Three-dimensional analysis of posed smile in adults: A scoping review

This scoping review investigated the evidence on the three-dimensional analysis of a posed smile in adults to discover any research gaps in this research area. Electronic searches of articles written in English were performed using the four databases of Embase, PubMed, Springer, and Web of Science with publications from 2010 to 2023. Reference lists were also manually searched to identify additional studies. The results showed that 13 cross-sectional descriptive studies from Asia, Europe, North and South America met our inclusion criteria. Studies mainly focused on linear and angle measurement for resting and smiling faces and landmark movement from resting to smiling. Most studies conducted analysis of smiles stratified by sex, ethnicity, smile type, dental occlusion, skeletal pattern, and age. Two studies compared smiling with the resting position and one study compared the attractive smiling group with the ordinary group. Our scoping review revealed the insufficiency of some measurement methods, such as those employing area, volume, and soft tissue thickness. Furthermore, few studies were conducted in Asian populations, and comparisons of various smile types, overjet types, horizontal skeletal patterns, and comparisons of smiles between people with untreated normal occlusion and those who had been orthodontically treated were lacking.


Introduction
Facial attractiveness in modern society assists in enhancing self-esteem and promoting relationships, employment opportunities, and business success. 1The mouth area, particularly smile characteristics, has been analyzed in terms of facial attractiveness.2e4 At present, smile esthetics tend to be people's principal concern when they pursue dental treatment. 5ubin was the first to examine the anatomical analysis of normal smile, and he divided it into three types: the Mona Lisa, the canine, and the full denture smile.He pointed out that changes in the anatomy of the muscle, soft tissue, and bones can affect the smile. 6Smile can also broadly be divided based on neurological control into "unposed" smile, which is involuntary related to emotion induced by enjoyment, and "posed" smile, which is voluntary and not related to emotion.It might be a formal welcome, a way to placate someone, or an effort to project confidence. 7,8ecause posed smile is repeatable over time, it has been mainly studied by orthodontists. 4,9,10For the treatment of patients with facial paralysis as well as for the diagnosis and evaluation of orthodontic treatment, a precise comprehension of the normal posed smile and precise quantitative measurements are required. 8,9uantitative smile analysis has been conducted.Hulsey 10 quantified the relationship of lips and teeth displayed when smiling and compared the smiles of people with untreated normal occlusion and those who had been orthodontically treated.He calculated the smile line ratio, smile symmetry ratio, and buccal corridor ratios and measured upper lip height curvature.11e15 Many methods have been used to evaluate smile esthetics, such as photographs, radiographs, model scanning, three-dimensional (3D) surface imaging, video, and clinical assessment. 16Two-dimensional (2D) approaches cannot precisely evaluate the complicated 3D soft tissue of the face. 17Studies reported that 3D orofacial imaging techniques can be used in daily dental practice, with digital animated models providing possibilities for quantitative diagnosis in three dimensions and the evaluation of treatment outcomes. 18ost smile analysis research, including the aforementioned studies, has employed 2D techniques.Few studies applied 3D techniques, and this scoping review was thus conducted to investigate the 3D of a posed smile in adults as well as to discover any research gaps to benefit people interested in researching smiles using 3D techniques that are increasing.

Search strategy
A literature review was performed following the checklist and definitions of the PRISMA extension for scoping review. 19Electronic searches of articles written in English were performed using the four databases of Embase, PubMed, Springer, and Web of Science with publications from 2010 to 2023.Reference lists were also manually searched to identify additional studies.The keywords used for the search were as follows: ((smile[MeSH Terms]) AND ((3D) OR (three dimensional) OR (Three dimensions) OR (3 dimensions) OR (stereophotogrammetry) OR (photogrammetry) OR (facial scanner) OR (facial scan) OR (structured light)).

