Introduction
Morphological analysis of teeth is critical for the diagnosis and treatment of patients with a wide variety of oral and craniofacial pathologies. There are multiple ways that have been described for measuring teeth [1, 2]. However, the methods that are pertinent to clinical management are generally focused on the clinical crown. Traditionally dental morphometrics have used linear metrics such as mesiodistal, buccolingual, and occluso-gingival dimensions [3]. Historically, tooth morphology has been studied using manual techniques which involve a variety of calipers or the Boley gauge; instruments that can only obtain linear measurements [4‐9]. These methods are limited to providing tooth size and are inherently incapable of detecting variations in tooth shape, form, and surface topography [10]. The establishment of more detailed methods has included identifying more landmarks on teeth, [11, 12] introducing angles within teeth, [13] and the use of occlusal polygons [14, 15]. A significant development was the combined use of high-definition photographs and computer technology [16]. Occlusal polygons were further incorporated into elliptical Fourier analysis (EFA) and used to analyze molar shapes [17]. Bernal (2007) used similar approaches but went further to subject the data to generalized Procrustes analysis (GPA) [10].
Geometric morphometric method (GMM) is increasing being applied to dental and craniofacial investigations [18]. Using aspects of GMM, Pavoni et al. evaluated the palatal morphology in children with impacted incisors, [19] Al Shaharani et al. evaluated the morphology of molars in patients with hypodontia [20] and Paoloni evaluated palatal morphology in Class II patients [21]. GMM has also been used to assess skeletal morphology in Class II and Class III and cleft lip and palate patients, [22‐25] as well as to evaluate dental arch morphology of different malocclusion groups [26].
Anzeige
Advances in digital imaging and scanning have facilitated the recording of landmarks as coordinates. Robinson et al. used this concept to study tooth form from a photographic image using two-dimensional (x, y) coordinate; [27, 28] thus introducing a novel application in the study of tooth morphology. Their methodology used two-dimensional (2-D) data, as opposed to 3-D, and therefore provided only partial description of shape [29]. GMM has been used to study arch form, [30] tooth surface reconstruction, [31] and dental anthropology [32, 33].
Three-dimensional imaging has found applications in orthodontics [34]. Archives of 3-D orthodontics study models produce images that are identical to the original study models, easing access, study and exchange of clinical data [35‐38].
There is no research investigating differences in tooth shape among the Hispanic population. The diagnosis and management of Tooth Size Discrepancy (TSD) for Hispanics has relied on standards that do not represent the group. The aim of this study was to determine whether there are significant differences in 3-D tooth shape between patients with Angle Class I, Class II, and Class III dental malocclusion, in the Hispanic population using GMM.
Materials and methods
Sample size and study sample
The sample size was determined based on previous studies which had used linear measurements as opposed to 3-D [39]. The sample size calculations showed that when two groups of 20 samples were compared a power of 80% power detected a 0.90 mm size difference. The total sample size was 120. Forty subjects in each group; 20 male and 20 female, with an age range between 12 and 55 years old. The subjects’ materials were obtained from the records of orthodontic patients at the graduate and faculty orthodontic clinics at University of Texas Health San Antonio, School of Dentistry. The study sample consisted of intra-oral scans (iTero scanner, Align Technologies, San Jose, CA) of Hispanic patients selected from the patient database. It included initial and final orthodontics study models of patients who had previous treatment. Successive cases that met the selection criteria were selected until the sample size was achieved. The following definitions and criteria were used to select subjects for the study groups: Group 1: Class I (ANB angle 0–4 degrees, Class I molar relationship), Group 2: Class II (ANB angle 4 or more degrees, Class II molar relationship), and Group 3: Class III (ANB − 1 or less degrees, Class III molar relationship).
Anzeige
Inclusion criteria and exclusion criteria
The study inclusion criteria were male and female participants from one demographic area (Southwest, Texas, USA) of Hispanic ethnicity and age between 12 and 55 years old with good quality orthodontics intraoral scan and no evident facial and dentoalveolar asymmetry.
