Accuracy of virtual facebow compared to conventional facebow: a systematic review

Abstract

Background

Digital technologies are rapidly evolving and have been introduced across many fields in dentistry, particularly prosthetic and restorative dentistry. In digital treatment planning, virtual articulators are helpful in simulating complex maxillomandibular movements and analyzing occlusion. Despite these advancements, evidence-based literature standardizing the transfer of clinical data to virtual environments remains limited. The aim of this review was to explore the accuracy of various methods in transferring maxillary digital models to virtual articulators compared with conventional facebow techniques.

Methods

The research (PICO) question focused on assessing the accuracy of virtual facebow transfer compared to that of conventional facebows. An electronic search was performed in 3 databases between January 2014 to December 2025: PubMed/MEDLINE, Scopus, and Web of Science. Eligibility criteria included in vitro studies, clinical trials, case reports, clinical reports, and clinical techniques that evaluated the accuracy of virtual articulation in fixed prosthodontics for dentate patients, published in English. Studies were excluded if they were not in English, involved orthognathic surgery, focused on edentulous patients, or were published before 2014. A comprehensive search was conducted in December 2025. Two reviewers (PE, HA) independently examined the identified papers in accordance with the predetermined qualifying criteria. For each included study, publication details, study characteristics, and methodological aspects were documented.

Results

The total number of articles retrieved was 1,916. Of these articles, 10 met the inclusion criteria and were used for qualitative synthesis. Several techniques were reported for transferring the maxillary model to virtual articulator, such as digital photographs, face scans with or without intraoral or extraoral scan body, smartphone applications, digital photographs using Kois glasses, CBCT, and stereophotogrammetry. Most studies showed comparable accuracy of different digital methods that transfer the maxillary cast to the articulator. However, one study showed that the photography-based method demonstrated superior trueness and precision compared to both facial scanning and the conventional facebow technique. Another study found that the accuracy of positioning a maxillary digital scan relative to a 3D facial scan using a virtual facebow technique is highly dependent on the facial scanning method employed.

Conclusions

Virtual facebow transfer can be recorded using techniques such as digital photographs, face scans, virtual facebow software, and smartphone scanning. Virtual facebow systems demonstrate promising repeatability and clinically acceptable transfer accuracy in selected workflows; however, due to heterogeneity among available studies, definitive conclusions regarding any improvement over conventional techniques cannot yet be drawn. Further standardized in vitro and clinical studies with large sample sizes, standardized assessments, and comparison methods are needed to draw meaningful conclusions in order to prove precedence of one method over the other. However, the use of a device to facilitate the transfer of a virtual maxillary cast to a virtual articulator has been promising.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12903-026-08869-4.

Introduction

Accurate transfer of the maxillary cast into a Class III or semi-adjustable articulator is a fundamental step in Prosthodontic treatment planning and prosthesis fabrication [1]. Conventional transfer of the maxillary cast requires a facebow record to relate the maxilla to the transverse horizontal axis of the articulator [2]. Various reference planes and hinge-axis concepts have been described for this purpose, including the Frankfort plane, axis-orbital plane, Camper plane, arbitrary or true hinge axis, and the true horizontal (gravity) plane [2, 3]. The use of a facebow has been reported to facilitate accurate horizontal jaw relationship recording, appropriate vertical positioning of the maxillary cast within the articulator, and proper anteroposterior orientation relative to the mandibular condyles [4, 5]. However, clinical evidence supporting the superiority of a specific articulator system or facebow transfer method remains inconclusive [2].

The integration of computer-aided design (CAD) technology into prosthodontics has enabled the use of virtual articulators within digital workflows. In comprehensive rehabilitative cases, a facebow record is often required to transfer the maxillary digital scan into a virtual articulator to approximate patient-specific jaw movements. Virtual articulators are designed to replicate the structure and function of mechanical articulators [5–8]. The virtual facebow (VF) method mounts a digital scan of the maxillary arch to a virtual articulator based on its spatial relationship with one or more anatomic reference points [9, 10]. This relationship is typically established by aligning the maxillary digital scan with three-dimensional (3D) facial scan data. Accurate virtual facebow registration is essential for the correct positioning of the maxillary digital cast within the virtual articulator. Despite the increasing availability of VF techniques, limited comparative evidence is available regarding their clinical transfer accuracy, even with (or considering) rapid technological advances [5, 6, 8].

