Reimagining blepharoplasty: the role of three-dimensional imaging and artificial intelligence in personalized eyelid surgery

Blepharoplasty remains one of the most nuanced procedures within oculoplastic surgery, demanding precise attention to anatomical detail and surgical technique. This procedure targets key anatomical structures, including eyelid skin, orbicularis oculi muscle, and orbital fat, to simultaneously address aesthetic concerns and functional impairments. Clinically, upper eyelid blepharoplasty (UEB) is predominant, typically involving skin excision, orbicularis muscle resection or tightening, orbital fat removal or repositioning, as well as correction of eyelid crease formation and ptosis. In contrast, lower eyelid blepharoplasty primarily focuses on aesthetic rejuvenation through skin tightening, fat removal or repositioning, muscle suspension, and correction of eyelid laxity. The success of blepharoplasty depends heavily on submillimeter surgical precision, which can be challenging due to the dynamic anatomy of the orbital area and subjective traditional assessments.

Conventional preoperative planning relies on 2D photography and clinical examination to evaluate parameters such as margin-reflex distance (MRD), skin redundancy, and fat prolapse. However, these methods suffer from inherent limitations: photographic distortion can alter true anatomical proportions, while manual measurements often show significant inter-observer variability. Such inconsistencies risk asymmetrical outcomes, under correction, or over-resection.

Recent advancements in three-dimensional (3D) surface imaging have significantly improved the accuracy and consistency of periocular surgical planning. Several 3D imaging modalities are currently utilized in aesthetic and reconstructive surgery, each offering distinct advantages (Figure 1). Stereophotogrammetry, used widely in systems like Canfield Vectra, creates accurate 3D models from multiple photos. This method achieves submillimeter accuracy (mean error: 0.1–0.3 mm) while being non-invasive and radiation-free (1,2). Laser-based scanning captures surface topography with exceptional depth resolution by analyzing laser reflections from the skin surface. While highly accurate, this technique is more time-consuming and sensitive to patient movement. Structured light scanning, such as Bellus3D, projects a pattern onto the skin and calculates distortions to generate a 3D model, offering rapid capture speeds and affordability compared to laser scanning (3).

Figure 1 Representative 3D imaging modalities used in facial analysis. Canfield VECTRA M3 stereophotogrammetry system (A). Canfield VECTRA H2 portable handheld 3D imaging system (B). Konica Minolta Vivid 910 laser-based scanner system (C). Bellus3D Face Camera Pro, a structured light-based system (D). 3D, three-dimensional.

These 3D imaging methods provide substantial advantages over traditional 2D imaging. Standardized imaging protocols also allow consistent and objective comparisons over time. As artificial intelligence (AI) continues to integrate with 3D imaging, automatic landmark detection, and predictive modeling are further refining its clinical applications, paving the way for more personalized, precision-driven blepharoplasty.

Despite recent improvements in surgical techniques and imaging methods, clear quantitative standards for surgical decisions remain lacking in periocular surgery. Choosing between levator resection and frontalis suspension is highly surgeon-dependent for ptosis correction, resulting in reoperation due to inadequate margin-reflex distance (MRD1) adjustments. Lower eyelid rejuvenation lacks objective guidelines for fat redistribution, leading to postoperative complications such as hollowing or ectropion. Furthermore, ethnic-specific procedures, such as double eyelid creation in Asian patients, suffer from the absence of standardized anthropometric criteria, thus increasing the risk of postoperative asymmetry (4). Traditional outcome assessments, such as the FACE-Q and Barton grading system, rely on subjective grading scales that may not accurately reflect objective three-dimensional contour changes. Although emerging AI and 3D imaging technologies, including deep-learning-based landmark detection and predictive morphing, show promise, their application in periocular surgery remains limited and requires further investigation.

Our perspective centers on three interconnected key areas: ethnographic adaptation of surgical techniques, rigorous clinical validation of imaging methodologies, and the integration of AI for enhanced surgical planning and outcome prediction.

The integration of 3D imaging represents a shift in blepharoplasty, overcoming fundamental limitations of two-dimensional assessments through millimeter-precision quantification and dynamic visualization. Beyond technical advancement, its true clinical value emerges in two critical domains: enabling ethnographically tailored approaches and establishing evidence-based validation frameworks—both essential for standardizing surgical decisions amid anatomical diversity.