Participants were excluded if their teeth were not fully recorded on the intraoral scan, had extensive dental restoration, had traumatized or severely worn teeth, or were patients with craniofacial anomalies.
Scanning and landmarks
All the orthodontic study models were scanned with maximum resolution using an intra-oral scanner (iTero scanner, Align Technologies, San Jose, CA). The dental landmarks were tooth specific, where 19 points of the landmark were used for molars, 16 for premolars, and 12 for anterior teeth. Table 1 provide a list and definitions of all landmarks used for geometric morphometric analysis, 19 landmark for molars, 16 for premolars, and 12 for anterior teeth. The 3-D scanned models were saved in the STL format and identified by the same investigator using the software Checkpoint (Stratovan Corporation, Davis, CA). [40] The x, y, z coordinates defining the landmarks were exported as simple data text files and uploaded onto MorphoJ 1.07a. [41] a software designed to perform geometric morphometric analysis.
Table 1
Landmarks used for Geometric Morphometric Analysis
No. | Upper and Lower Molars | Upper and Lower Premolars | Upper and Lower Anterior Teeth |
|---|---|---|---|
1 | Mesial contact points | Mesial contact points | Mesial contact points |
2 | Distal contact points | Distal contact points | Distal contact points |
3 | Occlusal extent of buccal groove | Lingual cusp tip | Gingival margin lingual middle point |
4 | Occlusal extent of lingual groove | Buccal cusp tip | Incisal middle point |
5 | Mesial lingual cusp tip | Mesial point of buccal cusp | Mesial point incisal |
6 | Distal lingual cusp tip | Distal point of buccal cusp | Distal point incisal |
7 | Mesial buccal cusp tip | Mesial pit | Gingival margin buccal most mesial point |
8 | Distal buccal cusp tip | Distal pit | Gingival margin buccal most distal point |
9 | Central pit | Gingival margin buccal most mesial point | Gingival margin buccal middle point |
10 | Gingival margin buccal most mesial point | Gingival margin buccal most distal point | Middle point buccal |
11 | Gingival margin buccal most distal point | Gingival margin buccal middle point | Gingival margin lingual most mesial point |
12 | Gingival margin buccal middle point | Middle point buccal | Gingival margin lingual most distal point |
13 | Middle point buccal | Gingival margin lingual most mesial point | |
14 | Gingival margin lingual most mesial point | Gingival margin lingual most distal point | |
15 | Gingival margin lingual most distal point | Gingival margin lingual middle point | |
16 | Gingival margin lingual middle point | Middle point lingual | |
17 | Middle point lingual | ||
18 | Distobuccal cusp tip (lower first molar) | ||
19 | Occlusal extent of distobuccal groove (lower first molar) |
Geometric morphometric analysis
The landmarks’ x, y and z coordinates for each tooth were uploaded onto the software MorphoJ 1.07a. [41]. The first pre-analysis process was to detect outliers. MorphoJ 1.07a. provided an output comparing each specimen to the mean output of all individual specimen. The output also included a plotting of the distribution of specimen distances compared to the mean shape of all specimens in a group [41]. The extreme outliers were discarded. Previous work has attributed the extreme outliers to errors in instrumentation.
The individual outcomes were rotated, centered, and scaled to remove all non-shape related variations, using General Procrustes Analysis (GPA). [42, 43] This was followed by canonical variates analysis (CVA) using MorphoJ 1.07a to delineate the features of shape that are unique to each of the four groups. These were displayed as wireframe graphs which were used as the read-out for differences in morphological shape among the groups. A discriminant function analysis (DFA) was used to create wireframe graphs displaying the differences between any two groups.
Validation of landmark reproducibility
For intraobserver error, six permanent teeth measured: Lower left first molar, lower left canine, lower right second premolar, upper right first molar, upper right central incisor, and upper left first premolar. These teeth were identified on intra-oral scans obtained as described above. The specific tooth landmarks were obtained on each of the scanned images, and this was repeated three times with intervals of one week using Stratovan Checkpoint software by one examiner. The data was processed by a Procrustes ANOVA in MorphoJ 1.07a and the digitization error assessed.