Study selection as well as inclusion and exclusion criteria
The inclusion criteria were as follows: studies must have (1)  reported measurements and conducted analyses of posed normal smile; (2) involved a group of adult participants with no obvious skeletal discrepancy, congenital defect, maxillofacial trauma, facial paralysis or history of orthodontic treatment; and (3) used a 3D camera.The exclusion criteria were as follows: studies that were (1) literature reviews; (2) scoping reviews; (3) systematic reviews and metaanalyses; or (4) case reports.
Two impartial reviewers independently extracted the data, and any discrepancies were settled with the help of a third reviewer.The information that was taken from the study's data included the study's participants, measurements, comparisons, results, and conclusions.

Selection of sources of evidence
A total of 2894 records were collected through the electronic search.After duplicates were removed, 2331 records were identified.According to the PRISMA statement, we screened the abstracts and assessed the available full-text articles based on the eligibility criteria.Including one record retrieved in the manual search, 13 articles were examined in this scoping review.The process of the selection of sources of evidence is detailed in the PRISMA flowchart (Fig. 1).

Characteristics of sources of evidence
The characteristics of the included studies are described in Table 1.All studies had a cross-sectional descriptive design.Four studies were conducted in Europe, three in North America (the United States), three in Asia, and one each in South America and Turkey.The participants have an average age ranging from 18.62 to 34 years or between 15 and 60 years, except for participants age over 60 years in two studies. 20,21nthesis of results regarding type of measurement (dependent variable) Linear and angle measurement of resting and smiling faces Five studies reported only linear measurement of resting and smiling faces, 20e24 and two studies reported both linear P. Banditsaowapak, J.H.-C.Cheng, D.D.-S.Chen et al. and angle measurements. 25,26Maxilla height, lower face height, mandible height, nasal bridge length, width of the mouth (intercommissural width), upper lip length, upper lip vermillion length, lower lip length, and lower lip vermillion length comprise most of the linear dimensions measured at resting and smiling positions.However, the interlabial gap, gingival display, maxillary incisor display, and maxillary intercanine width, are also measured at the smiling position.The angle measurements used in these studies are the nasolabial angle, upper lip angle, and lower lip angle.

Curving line
Only one study by Tanikawa et al. 21reported the curving lines.They extracted 142 measurements to describe five categories of curving lines: inter-landmark contours, sagittal sections, axial sections, facial outlines, and supraorbital ridge outlines.Their curving lines in five categories are composed of many cross-sectional lines and angles of the 3D image surface projected across three planes.

Landmark distance ratio
Four studies involved proportional measurements using only smile records. 22,24,25,27Li et al. 22 and Toth et al. 24 examined the smile index (intercommissural width/interlablal gap), whereas Demir and Baysal 25 analyzed the smile index and buccal corridor ratio.Ceinos et al. 27  One study involved ratios during smiling and resting postures. 21Those ratios related to smiling include total midface/lower face height, midface/lower face height, lower face/face height, mandible/upper face height, mandible/lower face height, chin/lower face height, facial index, upper face index, face height/mandible width, mandibular index, and mouth/face.

Landmark position
There are two studies reporting the landmark position.Masoud et al. 23 showed the positions of the left and right alar curvature, subspinale, subnasale, labrale superius, sublabiale, labrale inferius, and pogonion relative to the coronal and axial planes.The other, Li et al., 28 investigated the maxillary central incisor position relative to the coronal plane.