Multiple digital facebow methods have been described in the literature, including cephalometric radiographs, photographs, extraoral digitalization of analog facebow records, cone beam computed tomography (CBCT) images, facial scans alone or combined with an intraoral or extraoral scan body, or accompanied with CBCT images, in addition to jaw tracking systems that locate the true hinge axis [10–14]. Several studies have compared these systems with conventional articulation techniques in static and dynamic occlusion [15–17]. These studies generally reported comparable accuracy between virtual and mechanical articulation methods, supporting the clinical feasibility of virtual articulation in digital prosthodontic workflows.

With the advent of virtual articulators, intraoral scanners, virtual facebows, and virtual occlusal records, both the method and the final restoration design are now showing high accuracy and precision [18–20]. However, the accuracy of digital cast transfer is influenced by factors such as scanner precision, facial scanning technique, and the digital workflow employed. Moreover, the performance of facial scanners may vary depending on scanning protocol, acquisition time, and scanning technology [21, 22]. Recent reviews have highlighted increasing interest in virtual articulator systems and digital articulation workflows as alternatives to conventional mechanical articulation. These reviews suggest that virtual articulation offers advantages in digital integration, visualization, workflow efficiency, and clinician–laboratory communication. However, they also emphasize that articulation accuracy remains dependent on accurate digital data acquisition, registration protocols, facial landmark identification, and software alignment algorithms. Moreover, although promising results have been reported, evidence remains heterogeneous, and clinical validation remains ambiguous due to the variability in study methodologies and outcome measures [4, 23–25]. While recent systematic reviews have focused on the general accuracy of virtual articulation, evidence specifically evaluating virtual facebow transfer accuracy relative to conventional facebow transfer remains limited, justifying the need for the present focused systematic review.

The aim of this systematic review was to summarize and critically appraise the evidence on the accuracy of transferring a dentate maxillary model using virtual facebow in comparison to a conventional facebow. Based on the available literature, the null hypothesis was that no significant difference exists in transfer accuracy between virtual facebow techniques and conventional facebow transfer methods.

Materials & methods

Protocol and registration

This systematic review was retrospectively registered in the PROSPERO database (Registration No. CRD42024612659) and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) statement. A completed PRISMA 2020 checklist was submitted as a supplementary related file. The review methodology, including literature search, screening, eligibility assessment, and qualitative synthesis, was conducted following PRISMA’s recommendations.

Research question

The research question of this systematic review was: How accurate is virtual facebow transfer compared with conventional facebow transfer in mounting maxillary digital casts to articulators?

• P (Population): dentate patients / maxillary cast transfer

• I (Intervention): virtual facebow transfer techniques

• C (Comparison): conventional facebow transfer techniques

• O (Outcome): transfer accuracy (trueness / precision / reproducibility) In this review, “accuracy” was considered the conclusive term encompassing both trueness (closeness to a reference value) and precision (repeatability of repeated measurements). Because included studies used heterogeneous measurement methods and reported outcomes in different units (micrometers, millimeters, and angulations), direct quantitative comparison among studies was limited.

Information sources and search strategy

An electronic search was performed on 3 databases between January 2014 to December 2025: PubMed/MEDLINE, Scopus, and Web of Science using the following keywords: Articulator, Face bow, Jaw relation record, Virtual face scan, Accuracy, Precision, and Trueness. The Medical Subject Headings (MeSH) database was applied, with appropriate Boolean operators to refine the search strategy as presented in Table 1. Additionally, a manual review of the reference lists from the included studies was carried out to identify any relevant literature. A comprehensive search was completed in December 2025. A research flowchart is presented in Fig. 1.

Table 1.