Results
Of the 183 sets of dental casts, 120 were analyzed. Forty subjects in each group; 20 male and 20 female in each group, with an age range between 12 and 55 years old. The other 63 dental castes were excluded due to either teeth were not fully recorded on the intraoral scan, presence of extensive dental restorations, traumatized or severely worn teeth, or patient with craniofacial anomalies. The result of the intraobserver analysis showed an excellent landmark reproducibility, where the mean squares (MS) of shape digitization errors were smaller than MS of individuals, Table 2.
Table 2
Digitization Error of Shape
Tooth# | Effect | Error of Shape | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Individual | Digitization Error | |||||||||
SS | MS | DF | F | P | SS | MS | DF | |||
upper right first molar | 0.38238947 | 0.0009656300 | 396 | 36.40 | < 0.0001 | 0.02334273 | 0.0000265258 | 880 | ||
upper right central incisor | 0.92324788 | 0.0035373482 | 261 | 13.36 | < 0.0001 | 0.15354333 | 0.0000337027 | 580 | ||
upper left first premolar | 0.45683295 | 0.0012380297 | 369 | 36.73 | < 0.0001 | 0.02763624 | 0.000038 | 820 | ||
lower left first molar | 1.01743469 | 0.0022609660 | 450 | 132.70 | < 0.0001 | 0.01703879 | 0.0000170388 | 1000 | ||
lower left canine | 2.39298744 | 0.0091685342 | 261 | 26.99 | < 0.0001 | 0.19702193 | 0.0003396930 | 580 | ||
lower right second premolar | 1.08931960 | 0.0029520856 | 369 | 94.50 | < 0.0001 | 0.02561479 | 0.0000312375 | 820 | ||
The multivariate analysis of variance test criteria, F approximations, and P-values for the hypothesis of no overall malocclusion effect using Wilks’s Lambda test showed that shape in all the groups was significantly different for all teeth (Tables 3 and 4). The changes in shape are displayed in wireframe graphs associated with each CV, where the light blue representing the mean configuration of all the individual shapes and the dark blue determines a 5 Mahalanobis distance units change. Figure 1 shows an example of upper lateral incisors between subjects with Angle Class I, II, and III malocclusions. The results yielded a significant difference in the shape of the right and left lateral incisors when compared between groups of different dental malocclusion, where the first canonical (CV1) explains 77.59% of the total variation, followed by 22.40% for the second canonical (CV2). CV1 separates Class II (positive axis) from Class I and III (negative axis). While CV2 separates Class I and Class II (positive axis) from Class III (negative axis).
Table 3
MANOVA Test Criteria and F Approximations for the Hypothesis of No Overall Malocclusion Effect using Wilks’s Lambda for upper Permanent teeth
Statistic | Value | F Value | Num DF | Den DF | Pr > F |
|---|---|---|---|---|---|
Upper right second molars | 0.57674501 | 2.41 | 20 | 152 | 0.0014 |
Upper right first molars | 0.52859224 | 3.79 | 20 | 202 | < 0.0001 |
Upper right second premolars | 0.45628836 | 4.47 | 20 | 186 | < 0.0001 |
Upper right first premolars | 0.51748010 | 3.94 | 20 | 202 | < 0.0001 |
Upper right canines | 0.68827764 | 1.70 | 20 | 166 | 0.0369 |
Upper right lateral incisors | 0.57883233 | 2.80 | 20 | 178 | 0.0002 |
Upper right central incisors | 0.64148491 | 2.24 | 20 | 180 | 0.0029 |
Upper left central incisors | 0.69620727 | 2.00 | 20 | 202 | 0.0085 |
