3D Breast Scanning in Plastic Surgery Utilizing Free iPhone LiDAR Application: Evaluation, Potential, and Limitations

Abstract

Background

Three-dimensional (3D) imaging enhances surgical planning and documentation in plastic surgery, but high costs limit accessibility. Mobile light detection and ranging (LiDAR) technology offers a potential cost-effective alternative.

Objectives

The objective of this research was to evaluate the accuracy and clinical utility of iPhone-based LiDAR scanning for breast measurements compared to traditional methods, and to establish standardized protocols for clinical implementation.

Methods

In this prospective validation study, 25 consecutive patients (mean age: 44 years; range: 32-64 years; mean BMI: 23.2 kg/m²) undergoing breast procedures were evaluated with the “3D Scanner App” on the iPhone 15 Pro (Apple Inc., Cupertino, CA). Three standardized measurements (sternal notch-to-nipple, nipple-to-midline, nipple-to-inframammary fold) of LiDAR and manual techniques were compared. Technical error of measurement (TEM) and relative TEM (rTEM) were calculated. Interrater reliability, learning curve assessment, and cost-effectiveness analysis were performed.

Results

LiDAR measurements showed very good accuracy for sternal notch-to-nipple (rTEM 1.43%, 95% CI: 1.21-1.65) and nipple-to-midline (rTEM 2.83%, 95% CI: 3.12-3.78) distances. Nipple-to-inframammary fold measurements showed poor accuracy (rTEM 13.05%, 95% CI: 8.21-9.39). Interrater reliability was excellent (ICC = 0.92). Learning curve analysis demonstrated measurement stability after 5 cases. Cost analysis revealed 97.8% reduction in initial investment compared to commercial 3D imaging systems.

Conclusions

Mobile LiDAR offers a cost-effective tool for breast surgery planning and documentation. While measurements like sternum-to-nipple and nipple-to-midline are highly accurate, the relatively poor accuracy of nipple-to-IMF highlights limitations with complex curves. As this technology continues to evolve, further improvements in accuracy are anticipated, expanding its role in clinical use.

Level of Evidence: 4 (Therapeutic)

Audio Abstract

Conventional medical photography has inherent limitations in accurately capturing volumetric details, necessitating a shift toward 3-dimensional (3D) modeling of patients for various medical applications. The utilization of 3D photography and computer modeling has been demonstrated to significantly enhance the learning curve for surgical trainees in reconstructive facial surgery, improving both knowledge and satisfaction levels.^1,2 Plastic surgery has been at the forefront of adopting and innovating 3D technology applications.^2-6

Traditionally, 3D imaging in plastic surgery has been dominated by systems such as Vectra (Canfield Scientific, Parsippany, NJ). These systems, including the Vectra M1, M3, M5, XT, and H1, are recognized for their accuracy and reliability in capturing 3D images for various clinical applications, including breast reconstructive surgery, facial morphology assessment, and lymphedema volume measurement.^7-9 However, the high cost of these systems often limits their accessibility.⁷

Light detection and ranging (LiDAR) technology operates on the principle of light reflection. A light beam is directed towards a surface, and the time taken for the beam to reflect back to its source is precisely measured. By analyzing the wavelength and travel time of the reflected light, the LiDAR system can accurately compute distances. These measurements are then utilized to generate a detailed digital representation of the object. This technology has demonstrated potential applications in 3D breast scanning, offering precise and noninvasive imaging capabilities.¹⁰

The incorporation of LiDAR technology in smartphones has significantly increased the accessibility of 3D scanning.¹¹ In a recent study an attempt was made to demonstrate the effectiveness of LiDAR technology in breast scanning by developing a software application based on iOS devices with built-in LiDAR sensors.¹²

Recognizing the proficiency of existing applications in the field of LiDAR, photogrammetry, neural radiant fields (NeRF), and augmented reality technologies, we opted to utilize the commercially available 3D Scanner App by Laan Labs (New York, NY), rather than developing a new application.¹³ This approach diverges from that outlined in the previously mentioned study.¹²

The 3D Scanner App is a commercially available, free application that utilizes the iPhone (Apple Inc., Cupertino, CA) LiDAR sensor for 3D scanning as well as distance and volume measurements.