MeSH terms used in electronic search

Database	MeSH terms
PubMed	((“dental articulator” OR articulator OR “face bow” OR Articulation OR “Jaw Relation Record”) AND (Virtual OR digital OR “Face scan” OR “Facescan” OR “Facial scanning” OR “digital dentistry”) AND (Accuracy OR Precision OR Trueness OR Verificat*)) NOT (review)
Scopus	TITLE-ABS-KEY ((“dental articulator” OR articulator OR “face bow” OR articulation OR “Jaw Relation Record” ) AND ( virtual OR digital OR “Face scan” OR “Facescan” OR “Facial scanning” OR “digital dentistry” ) AND ( accuracy OR precision OR trueness OR verificat* ) ) AND PUBYEAR > 2013 AND PUBYEAR < 2025
Web of Science	(“dental articulator” OR articulator OR “face bow” OR Articulation OR “Jaw Relation Record”) AND (Virtual OR digital OR “Face scan” OR “Facescan” OR “Facial scanning” OR “digital dentistry”) AND (Accuracy OR Precision OR Trueness OR Verificat*)

Fig. 1.

PRISMA study flowchart

Eligibility criteria

The inclusion criteria were defined as in vitro studies, clinical trials, case reports, clinical reports, and clinical techniques assessing the accuracy of virtual articulation in fixed prosthodontics of dentate patients with comparison to conventional facebow transfer or accepted articulator mounting references. Exclusion criteria included articles not published in English, studies involving orthognathic surgeries, studies on edentulous patients, and publications prior to 2014.

Review process

The identified articles were independently screened by two reviewers (PE, HA) based on predefined eligibility criteria. Initially, titles were assessed to identify potentially relevant studies, followed by abstract screening to exclude those that did not meet the criteria. Full-text articles of the remaining studies were then reviewed, and final inclusion decisions were made according to the established inclusion and exclusion criteria. Any disagreements during the selection process were resolved through discussion, with the involvement of a third reviewer (AK) when consensus could not be reached.

Data collection process

Data extraction forms were used to gather information, including the authors’ names, publication year, type of study, aim, sample size, study grouping, scan method/type, articulator type, anatomical references selected, alignment system utilized, mounting process, CAD software program used in alignment, results, and conclusion.

Risk of bias / quality assessment

The methodological quality of the included studies was assessed according to the study design using appropriate critical appraisal criteria. Overall, included studies demonstrated moderate methodological quality; however, limitations included small sample sizes, lack of standardized reference protocols, operator-dependent methodology, and heterogeneity in reported outcome measures.

Synthesis of results

This review offers a descriptive analysis of the included studies, considering all methodological aspects, including the materials used and the assessment methods employed.

Results

The total number of articles retrieved, irrespective of the type of article, was 1,916. After removing duplicates, the number of articles was narrowed down to 1,637. Of that, 179 full-text articles were assessed for eligibility; 162 were excluded because of an irrelevant population, a review design, a lack of quantitative outcome assessment, or outcomes unrelated to virtual facebow transfer accuracy. Of the 17 potentially eligible studies, 7 were excluded after full-text assessment, leaving 10 included in the qualitative synthesis, as presented in Fig. 2. Of the 10 included studies, 4 were in vitro investigations performed under standardized laboratory conditions, 2 were comparative in vivo clinical studies, and 4 were clinical technique reports describing digital workflows and transfer protocols. This distribution highlights that most currently available evidence is derived from controlled laboratory or technical workflow investigations rather than robust comparative clinical studies. Among the included clinical studies, only a limited number directly compared conventional facebow transfer with digital mounting methods under comparable conditions, while several other reports primarily described digital transfer workflows without a controlled conventional comparator. The outcome of the search was described descriptively, with all relevant methodological information, and is presented in Table 2.

Fig. 2.

Included studies based on technology utilized for virtual facebows

Table 2.

Summary of articles included in the systematic review on virtual facebow techniques