Upper left lateral incisors | 0.62971370 | 2.37 | 20 | 182 | 0.0015 |
Upper left canines | 0.58521175 | 2.64 | 20 | 172 | 0.0004 |
Upper left first premolars | 0.59689648 | 2.91 | 20 | 198 | < 0.0001 |
Upper left second premolars | 0.45628836 | 4.47 | 20 | 186 | < 0.0001 |
Upper left first molars | 0.61148273 | 2.68 | 20 | 192 | 0.0003 |
Upper left second molars | 0.66878648 | 1.60 | 20 | 144 | 0.0590 |
Table 4
MANOVA Test Criteria and F Approximations for the Hypothesis of No Overall Malocclusion Effect using Wilks’s Lambda for lower Permanent teeth
Statistic | Value | F Value | Num DF | Den DF | Pr > F |
|---|---|---|---|---|---|
Lower left second molars | 0.50355871 | 2.74 | 20 | 134 | 0.0003 |
Lower left first molars | 0.38033365 | 5.53 | 20 | 178 | < 0.0001 |
Lower left second premolars | 0.42985192 | 5.04 | 20 | 192 | < 0.0001 |
Lower left first premolars | 0.41738604 | 4.99 | 20 | 182 | < 0.0001 |
Lower left canines | 0.60909834 | 2.11 | 20 | 150 | 0.0060 |
Lower left lateral incisors | 0.66541502 | 2.10 | 20 | 186 | 0.0055 |
Lower left central incisors | 0.74419686 | 1.54 | 20 | 194 | 0.0705 |
Lower right central incisors | 0.64206521 | 2.55 | 20 | 206 | 0.0005 |
Lower right lateral incisors | 0.57122396 | 2.91 | 20 | 180 | < 0.0001 |
Lower right canines | 0.50307833 | 3.24 | 20 | 158 | < 0.0001 |
Lower right first premolars | 0.58449710 | 2.65 | 20 | 172 | 0.0004 |
Lower right second premolars | 0.53326385 | 3.25 | 20 | 176 | < 0.0001 |
Lower right first molars | 0.54459802 | 3.37 | 20 | 190 | < 0.0001 |
Lower right second molars | 0.38576124 | 3.42 | 20 | 112 | < 0.0001 |
Fig. 1
Left side - Scatter Plots of Principal Components CV1 and CV2 of Shape Variables in The Right Lateral Incisors. Right side - Shape Changes in The Right Lateral Incisors Displayed in Wireframe Graphs from Discriminant Function Analysis (DFA) between Each Two Groups of Malocclusion. Light Blue: Average Configuration of All Individuals. Dark Blue: Change on the Positive Axis in Mahalanobis Distance Units
×
![]()
Anzeige
Discriminant function analysis (DFA) generated comparative a wireframe graphs between each two groups. Among the three malocclusion groups, the most morphological difference was between Class I and Class II groups for the following teeth: Upper right second molars, Upper right first molars, Upper right second premolars, Upper right first premolars, Upper right canines, Upper right lateral incisors, Upper right central incisors, Upper left lateral incisors, Upper left canines, Upper left canines, Upper left second premolars, Upper left first molars. Lower left first molars, Lower left second premolars, Lower left lateral incisors, Lower left central incisors, Lower right second premolars, Lower right second molars.
Another morphological difference was noted between Class I and Class III groups for upper left central incisors, and Lower right central incisors. For Class II and Class III groups, morphological difference noted in upper left second molars, lower left second molars, lower left first premolars, lower left canines, lower right lateral incisors, lower right canines, lower right first premolars, and lower right first molars. Tables 5 and 6 showed details description of the Mahalanobis Distances from Conical Variates Analysis for upper and lower Permanent dentition.