Periocular anatomy varies significantly across ethnic groups, with differences in eyelid crease formation, orbit fat distribution, skin thickness, and overall morphology. These variations directly impact surgical planning and outcomes in blepharoplasty, making ethnically-tailored approaches essential for optimal results. Traditional two-dimensional measurements often fail to capture these nuanced anatomical differences, leading to standardized surgical techniques that may not address patient-specific needs. 3D imaging has emerged as a critical tool for documenting, quantifying, and addressing these ethnic variations, as summarized in Table 1. 3D imaging has enabled the development of comprehensive, population-specific morphometric databases that provide surgeons with evidence-based reference points for diverse patient populations. These databases capture key anatomical metrics with submillimeter precision, allowing for truly individualized surgical planning. Lu et al. quantified significant ethnic differences in upper eyelid crease height (UECH), reporting averages of 4.91 mm for Chinese populations versus 8.33 mm for Malay populations, as well as variations in pretarsal skin height, highlighting the necessity for distinct surgical approaches when creating or modifying eyelid creases (6). Further, ethnic-specific 3D analyses have been conducted for other periocular features. Chi et al. validated a standardized 3D imaging protocol for lower eyelid anthropometry involving 58 participants, demonstrating excellent measurement reliability across diverse Asian phenotypes [intra-rater intraclass correlation coefficient (ICC) ≥0.95, MAD =0.22 units; inter-rater ICC ≥0.95, MAD =0.53 units; intra-method ICC ≥0.95, MAD =0.71 units] (12).

Gao et al. performed a detailed comparative analysis between Caucasians and Asians using 3D stereophotogrammetry, highlighting significant ethnic differences. Caucasians had notably larger iris diameter, lower palpebral margin length, lateral canthal angle, canthal tilt, and canthal angular index, but smaller inner and outer intercanthal distances and canthal indices compared to Chinese subjects. Additionally, Caucasians exhibited generally more prominent double-eyelid folds, underscoring the necessity for ethnically tailored surgical approaches (5). Moreover, Ju et al. assessed periocular asymmetry in a Caucasian population using 3D imaging, revealing that asymmetry, particularly in eyelid fissure height and upper eyelid crease, increased significantly with age and was more pronounced in males. For instance, upper eyelid crease asymmetry was higher in males (0.90±0.94 mm) compared to females (0.65±0.79 mm), emphasizing the importance of incorporating age and gender considerations into surgical planning for Caucasian patients (7).

By establishing ethnically-relevant reference values and thresholds, 3D imaging enables surgeons to move beyond the “one-size-fits-all” approach that has dominated traditional blepharoplasty. These precise, population-specific benchmarks offer a robust foundation for developing customized surgical techniques that respect and preserve ethnic identity while addressing individual aesthetic goals and functional concerns.

3D imaging has undergone clinical validation across multiple metrics, demonstrating superior reliability, accuracy, and clinical utility compared to traditional assessment methods in blepharoplasty.

In upper eyelid analysis, Guo et al. validated 3D imaging for upper eyelid area measurement, achieving high reliability (ICC: 0.982 intra-rater, 0.969 inter-rater, 0.917 intra-method) with low measurement error (MAD ≤0.36 cm², relative error of measurement ≤6.5%), supporting its precision for surgical planning, such as skin excision (13). Additional studies employed 3D imaging to analyze eyelid dermatochalasis and aging in greater detail, quantifying morphological changes associated with periocular aging, including palpebral fissure dimensions, crease positions, and soft tissue laxity (9,10,14). Additional validation was provided by Qu et al., who employed 3DSI to quantitatively evaluate surgical outcomes of combined subbrow blepharoplasty and brow fat pad transfer in treating dermatochalasis and upper eyelid depression, reporting significant reductions in depression depth (preoperative 1.817 mm vs. postoperative 1.345 mm) and volume (preoperative 0.212 cm³ vs. postoperative 0.162 cm³) at 6 months postoperatively (15). These findings were corroborated by consistent patient and physician satisfaction scores, thus validating 3D imaging as an objective, reliable tool for assessing surgical effectiveness and aesthetic outcomes in clinical practice.