The primary objective of this study was to evaluate the accuracy of the 3D Scanner App for breast measurements compared to traditional tape measurements in females undergoing breast reconstruction. Specifically, we aimed to determine whether the app could produce reliable and accurate 3D models that could effectively provide 3D documentation of patients before and after surgery. The hypothesis was that the app-based measurements would demonstrate good agreement with manual measurements and offer advantages of accessibility, cost-effectiveness, and utility in 3D documentation.

METHODS

Study Design and Participants

This prospective study was conducted with data collection planned before the index test and reference standard were performed. Twenty-five females scheduled for breast reconstruction were consecutively enrolled from our institution's plastic surgery department between November 2023 and May 2024. Eligibility criteria included females ages 18 years or older willing to participate in the study, undergoing aesthetic breast augmentation, breast mastopexy, or reduction. Exclusion criteria included a body mass index > 28, current pregnancy, and pseudoptosis. The study was conducted in accordance with the guidelines set forth in the Declaration of Helsinki. Informed consent was provided by patients.

Sample Size

A sample size of 25 participants was determined based on feasibility and the exploratory nature of this study.

Index Test

The 3D Scanner App on an iPhone 15 Pro was utilized to create 3D models of the breast in patients undergoing reconstructive and aesthetic procedures. The app’s built-in measurement tools were employed to measure sternal notch-to-nipple distance, nipple-to-inframammary fold distance, and nipple-to-midline distance. Scans were performed by I.K., who has experience with 3D scanning technologies and software and was blinded to the manual measurement results (Figures 1A, 1B).

Reference Standard

To assess interrater reliability, 2 surgeons (M.H., K.W.) performed manual tape measurements, and LiDAR scanning was conducted by the surgeons (I.K., K.W.). Each observer performed measurements twice. All measurements were recorded independently, and observers were blinded to each other's results. For tape measurements, measurements were recorded without access to LiDAR scan data, and vice versa. Tape measurement was chosen as the reference standard due to its widespread use in clinical practice.

Test positivity cut-offs were not applicable, because the measurements were continuous variables. Clinical information was available to plastic surgeons performing both the app measurements and the manual measurements.

Standardized Scanning Protocol

1. Environment setup

   • ○  Room lighting: 800-1000 lux diffuse lighting

   • ○  Background: Nonreflective blue surgical drape

   • ○  Patient positioning: Standing, arms at sides, marked reference points

   • ○  Device distance: 1-1.5 meters from patient

2. Patient positioning

   • ○  Patients were positioned standing with arms relaxed down and gently on the back, maintaining natural breast position and chest wall dynamics. The standardized positioning also facilitated comparison with traditional clinical photography and anthropometric measurements.

3. Scanning technique

   • ○  Device held at chest height

   • ○  Circular movement pattern at constant speed (approximately 10 seconds per revolution)

   • ○  Three complete revolutions per scan

   • ○  Real-time quality assessment with app's mesh visualization

   • ○  Immediate scan verification and repetition if quality metrics not met

If any quality issues were identified, such as holes in the mesh, blurred areas, or misaligned sections, the entire scan was discarded and repeated rather than attempting partial rescans. This comprehensive approach to quality control ensures consistent data quality and prevents mesh artifacts that could affect measurement accuracy.

• 4. Measurement protocol

  • ○  Anatomical landmarks marked prescan

  • ○  Three independent measurements per distance

  • ○  Automated and manual measurement comparison

The sternal notch-to-nipple (SN-N) distance was measured vertically as a straight vertical line, as it is done with the measuring tape. This approach was consistently applied in both LiDAR and manual tape measurements to ensure comparability.

Data Analysis

To directly compare the measurements from the 2 methods (software vs tapeline) and calculate the error ratios, the technical error of measurement (TEM) and the relative technical error of measurement (rTEM) were utilized. TEM measured the precision of the measurements by calculating the standard deviation of the differences between the 2 sets of measurements, and rTEM expressed this error relative to the mean measurement. The TEM was calculated with the following formula,

in which D represented the difference between the app and tape measurements for each data point, and N represented the number of participants. The rTEM was then calculated as

The TEM values were interpreted as follows: <1% excellent, 1-3.9% very good, 4-6.9% good, 7-9.9% moderate, >10% poor.¹⁴ There were no indeterminate results. Any missing data points were excluded from the analysis.

