Suitability of 3D scanning as a screening tool for reduction mammoplasty referrals for plastic surgery—a pilot study

Introduction

Reduction mammoplasty is a common procedure in plastic surgery. Women seeking to relieve symptoms caused by hyperplastic breasts represent a significant population. It has been shown that health-related quality of life improves significantly after reduction mammoplasty (1). In Finland there is a universal health care coverage. For the patient to be eligible for breast reduction with public funding, the patient needs to present symptoms that can be caused by overly large breasts, and the amount of resected tissue needs to be at least 300 g per breast. In previous studies it has been shown that the quality of life increases when the resection weight per breast is minimum of 300 g (2,3). In many countries also insurance companies often require a minimum resection weight for provisional approval (4). Making accurate estimations of breast size and resection weight is important. Plastic surgery clinics in public sector get a large amount of breast reduction referrals, including also patients who do not fulfill the criterions for the operation. It is understandable that general practitioners in health centers might not have large experience for evaluating the breast size. These referrals lead to expenses and use of limited health care resources especially when the criteria for the reduction mammoplasty are not fulfilled, and the patient will not end up with the surgery. Therefore, it would be beneficial if there was a reliable and easy to use method, that would not involve radiation, of measuring the breast volume already in the referring clinic to avoid unnecessary referrals.

The estimation of resection weight is not easy. There has been an effort to create formulas that would reliably predict the resection weight after reduction mammoplasty. Clinical judgement and many scales (e.g., Schnur, Apple, Descamps, Galveston) have been used. These formulas are numerous, but they are not accurate representations of resection weights and thus are unreliable for surgical planning. There is no “one-scale-fits-all” paradigm (4-6).

Measurement of sternal notch-to-nipple distance (SN-N) is commonly used to estimate breast size for patient counselling, surgical decision making and even preoperative insurance claims (7). However, using only SN-N could be misleading in evaluation of the breast size. The breast volume can be the same with varying distances depending on the degree of mammary atrophy. In fact, an atrophic, ptotic breast with a longer SN-N can have less volume than a non-ptotic non-atrophic breast. SN-N has been used as a part of formulas and predictive models which evaluate the breast size and predict the amount of tissue that can be resected. It has been concluded in earlier studies that through multiple linear regression analysis, the results indicate that among the other indicators, preoperative SN-N significantly affect the resection weight. Furthermore, higher values of these indicators are associated with a greater weight of tissue resected during surgery (8).

The conventional estimation methods have limitations in capturing volumetric details and there has been a shift towards more precise methods as for example three-dimensional (3D) scanning. In a study conducted by Killaars et al., Vectra XT 3D imaging system was compared to magnetic resonance imaging (MRI) in estimation of breast volume. Vectra showed lower breast volumes than MRI, but there was a linear association with MRI and it was therefore considered suitable for clinical practice for plastic surgeons (9). Howes et al. also found that 3D scanning, with an established protocol, was equivalent to non-contrast MRI for the assessment of breast volume (10). However, also a handheld 3D camera, Vectra H2, has shown to provide highly accurate predictions of mastectomy volumes (11).

Since there are many publications supporting the use of 3D scanning for breast volume estimation, we wanted to explore whether a handheld and easily portable 3D scanner would give comparable measurements of breast volumes to measurements calculated from the mammography images, when the scanning is performed by a healthcare professional without previous experience in imaging after a short training to use the scanner. There are many different 3D devices commercially available. Traditionally, 3D imaging in plastic surgery has been dominated by systems such as Vectra (Canfield Scientific, Parsippany, NJ, USA). However, the high cost of these systems often limits their accessibility. Even smaller devices are now sufficiently powerful to allow the use of 3D cameras and to easily process 3D images. Kyriazidis et al. showed in their study that mobile light detection and ranging (LIDAR) technology on iOS devices demonstrates potential as an effective solution for 3D breast imaging in plastic surgery. The 3D Scanner App exemplifies the potential for accessible, cost-effective, and secure 3D imaging in plastic surgery, paving the way for broader adoption of this technology in clinical practice (12). The goal of the current study was to evaluate if a handheld easy to use 3D scanner would be usable as an assistive device for general practitioners in the breast size evaluation. If succeeding, it could be used as a tool to avoid inadequate referrals. The aim of the current study was not to validate a 3D scanner. We present this article in accordance with the TREND reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2025-1-577/rc).

