The potential to extrapolate accurate data from 3-dimensional (3D) images of the breast is enormous and will greatly improve our ability to qualitatively determine differences in shape, size, and contour. The validity of these calculated measurements is important and needs to be determined before any meaningful data can be evaluated.
Premastectomy 3D images (3dMD patient) were obtained on 19 breasts (14 patients). The volume of the mastectomy specimen was determined intraoperatively using water displacement. Two independent raters then calculated breast volumes using the 3D images and software, and these were compared with the intraoperative volume. Inter- and intrarater reliability was determined. Part II: Surface measurements (nipple to notch) were then evaluated on 20 breasts (10 patients) by comparing the 3D image determined distance to the known measurements.
The average breast volume was 500 mL, compared with 489 mL for rater 1 and 490 mL for rater 2. The relative difference between the measured volume and the calculated volume for rater 1 and rater 2 was about −2%, with a standard deviation of ± 13% to 16%. The coefficient of reproducibility for each reader was excellent, at 0.80 for rater 1 and 0.92 for rater 2. The level of agreement between the readers was also high at 0.975. Part II: The average nipple to notch measurement for each patient was 27.1 cm, compared the calculated average of 25.1 cm for rater 1 and 26.1 cm for rater 2. The mean relative difference between the measured and calculated distances for raters 1 and 2 was about −6%, with a standard deviation of ± 6% to 7%. The level of agreement between readers was high, at 0.975.
The ability to objectively determine breast volume and surface measurements using 3D imaging technology is now available with consistent and reproducible accuracy. Measurements are typically underestimated, with more variability when calculating volumes. Although inherent subjectivity will always exist when evaluating breast measurements, 3D technology provides invaluable information, particularly in the longitudinal evaluation of results.