Author & Year	Type of Study	Aim	Sample	Study Group	Scan Method/Type	Articulator Type	Anatomical References	Alignment System	Mounting Process	CAD Software Program to Alignment	Conclusion
Kalman, 2016 [28]	Clinical technique	Develop a digital alternative and arbitrary mounting to address the shortcomings of the traditional facebow and provide additional clinical information.	2 separate sets of models poured in Microstone by each clinician technique	Clinician A Timed analog facebow (AF)	Conventional: CounterFit impression material & Quick Bite	N/A	Analog facebow & wax bite registration	Analog facebow & wax bite	Number of occluding teeth checked with 8 μm shim stock.	Mobile technology- application	The efficiency, accuracy, cost, and comfort of the VF provide the clinician with an alternative tool for records and essential technician communication.
				Clinician B Virtual facebow (VF)		Semi-adjustable articulator Whip Mix	Virtual facebow application & record	Face and dentition capture with reference grid and gyroscope
Lam, 2018 [5]	Clinical technique	Present a digital workflow for registering the patient’s horizontal plane in natural head position (NHP) on a virtual articulator.	1 patient	5 attempts of scans were done for the same patient	Stereophotogrammetry (SP) device (3dMDface, 3dMD Inc) used to scan the face. Teeth scanned with intraoral scanner (True definition; 3 M). Another SP scan with virtual facebow in position.	Denar Mark II articulator, Whip Mix	Buccal relationship of maxillary teeth & virtual facebow, 3D facial photograph, full-face cone beam computed tomography (CBCT).	Orientation reference board calibrates a stereophotogrammetry device and a 3D facial photograph of patient in NHP.	Load virtual patient model into Exocad, match THA to condylar element of the virtual articulator.	Exocad; GmBH	Casts mounted with NHP techniques have higher position in the articulator than those mounted with anatomic facebow technique.
Maveli, 2018 [29]	In vitro	Evaluate the coronal and sagittal orientation of the occlusal plane on maxillary casts mounted with an electronic application facebow system onto a semiadjustable arcon articulator.	Sample size: 30, 2 groups	Control group: 15 impressions of maxillary typodont	Polyvinyl siloxane impression material used to take impression of maxillary typodont attached to phantom mannequin head that has anterior landmarks (infraorbital rim & foramina).	Semi- Adjustable arcon Whip Mix Model 3040 Articulator	Frankfort horizontal plane and occlusal plane (OP)	Tablet calibration tool used, casts positioned on the stand with central incisors to a specific line. After alignment, another photograph was made and overlaid with the photograph of the phantom head.	Overlay of photograph of the phantom head aligned to central incisors	MaxAlign LabStand; Whip Mix Corp	MaxAlign system transferred the OP orientation more accurately for the sagittal plane than the coronal plane. Significant variation in coronal OP between test and control group that could have significant esthetic and functional implications. No significant differences were found in the sagittal orientation of the OP.
				Test group: 15 records using MaxAlign system (Whip Mix Corp) digital application on tablet	Tablet was placed on a stand and calibrated, 23 cm away and parallel to phantom head, photograph was taken from the application.
Amezua, 2022 [11]	In vitro	Analyze the influence of the facial scanning method on the accuracy with which a maxillary digital scan is transferred to a 3D face scan in a virtual facebow technique.	2 groups, 21 pairs of maxillary digital scans	Structured white light (SWL) technology (10 scans).	Maxillary digital scan (industrial reference scanner), and 3D face scan	N/A	Maxillary digital scan and standard virtual patient	Virtual facebow transfer (VFT) by intraoral transfer element (IOTE) kept in during face scans, removed, scanned with the impression material inside it, and another face scan without IOTE placed.	N/A	Geomagic Studio 2013, closest point algorithm/iterative closest point (ICP)	The accuracy with which a maxillary scan is transferred to a 3D face scan in a virtual facebow technique is strongly influenced by the facial scanner. Inaccuracies have been introduced when transferring a maxillary digital scan to a 3D face scan. Significant differences among trueness values when using different facial scanning methods with a very large effect size. Trueness of 0.138 mm and precision of 0.022 mm obtained by using SWL compared to trueness of 0.416 mm and precision of 0.095 mm was acquired from SFM scanning method.