For lower eyelids, the volumetric assessments of 3DSI offer advantages over conventional methods. Cristel et al. demonstrated precise mapping of lower eyelid fat compartments, quantifying postoperative fat redistribution with mean volume gains of 2.84 mL (right eye) and 2.87 mL (left eye) (16). This precision enables surgeons to avoid over-resection while maintaining natural contours—a significant improvement over subjective visual assessments traditionally used. In support of these findings, Miller et al. documented sustained volume increases of 0.64 mL in the tear trough area at 12 months post-fat repositioning, establishing 3D imaging as superior to 2D photography for standardized, reproducible outcome tracking across multiple time points (17).

Despite these advances, certain limitations remain. Fan et al. reported variable reliability in measuring upper eyelid area and volume using portable (VECTRA H2) and static (VECTRA M3) 3D imaging systems. Specifically, excellent intra-rater reliability was observed for the upper eyelid area (ICC: H2=0.985, M3=0.992), whereas poor inter-method reliability was noted for upper eyelid volume measurements (ICC =0.178), likely due to anatomical complexity and methodological limitations (18). Further, Miranda et al. found systematic discrepancies between preoperative 3D volume simulations and actual surgical outcomes, with postoperative eyelid volumes exceeding simulated predictions by 0.30 mL (right eye) and 0.24 mL (left eye) at six months (19). These findings reveal limitations in current 3D modeling algorithms that don’t account for dynamic tissue properties, highlighting the need for standardized imaging protocols not required with traditional methods.

Building on the established imaging capabilities and clinical validation discussed previously, the integration of AI with 3D imaging represents the next evolutionary step in periocular assessment. It will revolutionize surgical workflows and further enhance the precision and predictive accuracy of blepharoplasty procedures.

A critical advancement offered by AI in oculoplastic surgery is the standardization of eyelid measurements, historically subject to significant inter-examiner variability. Modern AI algorithms, particularly convolutional neural networks (CNNs), enable automated segmentation of eyelid structures in both photographs and 3D scans, accurately quantifying essential metrics such as palpebral fissure width, eyelid crease position, and brow height. Al-Baker et al. demonstrated this potential with their patch-based CNN model trained on 408 annotated 3D facial scans, achieving exact detection of 37 facial landmarks with a mean localization error of 0.83±0.49 mm (20). This algorithm surpassed manual precision by accurately localizing 95% of landmarks within 1 mm, showing particular effectiveness for periocular structures such as the medial canthus and eyelid crease.

Complementing deep learning approaches, optimization algorithms have further refined quantitative assessments and surgical planning. Liu et al. developed a structured light scanning method with a histogram of oriented gradient (HOG) active appearance model to detect 36 key landmarks and calculate 22 eyelid-related parameters. Their system demonstrated high precision with relative errors <10% in normal subjects and <13% in thyroid-associated ophthalmopathy patients (21).

These automated 3D methodologies significantly improved objectivity and reproducibility, reducing inter-observer variability and enhancing predictive accuracy for functional and aesthetic eyelid surgeries.

Consequently, surgeons enter the operating room equipped with standardized, objective eyelid metrics, enhancing surgical planning accuracy and providing clear postoperative benchmarks.

In blepharoptosis surgery, AI facilitates more informed decision-making. Song et al. developed an AI-based decision model using a gradient-boosted decision tree to analyze 2D and 3D facial features, detect ptosis, and advise appropriate surgical approaches. Their system improved accuracy when combining 2D and 3D data, enhancing early detection and surgical planning, especially for non-specialists (22). This highlights that 3D analysis is not just academic, but it can lead to measurably better clinical outcomes, probably by guiding surgeons to remove or reposition tissue more precisely.

In the future, integrating 3D imaging and AI will transform blepharoplasty by moving beyond static anatomical assessments toward dynamic, real-time intraoperative guidance and comprehensive postoperative evaluation. Emerging technologies are poised to significantly enhance surgical precision, safety, and personalized patient outcomes.

Advances in high-frame-rate stereophotogrammetry (≥60 fps) show promising potential for future applications in real-time functional analyses of eyelid kinetics, which may enable precise quantitative measurement of eyelid movement patterns, lagophthalmos severity and ptosis measurement in clinical settings (23,24). AI-driven dynamic models could further refine surgical candidacy selection by correlating levator muscle function with dynamic MRD1 variations, achieving predictive accuracies for outcomes of frontalis suspension procedures.