Landmark displacement from resting to smiling
Five articles reported the movement of oral landmarks related to the x-plane, y-plane, and z-plane from the resting position to the posed smiling position. 16,22,24,25,29he key landmarks used in these articles are the labiale superius, labiale inferius, and cheilion (right and left), and other landmarks including the subalare (right and left), subnasale, nasolabial fold (right and left), crista philtre (right and left), upper mid-lateral lip (right and left), and lower mid-lateral lip (right and left).Demir and Basal, 25 Li et al., 22 and Toth et al. 24 divided the average displacement of each landmark into x, y, and z vectors, whereas Campbell et al. 16 calculated the distance movements averaged over landmarks across three planes.In addition to the two    -Surface comparison analysis: The mouth corners of the older group at rest and smile were narrower along the x-axis.The older group had lower-positioned cheeks and chin vertical positions during rest and smile, according to the yaxis.When the older group was at repose, their mouth corners and lower lips were situated in more inferior positions, and when they smiled, their upper lips were situated in more inferior positions.
Along the z-axis, when an older group was at rest and smiling, the nasal bridge and infraorbital areas were more protuberant along the zaxis.At rest, the older group had more retruded mouth corners.
-Inter-landmarks distances and ratios at rest in the order group: sagging skin in the cheeks of the zygomatic region and facial shape, protrusion of the nasal wing, wider lower face, mouth placed lower on the face, and flabby cheeks at the level of the lips and chin.When smiling, the order group features the following: a lowerpositioned mouth corner, a convex subnasal profile, sagging skin at the mandibular angle, and a larger mouth protrusion.
-Landmark displacement analysis: In the older group, 66 variables shown significant shifts from resting to smiling.Significant changes in the younger group were seen in 144 variables.
-Discriminant analysis for resting vs. smiling: From rest to smile, the   29 further calculated the angle and distance of movement of landmarks across 2D and 3D planes.

Surface comparison method based on 3D meshes and closest point-to-point distances
Three studies used comparison methods based on 3D meshes and the closet point-to-point distances to analyze smiles. 21,30,31With this approach, 3D-averaged faces (surface shell meshes) of rest and smile postures are created and superimposed, and then the root mean square (RMS) differences of the point-to-point distances are obtained.In addition to the RMS, the 75th percentile is computed in the study of Jandova ´and Urbanova ´,30 whereas Pucciarelli et al. 31 calculated labial surface areas and percentage modification in the resting and smiling positions.On the other hand, Tanikawa et al. 21created and superimposed the 3D-averaged faces of younger and older groups, then the vectors with x-, y-, and z-values from average mesh points of the younger group to those of the older group were calculated.

Synthesis of the results of subgroup analysis (independent variable) Sex
Seven articles compared smile measurements between men and women. 20,23,25,27,29,30Among them, four articles focusing on smile movement reported no statistically significant difference between sexes. 31However, the remaining two articles focusing on linear and angle measurement in the resting and smiling positions revealed that most of the measurements, including upper lip length, lower lip length, and width of mouth, are significantly larger for men than for women. 20,23hnicity Only one study compared smile dimensions between two ethnicities, namely African American and Caucasian. 20The researchers concluded that, with the exception of the length of the male upper lip, all measures are significantly greater in African Americans than in Caucasians.

Smile type
One study analyzed three types of smiles, including the "Mona Lisa smile," canine smile, and full-denture smile. 31MS values, which are computed after the superimposition of all smile models in the resting position, gradually increase as smiles transition from Mona Lisa smiles to fulldenture smiles.Statistically significant differences are noted for facial and labial models among the three smile types.

Dental occlusion and skeletal pattern
Four articles investigated dental occlusion and skeletal patterns. 16,22,24,25Demir and Basal, 25 Li et al. 22 and Toth et al. 24 evaluated smile characteristics in terms of different vertical skeletal patterns.Demir and Basal 25 and Li et al. 22 reported that the interlabial gap and the movement of upper lip landmarks have a tendency to become larger in people with higher vertical skeletal patterns.The smile index exhibits the opposite results.Toth et al. 24 found that there are moderate and weak correlations between vertical skeletal variables and soft tissue smile variables.According to the moderate correlations, as SN-GoGn and anterior facial height increases so does the interlabial gap while the smile index reduces.In addition, this study reported that along the y-and z-axis, the intercommissural width, smile index, and lower lip all have significant multiple regression models.Campbell et al. 16 compared the range of movement in the smiling position between people with normal overjet and those with increased overjet, and they determined that the average movement of the landmarks in the normal overjet group is greater than in the increased overjet group.

