A study of correlations between cephalometric measurements in Koreans with normal occlusion by network analysis

Analyzing the correlation between cephalometric measurements is important for improving our understanding of the anatomy in the oral and maxillofacial region. To minimize bias resulting from the design of the input data and to establish a reference for malocclusion research, the aims of this study were to construct the input set by integrating nine cephalometric analyses and to study the correlation structure of cephalometric variables in Korean adults with normal occlusion. To analyze the complex correlation structure among 65 cephalometric variables, which were based on nine classical cephalometric analyses, network analysis was applied to data obtained from 735 adults (368 males, 367 females) aged 18-25 years with normal occlusion. The structure was better revealed through weighted network analysis and minimum spanning tree. Network analysis revealed cephalometric variable clusters and the inter- and intra-correlation structure. Some metrics were divided based on their geometric interpretation rather than their clinical significance. It was confirmed that various classical cephalometric analyses primarily focus on investigating nine anatomical features. Investigating the correlation between cephalometric variables through network analysis can significantly enhance our understanding of the anatomical characteristics in the oral and maxillofacial region, which is a crucial step in studying malocclusion using artificial intelligence.

Three-dimensional phenotype characteristics of skeletal class III malocclusion in adult Chinese: a principal component analysis-based cluster analysis

BACKGROUND

Skeletal class III malocclusion has a diverse and complicated aetiology involving environmental and genetic factors. It is critical to correctly classify and define this malocclusion to be diagnosed and treated on a clinically sound basis. Thus, this study aimed to provide reliable and detailed measurements in a large ethnically homogeneous sample of Chinese adults to generate an adequate phenotypic clustering model to identify and describe the skeletal variation present in skeletal class III malocclusion.

MATERIALS AND METHODS

This is a retrospective cross-sectional study in which 500 pre-treatments cone-beam computed tomography (CBCT) scans of patients with skeletal class III malocclusion (250 males and 250 females) were selected following specific selection criteria. Seventy-six linear, angular, and ratios measurements were three-dimensionally analysed using InVivo 6.0.3 software. These measurements were categorised into 47 skeletal, 18 dentoalveolar, and 11 soft tissue variables. Multivariate reduction methods: principal component analyses and cluster analyses were used to present the most common phenotypic groupings of skeletal class III malocclusion in Han ethnic group of Chinese adults.

RESULTS

The principal component analysis revealed eight principal components accounted for 72.9% of the overall variation of the data produced from the seventy-six variables. The first four principal components accounted for 53.37% of the total variations. They explained the most variation in data and consisted mainly of anteroposterior and vertical skeletal relationships. The cluster analysis identified four phenotypes of skeletal class III malocclusion: C1, 34%; C2, 11.4%; C3, 26.4%; and C4, 28.2%.

CONCLUSION

Based on three-dimensional analyses, four skeletal class III malocclusion distinct phenotypic variations were defined in a large sample of the adult Chinese population, showing the occurrence of phenotypic variation between identified clusters in the same ethnic group. These findings might serve as a foundation for accurate diagnosis and treatment planning of each cluster and future genetic studies to determine the causative gene(s) of each cluster.

The Validity of Machine Learning Procedures in Orthodontics: What Is Still Missing?

Artificial intelligence (AI) models and procedures hold remarkable predictive efficiency in the medical domain through their ability to discover hidden, non-obvious clinical patterns in data. However, due to the sparsity, noise, and time-dependency of medical data, AI procedures are raising unprecedented issues related to the mismatch between doctors’ mentalreasoning and the statistical answers provided by algorithms. Electronic systems can reproduce or even amplify noise hidden in the data, especially when the diagnosis of the subjects in the training data set is inaccurate or incomplete. In this paper we describe the conditions that need to be met for AI instruments to be truly useful in the orthodontic domain. We report some examples of computational procedures that are capable of extracting orthodontic knowledge through ever deeper patient representation. To have confidence in these procedures, orthodontic practitioners should recognize the benefits, shortcomings, and unintended consequences of AI models, as algorithms that learn from human decisions likewise learn mistakes and biases.

Characterization of phenotypes of skeletal Class III malocclusion in Korean adult patients treated with orthognathic surgery using cluster analysis

OBJECTIVES

To characterize the phenotypes of skeletal Class III malocclusion in adult patients who underwent orthognathic surgery (OGS).