The development of portable, AI-powered 3D scanners and depth-sensing RGB-D cameras facilitates real-time intraoperative navigation, effectively serving as a surgical “GPS” (25). These systems provide surgeons with live, continuously updated 3D visualizations of facial anatomy, surgical instruments, and tissue adjustments.

AI-enhanced intraoperative visualization technologies are particularly impactful in complex orbital and eyelid reconstructive surgeries. By integrating patient-specific preoperative 3D imaging with real-time surgical tracking, surgeons receive immediate feedback regarding anatomical landmarks (e.g., optic nerve, levator aponeurosis) and precise guidance for surgical adjustments. Such technologies facilitate precise orbital decompressions and eyelid reconstructions, dynamically highlighting vital structures to minimize risks and optimize outcomes.

Despite its advantages, 3D surface imaging has notable limitations: it cannot visualize internal eyelid structures (such as the tarsal plate and muscles), nor can it capture transparent elements like the tear film and cornea. Future integration of surface imaging with functional and internal anatomical modalities may offer more comprehensive surgical insights by addressing these constraints. Integrating with complementary imaging technologies such as optical coherence tomography (OCT) and magnetic resonance imaging (MRI), particularly when assisted by AI algorithms, could provide more comprehensive surgical analysis by combining surface morphology with internal anatomical information.

Integrating 3D imaging and AI significantly enhances blepharoplasty by providing precise, reproducible anatomical measurements and standardized surgical planning. AI-driven automated landmark detection reduces variability and improves predictive accuracy. Ethnically tailored morphometric databases optimize outcomes across diverse populations. Future developments in real-time intraoperative navigation, dynamic functional analysis, and multi-modal data integration will further enhance precision and personalization. Ultimately, these innovations empower surgeons to make better decisions, reduce complications, and achieve outcomes that align more closely with patients’ functional needs and aesthetic aspirations.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Frontiers of Oral and Maxillofacial Medicine. The article has undergone external peer review.

Peer Review File: Available at https://fomm.amegroups.com/article/view/10.21037/fomm-25-9/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://fomm.amegroups.com/article/view/10.21037/fomm-25-9/coif). L.M.H. serves as an unpaid editorial board member of Frontiers of Oral and Maxillofacial Medicine from September 2024 to August 2026. The other 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.