				Structure-from-motion (SFM) technology (10 scans). 27 digital photographs of the mannequin’s face at different positions and angles. Photos uploaded to digital image-processing & 3D data-generation software program (Agisoft).
Li, 2022 [30]	Clinical technique	To investigate the trueness and precision of virtual facebow records using a smartphone as a 3D face scanner.	2 subjects	10 repeated virtual facebow records for each subject	Intraoral scan of maxillary arch (TRIOS), 3D facial scan with smartphone application (Hege 3D scanner).	Not mentioned	3 points: mesh spheres with diameter of 0.5 mm added to mesial buccal cusps of the left #14, right #3, and the mesial incisor angle of #9.	Facebow fork placed, and scan taken with it	N/A	Blender, version 2.8 and DentalCAD software (Exocad 2.2)	Virtual records made using smart-phone scans can capture maxilla position with high trueness and precision. Deviation anticipated to be around 1 mm in linear distance and 1º in angulation.
Inoue, 2024 [4]	Pilot clinical study	To determine differences in measurements among casts mounted virtually by using the average values of the Bonwill triangle and the Balkwill angle, casts mounted by using facebow records, and casts mounted from 3D facial scans.	10 participants (3 men, 7 women) with complete maxillary and mandibular dentitions	Facebow preservation group (FPG)	Intraoral scans, 3D printed models, face bow record (Artex Rotofix; Amann Girrbach AG), desktop scanner (Freedom HD; Degree of Freedom Inc) used to preserve the mounting	Semi-adjustable arcon articulator- (Artex CR; Amann Girrbach AG)	Right & left external auditory meatus, point at 42 mm inferior to nasal locator of the facebow	Facebow (Artex Rotofix; Amann Girrbach)	All mounted with same maxillomandibular relationship	ExodCAD; exocad GmbH, distances were measured from mesiobuccal tip of mandibular first molar to center of left condyle with incisal pin at 0 mm, 3 mm, and 5 mm.	1.Accuracies of FPG & FSG were comparable without statistical differences 2.AMG produced lower accuracy than FPG & FSG; however, increase in vertical dimension up to 5 mm was not significantly different 3.Use of 3D facial scanner to virtually mount dental casts provides a suitable option for the restorative dentist -Both anterior and posterior condylar measurements between the FPG & AMG were found to be sig different, while comparisons between the FPG & FSG were found not sig different. -All changes in vertical dimension were found not to be sig different with respect to both anterior & posterior measurements.
				Average mounting group (AMG)	Intraoral digital scans mounted virtually with Bonwill triangle and Balkwill angle concepts		Incisal edge of mandibular central incisor (anterior reference) and mesiobuccal cusp tips of mandibular first molars (posterior reference)	Average values of Bonwill triangle & Balkwill angle
				Facial scan group (FSG)	Casts digitized and superimposed on facial scans using a target system		Bergstrom point, glabella, and orbitale	Target facebow fork system with a facial scanner used to superimpose facial scans with STL files
Revilla-León, 2024 [26] (a)	In vitro	Compare the accuracy of the maxillary cast transfer into the virtual semi-adjustable articulator by using analog and digital methods.	3 groups n=10	Conventional facebow (CNV)	Reference scan; Industrial scanner (Gom ATOS Q3D 12 M). Study group: Max typodont; T170;Medit), Digitized facebow record	Panadent PCH Articulator, exocad GmbH	Reference glasses 3.0; Kois Center +Kois dentofacial analyzer system + LazrTrak	Reference platform (Kois adjustable platform) (DentalCAD)	Digitized facebow superimposed with the virtual reference Kois table (Kois Platform; exocad GmbH)	Dental CAD Software, v3.1 Rijeka, exocad GmbH	P group demonstrated best trueness & precision values. The maxillary cast transfer accuracy ranged from 137 ±44 μm to 453 ±176 μm among techniques. Midline evaluation a no trueness & precision. Significant posterior right anterior, posterior left, and overall trueness discrepancies among the groups. P group obtained best posterior R & L and overall precision and trueness. P & FS groups demonstrated best anterior trueness but no anterior precision discrepancies were found.
				Digital photograph (P group)	10 digital photographs (jpeg) with a DSLR D780, Marcolens 105 mm, and maxillary scan		Photographs with levelled Kois reference glasses	Rescaled photograph, maxillary scan positioned on Kois reference table & aligned with photo
