Three-dimensional scanning for improved bra design

Introduction and Objectives: Three dimensional (3D) scanning has potential to significantly improve bra cup design, sizing and fit by increasing accuracy in measuring parameters such as breast volume [1,2]. Accurate breast volume measurement depends on the ability of a scanner to visualize the entire breast. Inaccuracies have been reported when measuring the volume of large ptotic breasts due to the tendency of the inferior aspect of the breast to contact the abdominal wall when women are scanned in a standing position [1,3]. As this compromises the ability of the scanner to visualize the entire breast, it is likely that breast volume is underestimated. To improve breast visualization during 3D scanning, a prone position, where the breasts hang away from the trunk, has been proposed when scanning women with large ptotic breasts [3]. The objective of this study was to quantify the difference in breast volume determined when scanning women with large ptotic breasts in a standing compared to a prone position. It was hypothesized that scanning large ptotic breasts in a prone position would allow complete visualization of the breasts and, in turn, a greater breast volume, compared to scanning women while standing. Methods: Participant's were 49 women (mean age 37.1±13.0yr; mean BMI 9.5±5.2kg/m2; average cup size F, range D-I; average band size 16, range 8-18). Adhesive markers were placed on participants’ skin around the outline of each breast. The breasts were scanned using a hand held scanner (ArtecTM) while participants assumed three positions: i) standing, hands on hips, ii) standing, hands on head [1,2], and iii) lying prone. A 3D model of each breast was created from each scan and breast volume was calculated (mL; Geomagic software, 3D Systems, USA). Breast visualization was rated as “complete” or “incomplete” and the level of breast ptosis was characterized by measuring sternal notch to nipple distance [4] and by grading the ptosis from 0-3 [5]. A repeated measures ANOVA with one within factor (scanning position) was used to determine any significant (p <0.05) differences in breast volume among scanning positions. Bland-Altman plots were used to determine difference in agreement between scanning position for breast volume data. Frequency of “complete” and “incomplete” breast visualizations in the scanning positions was also compared. Results: Breast volume ranged from 302-2,265mL (mean 1,038±450mL). Ninety per cent of participants were classified as having ptotic breasts (mean stage 2; range stage 1-3) and the sternal notch to nipple distance ranged from 20.8-35.2cm (mean 28.5cm). The prone position allowed complete breast visualization in 100% of participants, compared to only 5% of participants in either standing position. Breast volume measured in the prone position was significantly greater than that measured in either standing position, which were not significantly different to each other (Table 1). Bland-Altman plots showed good agreement in breast volume data measured in the standing positions but poor agreement with a negative bias between breast volume measured in the standing position (hands on hips) and the prone position. The negative bias and consequential underestimation of breast volume in the standing positions increased as breast volume and ptosis increased (Fig 1). The error of underestimation in breast volume of both standing positions compared to the prone position was as high as 473mL at an individual level and 103mL at a group level. This equated to ~3% underestimation of breast 1079 volume for participants at the smaller end of the breast volume (size) spectrum (volumes 400-500mL) and 7-10% for breast volume at the larger end of the spectrum (volumes >500mL). Conclusion: The volume of large ptotic breasts is likely to be underestimated when women are scanned while standing, particularly for women with large breast volumes (>500mL). This error has the potential to negatively affect bra sizing,