Table 5
Mahalanobis Distances from Conical Variates Analysis for Upper Permanent Teeth
Group | Class I | Class II | Class III | |
|---|---|---|---|---|
Upper right second molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.80 | - | < 0.0001 | |
Class III | 2.95 | 3.53 | - | |
Upper right first molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.05 | - | < 0.0001 | |
Class III | 3.03 | 3.03 | - | |
Upper right second premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.56 | - | < 0.0001 | |
Class III | 3.49 | 3.47 | - | |
Upper right first premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.06 | - | < 0.0001 | |
Class III | 2.54 | 2.70 | - | |
Upper right canines | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.22 | - | 0.0020 | |
Class III | 2.03 | 1.74 | - | |
Upper right lateral incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.20 | - | < 0.0001 | |
Class III | 1.94 | 2.36 | - | |
Upper right central incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.58 | - | < 0.0001 | |
Class III | 2.43 | 2.06 | - | |
Upper left central incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.10 | - | 0.0001 | |
Class III | 2.17 | 1.73 | - | |
Upper left lateral incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.40 | - | 0.0003 | |
Class III | 2.05 | 1.79 | - | |
Upper left canines | Class I | - | < 0.0001 | 0.0002 |
Class II | 2.22 | - | < 0.0001 | |
Class III | 1.93 | 2.13 | - | |
Upper left first premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.20 | - | < 0.0001 | |
Class III | 2.65 | 2.40 | - | |
Upper left second premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.24 | - | < 0.0001 | |
Class III | 2.74 | 2.86 | - | |
Upper left first molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.39 | - | < 0.0001 | |
Class III | 2.67 | 2.68 | - | |
Upper left second molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.77 | - | < 0.0001 | |
Class III | 2.43 | 2.93 | - |
Table 6
Mahalanobis Distances from Conical Variates Analysis for Lower Permanent Teeth
Teeth | Group | Class I | Class II | Class III |
|---|---|---|---|---|
Lower left second molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.49 | - | < 0.0001 | |
Class III | 3.63 | 4.02 | - | |
Lower left first molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 5.16 | - | < 0.0001 | |
Class III | 4.63 | 4.56 | - | |
Lower left second premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.85 | - | < 0.0001 | |
Class III | 2.78 | 3.21 | - | |
Lower left first premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 3.59 | - | < 0.0001 | |
Class III | 3.61 | 4.03 | - | |
Lower left canines | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.46 | - | < 0.0001 | |
Class III | 2.60 | 2.64 | - | |
Lower left lateral incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.61 | - | < 0.0001 | |
Class III | 1.81 | 2.18 | - | |
Lower left central incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.00 | - | < 0.0001 | |
Class III | 1.76 | 1.98 | - | |
Lower right central incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.04 | - | < 0.0001 | |
Class III | 2.23 | 1.80 | - | |
Lower right lateral incisors | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.00 | - | < 0.0001 | |
Class III | 2.06 | 2.28 | - | |
Lower right canines | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.69 | - | < 0.0001 | |
Class III | 2.04 | 3.57 | - | |
Lower right first premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.17 | - | < 0.0001 | |
Class III | 3.01 | 3.11 | - | |
Lower right second premolars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.33 | - | < 0.0001 | |
Class III | 3.26 | 3.21 | - | |
Lower right first molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 2.67 | - | < 0.0001 | |
Class III | 3.32 | 3.72 | - | |
Lower right second molars | Class I | - | < 0.0001 | < 0.0001 |
Class II | 4.28 | - | < 0.0001 | |
Class III | 4.09 | 3.76 | - |
Anzeige
Discussion
The GMM methods that were used in this study ensured detailed and objective quantification of the shape of the study samples. In addition, the GMM methods circumvented the inability of traditional metric and angular analyses to separate the effects of size on shape [10, 18, 44, 45]. The relatively large number of landmarks and their optimal distribution facilitated the capture of enough data; thus, ensuring accuracy in describing shape. This contrasts to the techniques which have been used variously to quantify tooth shape. Most of these are derived from traditional linear measurements, especially the ratio of mesio-distal (MD) and bucco-lingual (BL) metrics [46, 47]. The information derived is limited and does not describe most of the in tooth shape. The adaptation of 3D GMM analysis of subjects who fit into three subgroups of malocclusion was not only more efficient, but also more reproducible.