Methods

The study population comprised of 50 women recruited for the study in Tampere University Hospital, Finland. Of these patients 25 were recruited from the gynecological outpatient clinic (GYN) and their reason for the visit was not breast related. 25 patients were recruited from the plastic surgery outpatient clinic (TUPO) and who were seeking reduction mammoplasty. The demographics of the patients are presented in Table 1.

Exclusion criteria for the study were age under 40 years or over 70 years, and a mammography performed within one year before the recruitment time. The age limit was set to 40 years, since in Finland the mammography is not routinely performed for patients under 40 years of age before not-cancer-related breast surgery. Patients who had had a mammography within one year were excluded to avoid extra exposure to radiation. The patients recruited from the plastic surgery clinic had reduction mammoplasty performed within six months after the recruitment, and at the operation the resected tissue was weighed.

Before recruiting and scanning the 50 study patients, we performed scanning for five patients to create proper setup for the scanning and educate the study nurses to do the scanning. The training how to use the 3D scanner before the first study patient took 2 hours. The 3D scanner used was Artec Leo and the software Artec Studio by Artec 3D. The size of the scanner was width 229 mm, depth 165 mm, and length 387 mm. The scanning was performed the patient standing against an adjustable table and the breast laying on the table. The position of the arms was not standardized. All patients also underwent mammography. Volume measurements and calculations from the mammography were performed by experienced breast radiologist. The breast volume was calculated from mammograms by making two measurements on the craniocaudal view and knowing the compression thickness. This formula has been shown to accurately and reproducibly determine the breast volume (13). According to study by Goto et al. the breast volume measured by mammogram is moderately different from that measured by computed tomography (CT) or MRI. The correlation was weaker between the CT- and mammogram-determined volumes than between the CT- and MRI-determined volumes. However, mammogram is the most common and easy method of breast examination. They also conclude, that the accuracy of breast volume measured by mammogram is acceptable for daily practice (14). Cropping of the breast borders from the 3D images was performed by experienced plastic surgeon and the calculations of the volume was conducted by a Master of Science specialized in the use of the scanner and the software. The 3D volumes were then compared to mammography volumes. Example of a 3D image in Figure 1. Resection weights were converted to volumes by using formulas described elsewhere (15).

Figure 1 3D image of a breast. This image is published with the patient’s consent. 3D, three-dimensional.

The study was conducted by following the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional scientific center ethics board of Tampere University Hospital (No. R22053). Written informed consent forms were obtained from the study participants.

Statistical analysis

Due to the small sample sizes, non-parametric statistics were reported. Demographics and different modalities of patients at TUPO and GYN were tested using Mann-Whitney’s U-test, Pearson’s Chi-squared or Fisher’s exact test. In further analysis, the values of larger breasts were used. Breast size according to SN-N was classified into tertiles with nearly equal number of patients: small (n=16; 22.0–27.0 cm), medium (n=16; 27.5–30.0 cm) and large (n=18; 31.0–39.5 cm) breasts.

Correlations between volumes were tested using Spearman’s rank correlation test showing results by rhos with 95% confidence intervals (CIs). The consistency of the mammography and 3D scanning volumes has been examined by using their difference. The agreement between volumes of mammography and 3D scanning was observed using the Bland-Altman plot.

Statistical analyses were carried out with SPSS (SPSS Inc., IBM Corp., Armonk, NY, USA), version 30. The 2-sided P value <0.05 was considered statistically significant.

Results

The patient populations from the two outpatient clinics were comparable when considering somatic diagnoses, psychiatric diagnoses, and body mass index (BMI), but the patients from the gynecology were younger than the patients from plastic surgery (Table 1). Patients who visited plastic surgery clinic had longer SN-N and larger breast volume both in mammography and 3D scanning (Table 2). In addition to breasts being smaller, the breast tissues were also denser according to Breast Imaging Reporting and Data System (BI-RADS) classification in the GYN.

