformation. The SR was the slope of the best-fit line through the four points of each lesion. Histology of the MC was correlated with SR. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of SR for discrimination between benign and malignant MC. Results: The SRs were higher in the malignant MC group (139.10 a.u.) than in the benign MC group (34.09 a.u.; p = 0.024). The SR between invasive carcinoma and carcinoma in situ showed no statistic significance (127.03 a.u. vs. 156.00 a.u.; p = 0.642). The area under the ROC curve for the diagnosis of benign and malignant MC was 0.825 (95% CI, 0.667 to 0.982). Conclusion: This study demonstrated that the SR detected by mammography showed potential for distinguishing benign from malignant MCs. Further in vivo investigations of larger cohorts are necessary to validate these results. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-08-01.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call