RESULTS

Twenty-five females (mean age: 44 years, range: 32-64 years) were recruited, meeting the inclusion criteria and agreeing to participate in the study. Mean BMI was 23.2 kg/m² (range: 19.1 to 27.3 kg/m²). Fourteen patients underwent breast reduction, 7 underwent bilateral mastopexy, and 4 cases involved aesthetic breast augmentation. All 3D models produced were sharp, clean, and high quality.

Interrater reliability analysis demonstrated high consistency for both measurement methods. For LiDAR scanning, intraclass correlation coefficient (ICC) values were: sternal notch-to-nipple distance 0.95 (95% CI: 0.92-0.97), nipple-to-midline distance 0.93 (95% CI: 0.90-0.95), and nipple-to-IMF distance 0.82 (95% CI: 0.78-0.86). For manual tape measurements, ICC values were consistently higher: sternal notch-to-nipple distance 0.96 (95% CI: 0.93-0.98), nipple-to-midline distance 0.95 (95% CI: 0.92-0.97), and nipple-to-IMF distance 0.97 (95% CI: 0.94-0.99). These results demonstrate excellent reliability for both methods, with manual measurements showing superior consistency for IMF measurements; direct tissue palpation provides an advantage over surface scanning technology.

The TEM and rTEM values between the application and the tapeline measurements were calculated. The rTEM values were 1.43% (very good) for sternal notch-to-nipple distance, 2.83% (very good) for nipple-to-midline distance, and 13.05% (poor) for nipple-to-inframammary fold distance (Table 1).

DISCUSSION

The LiDAR-based application demonstrated commendable accuracy in measuring anatomical landmarks, with rTEM values ranging from 1.39% to 2.89% for most parameters. However, the nipple-to-inframammary fold distance showed higher error rates (rTEM 13.05%), which can be attributed to the challenges posed by natural breast ptosis and the inability of the app, and by extension LiDAR technology itself, to calculate the distance behind the natural curve of the breast from its most caudal aspect upward to the inframammary fold.

Our interrater reliability analysis demonstrated that manual tape measurements maintain superior consistency compared to LiDAR scanning, particularly for measurements involving the inframammary fold (ICC 0.96 vs 0.82). This finding aligns with the physical limitations of surface scanning technologies in accessing occluded anatomical landmarks, whereas experienced clinicians can directly access and measure these areas with greater precision.

Although commercial 3D imaging systems like Vectra have proven reliable, they face several inherent limitations too. The accuracy of virtual implant sizing simulations can be variable, particularly in cases of significant ptosis. The nipple-to-inframammary fold (N-IMF) measurements present particular challenges due to breast overhang and the dynamic nature of the IMF position. We found similar challenges with the iPhone-based system, suggesting that these are limitations of surface scanning technology rather than specific to any 1 system. Previous studies have shown that repeat measurements with Vectra systems can show variability between iterations.⁷ This variability needs to be considered when performing any 3D scanning technology for surgical planning.

Regarding volume measurements, the LiDAR application calculates breast volume based on user-defined boundaries (which act as the “base-boundary” of the scan) and the 3D mesh created during the scanning process. The volume is computed from the 3D mesh generated as the user moves around the patient, which may introduce minor artifacts or variations if not done correctly, compared to systems like Vectra XT, in which multiple fixed cameras capture the image simultaneously. Vectra's stationary multicamera array provides an advantage in mesh creation accuracy by eliminating potential motion artifacts.

The scanning protocol requires precise execution to avoid mesh reconstruction artifacts. Although our protocol involves 3 complete revolutions, each revolution captures a different angle and height to ensure complete coverage without rescanning the same areas. Rescanning previously captured areas can create mesh misalignment because the LiDAR sensor may register slightly different point cloud data when revisiting an area from a different angle or under different lighting conditions. This technical limitation necessitates smooth, deliberate scanning motions and careful attention to maintaining consistent distance and speed throughout the capture process. That said, we need to underline that as mobile computational power improves these misalignments and artifacts are likely to be diminished in the near future.