Differences in tooth shape between various racial groups have been reported. Lavelle (1972) studied the dental crown diameters of a White, African American, and Southeast Asian population sample [48]. The study reported the smallest dimensions in the White sample, next was the Asian sample, and the African American subjects had the largest dimension. Merz et al. (1991) reported similar findings, with larger mesio-distal canine, premolar and molar crown dimensions of the African American subjects [49]. According to Yuen et al. (1997), Australian Aboriginals had larger tooth dimensions compared to the Hong Kong Southern Chinese population, and Caucasians who had the smallest dimensions of the populations studied [50]. In a study comparing the mesio-distal tooth width of White British males to British of Pakistani origin, Radnzic (1987) found no statistically significant differences between the groups; concluding that the populations may have shared a common Caucasian lineage [51]. Other studies concluded that the population in Iceland had larger tooth dimensions compared to other Europeans [52]. In addition, Brook et al. (2009) reported that a Southern Chinese population had the largest mesio-distal crown size compared to a White North American population sample, and the Romano British sample had the smallest dimensions in the study [53].
Tooth shape is influenced by several factors [54] These include genetic, epigenetic, and environmental, factors as well as evolutionary adaptation processes [55, 56]. In this study, tooth shape difference amongst teeth in dental malocclusions Class I, II, and III was reported for all the twenty-eight teeth that were analyzed. This contrasts to the difference in centroid size (CS) in the same sample; CS difference was detected in only four teeth. The differences in shape were very similar in principal and symmetrical between left and right. The shape changes were reproducible when analyzed and visualized using wireframes, and scatter plot of the first two conical variates graphs representing a change in Mahalanobis distance units. This finding was unusual from a basic tooth development perspective. Although tooth development is controlled by common morphogenetic pathways, each tooth germ develops as an independent biological entity [57].
The effects of dental malocclusion class on tooth shape in this study can be represented by the upper lateral incisor; a tooth that has been reported to contribute to TSD [58]. The results indicate did not show any significant malocclusion-related difference between the upper lateral incisors in both male and females samples. However, there were significant differences in the shape of lateral incisors among the different malocclusion groups. Buccal views of wireframe graphs show that Class I is wider in shape mesiodistally, and Class II is longer in shape. The shape of the lateral incisor in Class III malocclusion did not show any significant difference from either Class I or II.
Anzeige
The effects on shape reported for the upper lateral incisor in this study are not in conformity with what would have been expected based on previous studies. Benward et al. [58] reported a higher level of tooth deformities in the maxilla of Class III patients. Eustaquio Araujo [59] concluded that the anterior tooth size discrepancy was greater in Class III patients compared to Class I and Class II. In these studies, the maxillary discrepancies in Class III were attributed to the upper lateral incisor. Although it is conceivable that a larger sample size might produce results that would more closely reflect the previous studies, the dissimilarity may be due to differences attributable to the ethnic background of the population studied.
This study revealed differences in tooth shape between the different dental malocclusions on all twenty-eight teeth that were studied, and the pattern of shape differences varied between the dental malocclusions. In addition, the study showed some unique differences in shape of teeth compared to the more commonly studied population as exemplified by the upper lateral incisors. This suggests that the shape variation described is a unique entity inherent in the Hispanic population studied. This study provides some guidelines towards future directions. This includes using larger sample sizes, comparative population studies, and correlation with dental and craniofacial abnormalities, which was a limitation in this study.
Conclusion
The shape variation is a distinct entity inherent in the Hispanic population, where there is a significant difference in 3-D tooth shape between patients with Angle Class I, Class II, and Class III dental malocclusions compared to other populations.
Acknowledgements
The authors would like to acknowledge Dr. Peter Gakunga and Dr. Kate Spradley for their supervision, guidance, suggestions, and corrections. The authors also would like to thank the Deanship of Scientific Research, Qassim University, for funding publication of this project.
Declarations
Ethics approval and consent to participate
The study protocol was carried out in accordance with relevant guidelines and regulations. The protocol reviewed and approved by the Institutional Review Board at the University of Texas Health Science Center, San Antonio, Protocol number HSC20190749E. Informed consent was obtained from participants and/or the legal guardian.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.