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. Singh P, Bornstein MM, Hsung RT, et al. Frontiers in Three-Dimensional Surface Imaging Systems for 3D Face Acquisition in Craniofacial Research and Practice: An Updated Literature Review. Diagnostics (Basel) 2024;14:423. [Crossref] [PubMed]
2. Verhulst A, Hol M, Vreeken R, et al. Three-Dimensional Imaging of the Face: A Comparison Between Three Different Imaging Modalities. Aesthet Surg J 2018;38:579-85. [Crossref] [PubMed]
3. Parsa S, Basagaoglu B, Mackley K, et al. Current and Future Photography Techniques in Aesthetic Surgery. Aesthet Surg J Open Forum 2022;4:ojab050. [Crossref] [PubMed]
4. Yu AY, Yu AA, Boonipat T, Pu LLQ. A Conservative Crease-Height-Adjustable Asian Double-Eyelid Surgery Technique Based on Live Anatomical Studies. Plast Reconstr Surg 2025;156:29-40. [Crossref] [PubMed]
5. Gao T, Guo Y, Rokohl AC, et al. Racial and sexual differences of eyebrow and eyelid morphology: three-dimensional analysis in young Caucasian and Chinese populations. Quant Imaging Med Surg 2025;15:882-97. [Crossref] [PubMed]
6. Lu TY, Kadir K, Ngeow WC, et al. The Prevalence of Double Eyelid and the 3D Measurement of Orbital Soft Tissue in Malays and Chinese. Sci Rep 2017;7:14819. [Crossref] [PubMed]
7. Ju X, Rokohl AC, Fan W, et al. Periocular Asymmetry Index in Caucasian Populations Using Three-dimensional Photogrammetry Assessment. Aesthetic Plast Surg 2024;48:4489-99. [Crossref] [PubMed]
8. Jayaratne YS, Deutsch CK, Zwahlen RA. Normative findings for periocular anthropometric measurements among Chinese young adults in Hong Kong. Biomed Res Int 2013;2013:821428. [Crossref] [PubMed]
9. Chong Y, Li J, Liu X, et al. Three-dimensional anthropometric analysis of eyelid aging among Chinese women. J Plast Reconstr Aesthet Surg 2021;74:135-42. [Crossref] [PubMed]
10. Flores CA, Mundy JL, Byrne ME, et al. Quantitative 3-dimensional Geometry of the Aging Eyelids. Plast Reconstr Surg Glob Open 2019;7:e2512. [Crossref] [PubMed]
11. Kim YC, Kwon JG, Kim SC, et al. Comparison of Periorbital Anthropometry Between Beauty Pageant Contestants and Ordinary Young Women with Korean Ethnicity: A Three-Dimensional Photogrammetric Analysis. Aesthetic Plast Surg 2018;42:479-90. [Crossref] [PubMed]
12. Chi Y, Yang Y, Jin L, et al. Protocol Establishment and Reliability Verification of Three-Dimensional Digital Stereophotogrammetry in Lower Eyelid Anthropometry. Aesthetic Plast Surg 2024;48:1276-87. [Crossref] [PubMed]
13. Guo Y, Rokohl AC, Fan W, et al. A novel standardized approach for the 3D evaluation of upper eyelid area and volume. Quant Imaging Med Surg 2023;13:1686-98. [Crossref] [PubMed]
14. Li X, Rokohl AC, Fan W, et al. Quantifying Dermatochalasis Using 3-Dimensional Photogrammetry. Aesthetic Plast Surg 2024;48:1288-97. [Crossref] [PubMed]
15. Qu L, Liang Z, Wang J, et al. Combined Blepharoplasty and Brow Fat Pad Transfer for the Correction of Dermatochalasis and Sunken Upper Eyelids. Plast Reconstr Surg 2024;153:701e-10e. [Crossref] [PubMed]
16. Cristel RT, Caughlin BP. Lower Blepharoplasty Three-Dimensional Volume Assessment after Fat Pad Transposition and Concomitant Fat Grafting. Facial Plast Surg 2020;36:478-83. [Crossref] [PubMed]
17. Miller TR. Long-term 3-Dimensional Volume Assessment After Fat Repositioning Lower Blepharoplasty. JAMA Facial Plast Surg 2016;18:108-13. [Crossref] [PubMed]
18. Fan W, Rokohl AC, Maus J, et al. Evaluation of the Portable Next-Generation VECTRA H2 3D Imaging System for Measuring Upper Eyelid Area and Volume. Aesthet Surg J 2023;43:1114-23. [Crossref] [PubMed]
19. Miranda RE, Matayoshi S. Vectra 3D Simulation in Lower Eyelid Blepharoplasty: How Accurate is it? Aesthetic Plast Surg 2022;46:1241-50. [Crossref] [PubMed]
20. Al-Baker B, Ayoub A, Ju X, et al. Patch-based convolutional neural networks for automatic landmark detection of 3D facial images in clinical settings. Eur J Orthod 2024;46:cjae056. [Crossref] [PubMed]
21. Liu N, Liang G, Li L, et al. An eyelid parameters auto-measuring method based on 3D scanning. Displays 2021;69:102063.
22. Song X, Tong W, Lei C, et al. A clinical decision model based on machine learning for ptosis. BMC Ophthalmol 2021;21:169. [Crossref] [PubMed]
23. Tzou CH, Artner NM, Pona I, et al. Comparison of three-dimensional surface-imaging systems. J Plast Reconstr Aesthet Surg 2014;67:489-97. [Crossref] [PubMed]
24. Lootus M, Beatson L, Atwood L, et al. Development and Assessment of an Artificial Intelligence-Based Tool for Ptosis Measurement in Adult Myasthenia Gravis Patients Using Selfie Video Clips Recorded on Smartphones. Digit Biomark 2023;7:63-73. [Crossref] [PubMed]
25. Udhay P, Bhattacharjee K, Ananthnarayanan P, et al. Computer-assisted navigation in orbitofacial surgery. Indian J Ophthalmol 2019;67:995-1003. [Crossref] [PubMed]