However, despite these limitations, our testing demonstrates that the iPhone-based system provides valuable 3D documentation and volumetric approximation in settings in which more expensive systems are unavailable. The practical utility of this technology is supported by several factors demonstrated in our study: (1) the significant cost reduction compared to commercial 3D systems; (2) excellent reliability for most anatomical measurements (rTEM 1.43% for sternal notch-to-nipple and 2.83% for nipple-to-midline distances); (3) the ability to generate high-quality, clean 3D models suitable for medical documentation; (4) rapid learning curve with measurement stability after 5 cases; and (5) the portable and readily accessible nature of the technology with widely available devices. The integration of LiDAR technology into mobile devices represents a significant advancement in plastic surgery, offering a practical and cost-effective alternative to traditional 3D scanning systems, making advanced documentation tools accessible to a broader range of healthcare professionals.

In this study, we demonstrated the commendable accuracy of the 3D Scanner App in measuring anatomical landmarks and volumes, with relative technical error of measurement (rTEM) values ranging from 1.43% to 2.83% for sternal notch-to-nipple and nipple-to-midline distances. However, the nipple-to-inframammary fold distance showed a higher error rate of 12.05%, attributed to the challenges posed by natural breast ptosis and the limitations of LiDAR technology in calculating distances behind the natural curve of the breast.

Our choice of patient positioning with arms at the sides was supported not only by our current practice measuring with tape, but also by recent evidence, which has demonstrated that arm position significantly affects breast volume measurements in 3D imaging.¹⁵ Although arm abduction might provide better access to lateral breast borders, it can alter chest wall dynamics and breast tissue position. Our protocol prioritized capturing the breast in its natural anatomical position, enabling reproducible measurements and facilitating comparison with traditional clinical photography and anthropometric measurements.

In this study, we utilized the 3D Scanner App for 3D breast imaging, but it is crucial to recognize that there are numerous other applications readily available that could be employed for similar purposes. However, it is important to consider that not all of these applications deliver the same quality of results, which can be attributed to the specific features and algorithms of each application.

As mobile technology continues to evolve at a rapid pace, it is anticipated that an increasing number of applications will enter the market, each striving to capitalize on the ever-advancing hardware capabilities of mobile devices. This trend is likely to drive the development of more sophisticated 3D scanning applications, leading to significant improvements in the accuracy and resolution of the 3D models produced.

The 3D Scanner App offers a user-friendly interface that allows for direct manipulation of models on the device, enhancing patient privacy and data security by avoiding cloud processing. The app’s capabilities extend beyond measuring distances, enabling the calculation of surface areas and volumes, which is particularly helpful for assessing breast volumes following reconstruction or augmentation. Additionally, the ability to superimpose and compare preoperative and postoperative models provides valuable insights for surgeons and patients alike.

Our findings align with recent validation studies comparing iPhone-based 3D scanning to Vectra systems.⁷ Rudy et al demonstrated comparable accuracy between iPhone and Vectra H2 measurements, with average discrepancies less than 1 mm for key anatomical landmarks. Their color map analysis showed mean differences of 0.53 mm between systems. Whereas their study focused on the Vectra H2, our work extends these findings by comparing LiDAR scanning with traditional manual measurements. Together, these studies suggest that smartphone-based 3D scanning represents a viable alternative to more expensive systems for basic breast measurements and documentation. The consistency between studies with different comparative standards (Vectra H2 vs manual measurements) strengthens the case for adopting this accessible technology in clinical practice.

Our study advances Han's et al work in several key ways.¹² First, we utilize a commercially available application, without trying to reinvent the wheel, making our approach immediately accessible to practitioners worldwide. Second, we provide comprehensive scanning protocols and quality control metrics that enable reproducibility. Third, the application we employed also has the capacity to calculate volume.

This study describes the application of LiDAR-based 3D scanning technology primarily for breast revision and reconstruction. However, the technology has potential applications in other areas, such as breast reconstruction and body contouring (Figure 2) and facial reconstruction (Figure 3). The capability to generate precise 3D models can significantly enhance surgical planning and patient outcomes in these procedures. It is important to note that the accuracy of the technology may be limited by the fineness of the anatomical structures being scanned. Specifically, finer structures like the ear and nose may not be captured with the same level of detail due to inherent limitations of current LiDAR technology, which will be discussed in subsequent paragraphs.

It is worth noting that the most crucial aspect of obtaining accurate 3D models is the user's scanning technique. Creating high-quality models requires practice, steady movements, and maintaining an appropriate distance from the patient. When scanning, it is crucial to move slowly and steadily, maintaining a constant speed to ensure proper texture capture and avoid holes in the mesh or blurry output. A spraying gesture, similar to painting with spray paint, can be applied to cover every part of the patient evenly. It is important not to rescan areas that have already been covered, because this may lead to alignment issues. If a full 360-degree model is not required, the scanning process becomes even more straightforward.