The volumes obtained from mammography and 3D scanning were correlated (Spearman’s rho 0.76, 95% CI: 0.60–0.86). However, when classified into three categories according to the SN-N, there was no linear association between the mammogram volume and 3D volume in the group with middle sized breasts, rho =0.10, where the variation was greatest (Table 3). Thus, the relationship between the volumes was not necessarily linear (Figure 2). The best overall explanation for association of mammography and 3D scanning volumes was found by fitting a cubic fit line: for small breasts R²=90.3, the medium sized breasts still only R²=9.0, but the large sized reach explanation of R²=70.4.

Figure 2 The best fit for the association between mammography and 3D scanning volumes according to breast size. 3D, three-dimensional.

Mammography and 3D scanning volumes increase according to SN-N (Table 3). Likewise, their difference increases as SN-N distance increases. Information from resections was available only from the plastic surgery clinic, since the patients from the gynecology were not operated. From large breasts (measured by SN-N distance), a bigger percentage of the volume was resected (P<0.001) than from medium-sized breasts.

All the patients, the resection weight exceeded 300 grams at least on one of the breasts; in fact, only two had less than that resected from one of their breasts. Differences in volume values indicate that 3D scanning shows systematically around 300 cc smaller volumes than mammography.

In BI-RADS classes A–B both volumes were higher than in classes C–D (Table 4). Breast size and BI-RADS were associated with each other (P<0.001). There was no difference between the BI-RADS categories in terms of resected weight, volume, or proportion.

BI-RADS classification seems to affect how well scanning, both mammography and 3D, correlate with the resection volume. With Spearman’s correlation test they correlate in class A–B (rho varies from 0.80 to 0.89), but varies from 0.29 to 0.43 in class C–D.

Figure 3 shows agreement between mammography and 3D scanning volumes. Four outliers, according to the 95% agreement limit, had large or at least medium-sized breasts. Additionally, those patients situated inside but near the 95% agreement limit were patients including in the groups of medium- or large sized breasts. According to BI-RADS classification, outliers belong to the B or C classes.

Figure 3 Bland-Altman plot. The mean and difference of mammography and 3D scanning volumes were presented using the categorized sternal notch-to-nipple distance (small, medium or large-sized breasts) with BI-RADS values (A, B, C or D). The black line represents the mean, and the dashed line represents the 95% confidence interval of the mean. 3D, three-dimensional; BI-RADS, Breast Imaging Reporting and Data System; SN-N, sternal notch-to-nipple distance.

At the beginning of the study, it took 20–30 minutes to set up the patient and to do the scanning. After the nurses got more familiar with the scanner, the time required per patient shortened to 10 minutes.

Discussion

In this pilot study of 50 patients, we tested the suitability of a handheld 3D scanner for breast volume estimation for general practitioners. Our goal was to seek a reliable, cost-efficient and easy-to-use tool to identify already in the referring unit patients who will not qualify for reduction mammoplasty with public funding and avoid unnecessary referrals. We found that 3D scanning correlates with mammography in patients with smaller breasts and the correlation was almost linear. However, in the medium sized breasts we could not find any correlation. To further investigate this phenomenon, we will need a bigger number of patients and see if by dividing the medium size breast into two groups we can decrease the variance in this group of patients. Also, with medium and larger breasts the need for standardized positioning of the patients, need for more comprehensive training for the scanner, and refinement of the cropping of the breast in the 3D pictures becomes apparent.

For the accuracy of the study in evaluating the suitability of 3D scanning in screening we also included patients with smaller breasts. Therefore, half of the study patients were recruited from clinic visits not related to breasts. As expected, the patients from the plastic surgery clinic had bigger breasts that the patients from gynecological clinic, since the reason of the visit in plastic surgery was large breasts. In our study we cleared that the volumes from 3D scanning correlate with the volumes calculated from the mammograms. However, when further examined by the size of the breast (determined by SN-N distance) the correlation only persisted with smaller breasts. In patients with larger breasts the positioning of the breast is more challenging, and breast can form folds that can affect negatively the accuracy of 3D scanning. Patients with bigger breast also tended to have higher BMI more often than patients with smaller breasts, therefore forming obtrusive folds also on their sides hampering the cropping of the breast for calculations of the volume.

