Mammographic Determination of Breast Volume by Elliptical Cone Estimation

Background We propose a new breast volume calculation method, treating the breast as an elliptical cone on craniocaudal and medial lateral-oblique mammograms. This study aims to compare the accuracy and reproducibility of this proposed calculation method to the old Katariya method, which calculates breast volume as a circular cone on the craniocaudal mammogram only. Materials and methods From January 2005 to December 2006, 83 mastectomy patients with recorded breast weight and available preoperative mammographic films were included in the present study. Two surgeons independently measured breast height and width on the preoperative craniocaudal and medial lateral-oblique mammograms. Breast volume was calculated as a circular cone in the craniocaudal view, and as an elliptical cone in both the craniocaudal and the medial lateral-oblique mammograms. The accuracy of each method was determined and compared by linear regression analyses. Interobserver variability was assessed by bivariate correlation. Results The most accurate formula for calculating breast volume was the one that assumed the elliptical cone projection. The mean error of estimated breast volume using the elliptical cone (EC) formula is 3.8 cm(3) (standard deviation [SD] = 133 cm(3)). The mean error of traditional circular cone (CC) formula is -51.3 cm(3) (SD = 182 cm(3)). From a linear regression model, the correlation coefficient of estimated breast volume using EC formula measured 0.977, whereas that using the CC formula measured 0.952 (Fig. 1). Measurements were reproducible between the two independent observers; the Pearson correlation for the EC formula is 0.93 (p < 0.001), and that for the CC formula is 0.95 (p < 0.001). Conclusions Breast volume can be accurately determined from measurements made on mammograms. The proposed EC formula calculating breast volume on both the craniocaudal and the medial lateral-oblique mammograms may be more accurate, as the horizontal and vertical dimensions of the breast are not always the same. In addition, height measurement on the medial lateral-oblique film can eliminate the compression error as the craniocaudal film may miss the base of the breast and underestimate the breast volume. Taking both mammogram views for measurement is comparably reproducible to the traditional method. Reproducibility of measurement can be further enhanced by better defining the point of measurements.