				Facial scanning with a 3D extraoral scan body (FS)	Facial scan (Instarisa 3.0 Facial Scanner) with extraoral scan body in place on mannequin; exported as PLY, without scan body, mannequin scanned with laboratory scanner (Gom ATOS Q3D 12 M),		Facial scan with a 3D extraoral scan body with grid	Digitized extraoral scan body with Kois reference table
Revilla-León, 2024 (b) [31]	In vitro	Compare the accuracy of maxillary cast transfer into the virtual semi-adjustable articulator by using an analog and digital standardized photography technique	2 groups	Conventional facebow (CNV)	Laboratory scanner (T710; Medit) scan of maxillary dentate cast with a magnet, mannequin positioned in a tripod, reference scan of the assembled mannequin obtained with facial scanner	Panadent PCH Articulator; exocad GmbH	CNV group: Kois disposable tray in the facebow of the dentofacial analyzer system (Kois, Panadent), facebow acquired, digitized with same laboratory scanner	Digitized facebow superimposed with virtual reference Kois table of semi-adjustable by using disposable index as common information	N/A	DentalCAD, v.3.1 Rijeka; Exocad GmbH	Facebow record methods impacted trueness and precision of maxillary cast transfer into virtual semi-adjustable articulator. Photo group obtained better trueness in midline transfer compared to CNV. CNV obtained better trueness in anterior, posterior right & left, overall discrepancy compared to P group. P group obtained better precision in midline, posterior right & left, anterior and overall discrepancy of maxillary cast transfer compared to CNV. Overall discrepancy measured in CNV group was 0.620 ± 0.396 mm, in P group was 1.282 ± 0.118 mm. Sig trueness differences found in midline, anterior, posterior right, posterior left, and overall discrepancy between CNV & P groups.
							P group: 10 digital photos with levelled Kois reference glasses	Photographs rescaled, aligned with maxillary scan
Villanueva et al. 2025 [27]	Clinical dental technique	describes a digital technique for virtual mounting in a semi-adjustable articulator by integrating CBCT and intraoral and facial scans	Single patient	intraoral scans of the maxillary and mandibular arches + CBCT + facial scan	maxillary intraoral scan (reference scan) CBCT Facial scan		1. Mesiobuccal cusps of the bilateral first molars and the distoincisal angles of the maxillary lateral incisors as craniofacial reference points to establish a stable 4-point registration (align CBCT to intraoral scan) 2. Reference points on the cusp tips of both maxillary canines and the midpoint of the central incisors (align face scan to intraoral scan)	best-fit algorithm to refine the superimposition coincides at the defined reference points and along the occlusal plane ensuring symmetry with the facial midline and interpupillary line	aligning the Frankfort horizontal plane with the articulator reference plane using digital semi-adjustable articulator (BioArt A7 Plus)	Exocad; GmBH	By eliminating analog transfers, it improves accuracy, reduces variability, and enhances communication among specialists
Kim, 2025 [32]	In vivo	Compare accuracy of a novel virtual facebow transfer (VM) technique based on CBCT with that of conventional mounting (CM) technique using a facebow.	5 repeated mountings for with each technique for 15 participants	Conventional mounting technique	CBCT, preliminary impressions à Casts, one cast was digitalizedà STL	Hanau Modular Articulator System, WhipMix Corp scanned with 5 cube shaped markers with an industrial scanner as the reference articulator (RA-STL).	Anatomical Frankfort horizontal plane, infraorbital notch.	For CM group (Hanau Springbow; WhipMix Corp)	Mounted conventionally with facebow, then with cast on articulator, industrial scanner scan it all; TEMPLATE	Geomagic Control X v2018.0.1; 3D Systems	Significant differences found between 2 groups, VM more precise technique. Position of maxillary right central incisor indicated VM casts were positioned ahead of CM casts. Occlusal plane angle tended to be steeper in CM casts than VM casts. Mean distance between target points was significantly greater in the CM group than VM. Standard deviation between target points was sig greater in CM group than VM.
				Virtual facebow transfer technique			References from the skull: Bilateral porions, condyle, external acoustic meatuses, orbit, midline, sagittal suture, & spine	For VM group: CBCT, STL files of their skulls were obtained	Superimposing the scans with the CBCT.