We think that positioning of the patient and breast is a main factor contributing to this finding. For example, when performing mammography, the position of the patient and the breast are highly determined, providing scans with even quality. Tong et al. also found in their study, that they were able to obtain complete scans for smaller breasted individuals in all positions, but struggled to acquire complete scans for individuals with large breasts regardless of hand position due to shadowing (16). In a study by Kyriazidis et al. they concluded that accuracy of scanning shows dependence on both patient factors and operator technique. They also conclude that there is a notable learning curve for operators to achieve consistent, high-quality scans (12). We also observed, that with a more pronounced abdomen the board where the breast rested tended to move with breathing causing unwanted movement. In the current study the posture of the patients was otherwise assigned, but the positioning of the arms was determined on the patient basis for them to be least on the way of the breast when scanning. The earlier studies have not reached the consensus about the positioning of the arms (17,18).

In Tampere University Hospital when we get a referral for breast reduction, we require that photographs of breasts from the front and both sides are included. If it is clear in the photos, when evaluated by an experienced plastic surgeon, that the minimum amount of 300 g per breast is not reasonably removed, we do not accept the referral. If the estimation cannot be made from the photographs or it is a borderline case, we accept the referral and do the evaluation in the outpatient clinic. The referral will also be unaccepted if the BMI is >30 kg/m². Screening with the photographs seems to be beneficial, since in all the patients that were operated the resection weight was over 300 g in at least one of the breasts. The patients in plastic surgery clinic also had bigger breast when measured with all three different modalities, when comparing to patients from the gynecology. With this practice screening with photographs, we can bypass a considerable number of unnecessary visits that would not lead to operation.

However, the screening with photographs is not without problems. Density of the breast (BI-RADS class) might affect the estimated resection volume and there could be some patients left out that could fill the reduction mammoplasty criteria, when the screening of the breast size is based only on the photographs. As conventional medical photography has inherent limitations in accurately capturing volumetric details, there is an urge to a shift towards more precise methods. BI-RADS class affecting the correlation of resection volume with mammography and 3D scanning is most likely related to notion, that BI-RADS class A breasts tend to be smaller and in our study 3D scanning works better on smaller breasts. Density of the breast might affect for example its compressibility and ptosis, therefore explaining why BI-RADS classification has an effect on how well both mammography and 3D scanning correlate with resection volumes.

In the current study we set to explore, whether 3D scanning could be used as a screening tool for reduction mammoplasty when the scanning is performed by a healthcare professional unexperienced with imaging and after only a short training period. We found that 3D scanning correlates with mammography in patients with smaller breasts, but it gave systematically 300 cc smaller volumes than volumes obtained from mammograms. This phenomenon is most likely related to difficulties in cropping the breast and defining the border between the thoracic wall and breast. The need for standardized positioning of the patients, need for more comprehensive training for the scanner, and refinement of the cropping of the breast in the 3D pictures becomes apparent before its use in screening.

Conclusions

The number of referrals for reduction mammoplasty in plastic surgery is large. Our current practice of screening the referral with photographs seems to reduce the number of unnecessary visits not leading to surgery. However, more tools are needed to evaluate breast size more accurately especially in denser breasts. We set up a pilot study to explore whether 3D scanning could be easily implementable for wider use in referring units to screen eligible patients for the surgery. We conclude that with this study population we established that as it stands the handheld 3D scanning device has sufficient technical properties for breast scanning, but patient positioning requires standardization and further training to improve the quality of the scanning. With non-standardized scanning technique, the employed scanner is not ready to be used as such. The underestimation of volume in a described quantity is not acceptable and could lead to incorrectly rejected referrals. Before using this 3D device in daily practice, the standardization of the scanning process needs to be studied further. It is important to recognize these problems before larger use of these devices in breast volume evaluation in a daily practice. In the current study, we utilized the 3D Artic Leo Scanner for breast imaging, but it is crucial to recognize that there are numerous other readily available devices, that could be employed for similar purposes. This series of 50 patients serves as a pilot feasibility study to identifying barriers to implementation of 3D scanning in daily general practice.

Acknowledgments

We thank our study nurses Janika Pietilä, Marketta Rautio and Meri Järvinen for performing the 3D scanning.