Age
Two studies 20,21 21 the mouth corners of the older group at rest and smile are narrower along the x-axis.The older group has lower-positioned cheeks and chin vertical positions during rest and smile, according to the y-axis.When the older group is at repose, their mouth corners and lower lips are situated in more inferior positions, and when they smile, their upper lips are situated in more inferior positions.Along the z-axis, when an older group is at rest and smiling, the nasal bridge and infraorbital areas are more protuberant along the z-axis.At rest, the older group has more retruded mouth corners.Moreover, they also concluded that the older group observes less soft tissue movement during the 3D facial changes in facial expression development and the transition from resting to smiling than the younger group.

Resting versus smiling position
Two studies analyzed the discrimination of resting versus smiling posture. 24,26Parra et al. compared the length of the alar base and the nasolabial angle between resting and smiling positions. 26The results revealed that the mean value of the nasolabial angle is unchanged, whereas the length of the alar base changes significantly in the smiling position.Tanikawa et al. 21analyzed the discrimination of resting versus smiling posture in the younger group and older group.They concluded that from rest to smile, the younger group has a large retrusive movement of the lip commissure and upper lip, a decrease in the labio-mental fold, a retrusive movement of the lower lip, a decrease in protrusion at the level of the lower lip and at the chin, and an increase in facial width.On the contrary, the older group has a significant decrease in lower lip height but increase in cheek protrusion, and retrusive movement of nasal ala.

Attractive score
One study by Li et al. 28 compared the position of maxillary central incisors between the attractive smiling female sample and the ordinary smiling female sample.The measurements they analyzed were the anteroposterior distances from the maxillary central incisors (mFA) to four vertical lines.The result revealed the average maxillary incisor position of the attractive group was more posteriorly to all vertical lines than that position of the ordinary group.
In addition, they concluded that the mFA to the vertical line passed through soft tissue subnasale (SSA) and the mFA to Glabella distances are esthetically important factors in female profiles.