Several methods were used for transferring maxillary model to a virtual articulator, such as digital photographs, face scan alone or combined with intraoral or extraoral scan body, smartphone application, digital photographs using Kois glasses, CBCT, and stereophotogrammetry. Five studies used face scans, with one study [11] that used two different technologies: structured white light and structured motion technology that resulted in trueness of 138 μm and 416 μm respectively. Whereas another study [26] found the facial scanner to have trueness of 377 ± 106 μm. However, Inoue [4] found no significant differences between FPG (conventional) & FSG (facial scan), whereas the technique described by Villanueva [27] demonstrated that digital mounting procedures provide comparable accuracy with improved reproducibility and communication.

Three studies used mobile/smartphone device/application for aligning maxillarycast [28–30]. Two studies used mobile applications, namely Virtual Facebow by Kalman [28] and MaxAlign System by Maveli [29]. The former found that the application successfully replicated nearly twice the original occlusion as compared to the analog facebow, whereas the latter found that the mean difference from the control was 0.275 degrees in the sagittal plane and 1.27 degrees in the coronal plane, respectively. The remaining study by Li [30] used a smartphone as a 3D face scan and found trueness to be 1140 ± 400 μm and precision to be 1080 ± 520 μm.

Two studies done by the same author group used photography for virtual mounting and found accuracy ranging from 137 ± 44 μm in one study to 1282 ± 118 μm in the other study as compared to 453 ± 176 μm and 620 ± 396 μm for control groups, respectively [26, 31]. One study used CBCT and found nearly double the discrepancy in conventional mounting as compared to the virtual mounting group [32]. Lastly, one study used stereophotogrammetry and found it to be reliable for recording natural head position with deviations of less than 1 mm/1 degree in 5 attempts [5].

Collectively, these studies suggest that although the accuracy of transferring the maxillary cast to the articulator varies depending on the facebow recording technique, most methods produce results that are clinically acceptable. Notably, digital photography demonstrated particularly promising outcomes. Facial scanning also showed high levels of accuracy, though its effectiveness is influenced by the specific scanning method used.

Discussion

The purpose of this review was to summarize the available methods for transferring a maxillary digital model to a virtual articulator and to evaluate their accuracy in comparison with conventional facebow techniques based on the existing evidence. The present findings should be interpreted in the context of recent systematic reviews evaluating virtual articulation systems. Contemporary evidence suggests that virtual articulators generally demonstrate clinically acceptable static articulation accuracy and promising reproducibility, whereas dynamic occlusion simulation remains more variable due to differences in software algorithms, condylar motion simulation, and registration methods. Recent systematic reviews also emphasize that heterogeneity in reference standards, measurement techniques, and methodological design limits direct comparison across studies. Consistent with these observations, the present review found that virtual facebow transfer demonstrates encouraging repeatability and clinically acceptable transfer accuracy in selected workflows; however, robust comparative clinical evidence remains scarce, and much of the available literature is derived from controlled in vitro settings rather than true clinical conditions [4, 23, 24]. The methods identified included digital photography, facial scanning alone or combined with intraoral or extraoral scan bodies, smartphone-based applications, digital photography using Kois glasses, cone beam computed tomography (CBCT), and stereophotogrammetry. Owing to variability in previously published findings on digital mounting techniques and the absence of consensus on their comparative transfer accuracy, the null hypothesis of this review was that virtual facebow transfer techniques do not significantly differ from conventional facebow transfer methods in mounting accuracy.

With respect to photographic techniques, Revilla-León [31], found that the best trueness and accuracy values were found in the digital photography method test compared to the conventional facebow record. In another study by Revilla-León [26], the digital photo group outperformed the conventional group in terms of midline transfer trueness. However, for posterior right and left regions, anterior reference points, and overall maxillary cast discrepancy, the conventional facebow method demonstrated greater accuracy than photography-based transfer. These findings suggest that although digital photography may offer advantages in specific reference areas, its performance may be region-dependent.

Additional elements should be considered while using the photography method that could impact the distortion of the captured photos, including the camera’s focal length and lens angle. Moreover, this technique is highly operator dependent, and inaccuracies during image acquisition may contribute to increased error when extraoral data are captured using photographs [33].

Regarding digital facebow systems, Maveli [29] reported that the MaxAlign facebow system transferred occlusal plane orientation more accurately in the sagittal plane than in the coronal plane. Similarly, Kalman [28] demonstrated acceptable effectiveness and precision in using the virtual facebow application. Advances in digital technology also allow virtual articulators to be programmed using either patient-specific sagittal condylar inclination values or average values, which may further influence clinical accuracy.

In evaluating facial scanning methods, Inoue’s [4] clinical study, no statistical differences were noted between the face bow methods tested and 3D face scan accuracy levels. In addition, Amezua [11] showed that the facial scanning approach had a significant impact on how accurately a maxillary scan can be aligned with a 3D face scan in a virtual facebow approach. Li [30] reported that the expected deviation was approximately 1º in angulation and 1 mm in linear distance by using a smartphone application used in developing 3D face scan to record maxilla location whereas, Villanueva et al. [27] showed that digital mounting procedures provide comparable accuracy with enhanced reproducibility and communication, further advanced by integrating CBCT, intraoral, and facial scans into a single workflow.