Footnote

Reporting Checklist: The authors have completed the TREND reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-1-577/rc

Data Sharing Statement: Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-1-577/dss

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-1-577/prf

Funding: This work was supported by state research funding (VTR T66374).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2025-1-577/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional scientific center ethics board of Tampere University Hospital (No. R22053). Written informed consent forms were obtained from the study participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Lin Y, Yang Y, Zhang X, et al. Postoperative Health-related Quality of Life in Reduction Mammaplasty: A Systematic Review and Meta-Analysis. Ann Plast Surg 2021;87:107-12.
2. Saariniemi KM, Sintonen H, Kuokkanen HO. The improvement in quality of life after breast reduction is comparable to that after major joint replacement. Scand J Plast Reconstr Surg Hand Surg 2008;42:194-8.
3. Saariniemi KM, Keranen UH, Salminen-Peltola PK, et al. Reduction mammaplasty is effective treatment according to two quality of life instruments. A prospective randomised clinical trial. J Plast Reconstr Aesthet Surg 2008;61:1472-8.
4. Gupta N, Lava CX, Ferdousian S, et al. Resection Weight Formulas for Reduction Mammoplasty: A Systematic Review and Regression Analysis. Ann Plast Surg 2024;93:643-8.
5. Yan M, Bustos SS, Kuruoglu D, et al. Breast Resection Weight Prediction and Insurance Reimbursement in Reduction Mammaplasty: Which Scale Is Reliable? Plast Reconstr Surg 2022;150:723e-30e.
6. Boukovalas S, Padilla PL, Spratt H, et al. Redefining the Role of Resection Weight Prediction in Reduction Mammaplasty and Breaking the "One-Scale-Fits-All" Paradigm. Plast Reconstr Surg 2019;144:18e-27e.
7. Park JB, Rahmani B, Adebagbo OD, et al. Impact of sternal notch-to-nipple distance and resection weight asymmetry on complication rates and patient reported outcomes in reduction mammaplasty. J Plast Reconstr Aesthet Surg 2024;88:340-3.
8. Li H, Lin Y, Li Z, et al. A Predictive Model for Breast Tissue Resection Weight in Vertical-Scar Reduction Mammoplasty. Aesthetic Plast Surg 2025;49:4006-12.
9. Killaars RC, Preuß MLG, de Vos NJP, et al. Clinical Assessment of Breast Volume: Can 3D Imaging Be the Gold Standard? Plast Reconstr Surg Glob Open 2020;8:e3236.
10. Howes BH, Watson DI, Fosh B, et al. Magnetic Resonance Imaging Versus 3-Dimensional Laser Scanning for Breast Volume Assessment After Breast Reconstruction. Ann Plast Surg 2017;78:455-9.
11. Kim J, Matsumine H, Niimi Y, et al. Estimation of Mastectomy Volume Using Preoperative Mastectomy Simulation Images Acquired by the Vectra H2 System. Plast Reconstr Surg Glob Open 2023;11:e5180.
12. Kyriazidis I, Berner JE, Waked K, et al. 3D Breast Scanning in Plastic Surgery Utilizing Free iPhone LiDAR Application: Evaluation, Potential, and Limitations. Aesthet Surg J 2025;45:NP99-104.
13. Kalbhen CL, McGill JJ, Fendley PM, et al. Mammographic determination of breast volume: comparing different methods. AJR Am J Roentgenol 1999;173:1643-9.
14. Goto H, Fukumoto S, Hara Y, et al. Use of Three-Dimensional Surface Imaging to Measure Breast Volume in the Upright Position With Acceptable Accuracy. Eur J Breast Health 2025;21:230-6.
15. Sirimahachaiyakul P, Boonjunwetwat D, Meevassana J, et al. Weight to volume conversion: an easy and practical breast volume estimation. Gland Surg 2023;12:1387-94.
16. Tong OLH, Chamson-Reig A, Yip LCM, et al. Structured-light surface scanning system to evaluate breast morphology in standing and supine positions. Sci Rep 2020;10:14087.
17. Kawale M, Lee J, Leung SY, et al. 3D Symmetry Measure Invariant to Subject Pose During Image Acquisition. Breast Cancer (Auckl) 2011;5:131-42.
18. Kovacs L, Yassouridis A, Zimmermann A, et al. Optimization of 3-dimensional imaging of the breast region with 3-dimensional laser scanners. Ann Plast Surg 2006;56:229-36.