Lighting plays a significant role in the quality of the scan. Taking advantage of daylight can greatly improve the results, because LiDAR technology relies on light detection. Additionally, shiny or reflective surfaces, such as glasses, earrings and most metals, should be removed because they can interfere with the scanning process.

Finally, postprocessing is an essential step in achieving high-quality 3D models. The 3D Scanner App application offers in-app cropping tools, which can be applied to create a clean and polished final product. For more advanced editing, software on desktop computers is available, but this is beyond the scope of this article.

The primary limitations of LiDAR technology on mobile devices include the processing power of the devices themselves and the resolution of the models produced. These factors are influenced by the device's camera optics and LiDAR sensor capabilities, as well as the user's scanning technique. Despite these challenges, ongoing developments in mobile technology are expected to enhance the resolution and accuracy of these models in the near future.

It is worth acknowledging that more sophisticated systems, such as Vectra, offer a more streamlined approach to capturing a set of photographs from the patient, minimizing human-user error. These systems provide a controlled setting for capturing multiple angles with steady rotation, ensuring consistent and high-quality images. In addition to the higher-quality optic lenses and sensors, they also include specialized software that enables the simulation of different breast procedures, such as augmentation or reduction, which is not currently available in most mobile applications.

However, it is crucial to recognize that the primary limiting factors for mobile devices are the lack of specialized software and hardware capabilities to perform the entire processing on the device itself. As mobile technology continues to evolve, it is likely that these limitations will be addressed, allowing for more advanced features and improved accuracy in 3D scanning applications.

Limitations

Several important limitations of this study warrant discussion. The accuracy of LiDAR scanning shows dependence on both patient factors and operator technique. Although patient movement during scanning can introduce artifacts, there is also a notable learning curve for operators to achieve consistent, high-quality scans. This highlights an advantage of fixed multicamera systems like Vectra, in which simultaneous image capture eliminates motion artifacts and reduces operator dependence. Image quality can vary with ambient lighting conditions, although our standardized protocol helps mitigate this. Manual landmark selection is required because automated detection is not yet available, and long-term reliability across multiple follow-up visits needs further study. Although the system lacks some advanced features of dedicated 3D imaging platforms, our findings demonstrate its utility as an accessible 3D documentation tool when more sophisticated systems are unavailable.

The challenges mentioned in N-IMF measurement accuracy stem from 2 main factors. First, surface scanning technologies cannot directly capture the posterior surface where the breast meets the chest wall, particularly in ptotic breasts. Second, the sharp curvature at the IMF creates challenges for LiDAR point cloud generation. These technical limitations could be addressed through several future technological developments. Advanced mesh reconstruction algorithms could be specifically optimized for curved surfaces, and the implementation of machine learning–based landmark detection could improve measurement consistency. The integration of complementary sensing technologies might enhance tissue fold detection accuracy. Additionally, standardized patient positioning techniques could be developed to optimize IMF visualization without compromising overall scan quality. These improvements will likely emerge as the technology continues to evolve, particularly as LiDAR sensors become more sophisticated and processing algorithms more refined. In the interim, combining LiDAR measurements with traditional manual techniques remains the most reliable approach for accurate IMF assessment.

Despite these current limitations, the potential for mobile devices to provide accessible, cost-effective, and secure 3D imaging solutions in plastic surgery is significant. As more applications enter the market and take advantage of the ever-improving hardware capabilities of mobile devices, we can expect to see a narrowing gap between the performance of these applications and that of more sophisticated systems like Vectra. This development will ultimately benefit both surgeons and patients by providing more widely available tools for 3D documentation and analysis in plastic surgery.

CONCLUSIONS

LiDAR technology on iOS devices demonstrates potential as an effective solution for 3D breast imaging in plastic surgery. Although current limitations exist, particularly in measuring distances behind natural breast curves, the technology offers immediate benefits comparable to more sophisticated systems. 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. As mobile technology continues to advance, the accuracy and utility of 3D scanning applications are expected to improve further, potentially narrowing the gap between mobile applications and specialized 3D imaging systems.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The authors received no financial support for the research, authorship, and publication of this article.

REFERENCES

Author notes