Most facial scanners had reported accuracy of about 500 μm, which is within clinically acceptable limits. However, several variables, such as the type of scanner utilized, the geometry of the object being scanned, and the depth and speed of scanning, might affect the accuracy of the scanner [4]. Concave surfaces, like the lower face, are less accurate than the flat and convex surfaces of the upper face [34]. Areas with undercuts, including the labial sulcus, oral commissures, and oral fissure, are particularly challenging to capture accurately. The midfacial region has been reported to exhibit the highest precision and overall accuracy, whereas landmarks such as teeth, the external auditory canal, and the nostrils present additional challenges [35]. In a clinical study by Kim et al. [32], CBCT-based virtual mounting demonstrated an acceptable accuracy compared with conventional mounting methods. However, the mandibular condyle, external bone auditory meatus, and infraorbital bone boundary must all be visible for CBCT-based mounting to be feasible. Given the principle of keeping radiation exposure “as low as reasonably achievable” (ALARA), CBCT-based mounting may be justified for complex prosthetic rehabilitations, implant dentistry, or orthognathic surgical cases in which CBCT imaging is already indicated, but it may not be appropriate for routine or simple prosthodontic treatments [14, 36]. When examining the accuracy of stereophotogrammetry, Lam [5] demonstrated that stereophotogrammetry-based recording of natural head position showed deviations of less than 1 mm or 1 degree, indicating good repeatability. Nevertheless, this technique requires specialized software and is sensitive to lighting conditions, as excessive or insufficient ambient light can compromise image accuracy [37].

Interpretation of trueness in virtual facebow transfer remains complex because spatial accuracy is influenced by how each system defines craniofacial orientation, reference planes, and landmark registration. Variability in the use of the Frankfort horizontal plane, axis-orbital plane, natural head position, facial midline, or other craniofacial landmarks may produce systematic differences in occlusal plane orientation while still resulting in clinically acceptable mounting. This may also explain why several studies reported greater precision in midline transfer measurements, as midline landmarks are generally less ambiguous and more reproducible than broader frontal facial plane references.

A major limitation in interpreting the current review is that a substantial portion of the literature is based on in vitro investigations performed under highly standardized conditions using rigid mannequin or typodont models. Unlike laboratory models, the human face is a dynamic and deformable structure influenced by soft tissue mobility, facial expression, muscle contraction, head posture, and operator-dependent placement of external reference devices. Consequently, the reproducibility observed in laboratory conditions may overestimate the clinical repeatability and transfer accuracy achievable in vivo. Additionally, direct comparison between digital and conventional facebow transfer should also recognize that conventional articulation itself exhibits inherent variability related to operator technique, arbitrary hinge-axis approximation, landmark identification, and patient positioning. Accordingly, digital methods should be interpreted in relation to the reproducibility and accepted clinical tolerance of conventional methods rather than against a fixed absolute reference. Furthermore, the limited sample size reported in the available studies restricts the generalizability of these findings.

Conclusion

Current evidence suggests that virtual facebow transfer techniques demonstrate promising repeatability and clinically acceptable transfer accuracy in selected workflows. However, robust comparative clinical studies directly evaluating trueness and precision relative to conventional facebow transfer remain limited. Furthermore, much of the available evidence is derived from in vitro investigations performed under controlled laboratory conditions that may not fully reproduce important clinical confounding factors, including soft tissue mobility, facial expression, muscle activity, and operator-dependent variability in landmark positioning. Therefore, while virtual facebow systems show significant clinical promise, further standardized prospective clinical investigations are needed before definitive conclusions regarding superiority over conventional methods can be established.

Abbreviations

VFT

Virtual Facebow Transfer

CBCT

Cone-Beam Computed Tomography

3D

Three Dimensions

SP

Stereophotogrammetry

AF

Analog Facebow

NHP

Natural Head Position

OP

Occlusal Plane

FS

Face Scan

SWT

Structured White Light

Authors’ contributions

PE have made substantial contributions to the conception and design of the work; PE, HM, AK did the analysis, and interpretation of data; PE, HM, AK, KK, RSZ have drafted the work and substantively revised it. All authors approved the final version before submission.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

All data are available in the manuscript.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.