Discussion
This scoping review identified twelve studies published between 2010 and 2023 that involved the 3D analysis of a posed smile in adults.The included studies analyzed normal smiles in adults through 3D imaging and the following types of measurement: linear and angle measurement in the resting and smiling position, curving line, landmark distance ratio during smiling, landmark position, landmark displacement from resting to smiling, and the surface comparison method.According to our review, most of all studies contribute the same opinion that these techniques are completely non-invasive and non-contact imaging systems that offer superior accuracy, reliability, and speed compared to 2D photography.
Regarding the type of measurement, our results indicated that most of the included studies focused on linear and angle measurement in the resting and smiling positions, the landmark distance ratio during smiling, and landmark displacement from resting to smiling.Only two studies reported the landmark position, and three studies used the surface comparison method based on 3D meshes and the closet point-to-point distances.Because the lip framework is a critical smile component, 32 the researchers that conducted linear and angular measurements between landmarks may not gain a comprehensive understanding of the morphology of soft tissues.Other techniques applied in soft tissue analysis have potential application in smile analysis.For example, Ayoub et al., 33 Sforza et al., 34 and Ferrario et al. 35 conducted surface area and volume measurement at rest position only.In addition to the method that superimposed the 3D dental image with 3D facial soft tissue in the study of Masoud et al., 23 we could measure the lip thickness, which is an influent variable on the profile with respect to the position of the incisors. 36We, therefore, suggest conducting more research measuring the surface area, volume, and soft tissue thickness to examine smile morphology to fill the research gaps we discussed above.
The included studies involved subgroup comparisons in terms of sex, ethnicity, smile type, dental occlusion and skeletal pattern, and age, resting versus smiling position, as well as attractive score.We can determine that those subgroups are independent variables of their studies.Most of the included studies compared smiles between sexes, and a few studies compared smiles across ethnicities, smile type, ages, and attractive score.Rubin 6 was the first to classify smiles into the three following types: Mona Lisa, canine, and full-denture smiles.Each type varies in terms of its direction and the strength of the individual perioral muscles.By contrast, Pucciarelli et al. 31 evaluated the labial movement in terms of smile type.Thus, no study has analyzed and compared different smile types using 3D imaging techniques.For the dental occlusion and skeletal pattern subgroup, Campbell et al. 16 was the only study that measured the magnitude of movement during smiling on the basis of different anteroposterior dimensions of occlusion between people with normal and increased overjet.No study has implemented another measurement for analysis by the type of overjet.In terms of age, Souccar et al. 20 reported that the upper lip length during smiling and the mouth width in the resting position increase as age increases, whereas gingival and maxillary incisal display decrease with age.This result is in line with that of Sachdeva et al. 37 in their 2D imaging study.Souccar et al. 20 found no significant difference in the upper lip length at rest among age groups; by contrast, in their 2D imaging study, Dindaro glu et al. 38 noted an increased upper lip length at rest with increased age.From our review, we suggest comparing smiles in three dimensions across various overjet types, overbite types, horizontal skeletal patterns, ethnicities, smile types, and ages for the future search or even get the norm of smile characteristics according to those variables.If we could identify the average 3D smile measurements of normal dentoskeletal participants, we could then compare them with those who were graded with a high esthetic score of smiles using a questionnaire.This comparison would involve various measurements different from the study conducted by Li et al., 28 which primarily focused on the anteroposterior position of the maxillary central incisor.Since there is no study comparing smiles on the basis of different anteroposterior dimensions of occlusion, we might perform measurements for analysis between people with normal and irregular overjet or anteroposterior skeletal relationship.For the vertical dimension, we recommend analyzing the type of overbite as well.Furthermore, based on the classical studies of Hulsey 10 and Ackerman et al. 8 which compared the features of a posed smile between orthodontically treated patients and participants who were untreated with normal occlusion in 2D images, we might consider conducting another research study in a similar manner, but utilizing 3D imaging techniques to obtain more explicit results.
In addition, one study 27 did not group the participants, but instead, they compared the mean facial proportions when smiling to the golden ratio (1.618).The study reported that out of the four observed ratios, three of them, which were horizontal ratios, were close to the golden ratio.However, this study only benefited from the 3D camera in terms of photo accuracy because their observed ratios were calculated from the 2D measurements projected onto three planes.Therefore, we could advance future research by comparing the other 3D measurements we previously suggested with the golden ratio.
In conclusion, we recommend conducting more research, including the following types of measurements: the surface area, volume, and soft tissue thickness, to examine smile morphology.Regarding the independent variables, we suggest studying the effects of various overjet types, overbite types, horizontal skeletal patterns, ethnicities, smile types, and ages on 3D smile characteristics.Moreover, 3D quantitative analysis of attractive smiles, orthodontically treated smiles, and other ratios associated with the golden ratio should be further investigated.Our scoping review has some limitations.We could only include publicly available articles written in English.Therefore, relevant studies conducted and published in local journals or studies published in other languages were not included in the analysis.
documented measurement of the following ratios: IE-Me/SN-IE (incisal edge of the maxillary central incisorelower border of the chin/subnasaleeincisal edge of the maxillary central incisor); IE-DCr/DCr-LCr (incisal edge of the maxillary central incisoredistal edge of the right canine/distal edge of the right canineeright commissure), and CLr-CLl/DCr-DCl (intercommissural width/distal edge of the right canineedistal edge of the left canine).
20alyzed smile dimensions across age groups.Souccar et al.20grouped the age of participant into five groups: 20 to 30, 30 to 40, 40 to 50, 50 to 60, and >60 years.The researchers reported that the upper lip length during smiling and the mouth width in the resting position increase significantly as age increases.By contrast, gingival display and the maxillary crown length during smiling decrease as age increases.According to the study byTanikawa et al., in the upper jaw: cephalometric evaluation.Int J Clin Pediatr Dent 2019;12:391e7.37. Sachdeva K, Singla A, Mahajan V, Jaj H, Negi A. Esthetic and smile characteristics at rest and during smiling.J Indian Orthod Soc 2012;46:17e25.38.Dindaro glu F, Do gan S, Ertan Erdinc ¸A.Smile esthetics: age related changes, and objective differences between social and spontaneous smiles.J Clin Pediatr Dent 2011;36: 99e106.P. Banditsaowapak, J.H.-C.Cheng, D.D.-S.Chen et al.