MATERIALS AND METHODS

The sample consisted of 326 patients with Class III malocclusion treated with OGS (170 men and 156 women; mean age, 22.2 years). Using lateral cephalograms taken at initial visits, 13 angular variables and one ratio cephalometric variable were measured. Using three representative variables obtained from principal components analysis (SNA, SNB, and Björk sum), K-means cluster analysis was performed to classify the phenotypes. Statistical analysis was conducted to characterize the differences in the cephalometric variables among the clusters.

RESULTS

Class III phenotypes were classified into nine clusters from the following four major groups: (1) retrusive maxilla group, clusters 7 and 9 (7.1% and 5.5%; severely retrusive maxilla, normal mandible, severe and moderate hyperdivergent, respectively) and cluster 6 (9.2%; retrusive maxilla, normal mandible, normodivergent); (2) relatively protrusive mandible group, cluster 2 (20.9%; normal maxilla, normal mandible, hyperdivergent); (3) protrusive mandible group, clusters 3 and 1 (11.7% and 15.3%; normal maxilla, protrusive mandible, normodivergent and hyperdivergent, respectively) and clusters 8 and 4 (15.3% and 3.7%; normal maxilla, severe protrusive mandible, normodivergent and hypodivergent, respectively); and (4) protrusive maxilla and protrusive mandible group, cluster 5 (11.4%; protrusive maxilla, severely protrusive mandible, normodivergent). Considerations for presurgical orthodontic treatment and OGS planning were proposed based on the Class III phenotypes.

CONCLUSIONS

Because the anteroposterior position of the maxilla and rotation of the mandible by a patient's vertical pattern determine Class III phenotypes, these variables should be considered in diagnosis and treatment planning for patients who have skeletal Class III malocclusion.

Craniofacial growth predictors for class II and III malocclusions: A systematic review

OBJECTIVE

To evaluate the validity of craniofacial growth predictors in class II and III malocclusion.

MATERIAL AND METHODS

An electronic search was conducted until August 2020 in PubMed, Cochrane Library, Embase, EBSCOhost, ScienceDirect, Scopus, Bireme, Lilacs and Scielo including all languages. The articles were selected and analyzed by two authors independently and the selected studies was assessed using the 14-item Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS-2). The quality of evidence and strength of recommendation was assessed by the GRADE tool.

RESULTS

In a selection process of two phases, 10 articles were included. The studies were grouped according to malocclusion growth predictor in (1) class II (n = 4); (2) class III (n = 5) and (3) class II and III (n = 1). The predictors were mainly based on data extracted from cephalometries and characterized by: equations, structural analysis, techniques and computer programs among others. The analyzed studies were methodologically heterogeneous and had low to moderate quality. For class II malocclusion, the predictors proposed in the studies with the best methodological quality were based on mathematical models and the Fishman system of maturation assessment. For class III malocclusion, the Fishman system could provide adequate growth prediction for short- and long-term.

CONCLUSIONS

Because of the heterogeneity of the design, methodology and the quality of the articles reviewed, it is not possible to establish only a growth prediction system for class II and III malocclusion. High-quality cohort studies are needed, well defined data extraction from cephalometries, radiographies and clinical characteristics are required to design a reliable predictor.

Prognostic approach to Class III malocclusion through case-based reasoning

OBJECTIVE

This investigation evaluates the evidence of case-based reasoning (CBR) in providing additional information on the prediction of future Class III craniofacial growth.

SETTINGS AND SAMPLE POPULATION

The craniofacial characteristics of 104 untreated Class III subjects (7-17 years of age), monitored with two lateral cephalograms obtained during the growth process, were evaluated.

MATERIALS AND METHODS

Data were compared with the skeletal characteristics of subjects who showed a high degree of skeletal imbalance ('prototypes') obtained from a large data set of 1263 Class III cross-sectional subjects (7-17 years of age).

RESULTS

The degree of similarity of longitudinal subjects with the most unbalanced prototypes allowed the identification of subjects who would develop a subsequent unfavourable skeletal growth (accuracy: 81%). The angle between the palatal plane and the sella-nasion line (PP-SN angle) and the Wits appraisal were two additional craniofacial features involved in the early prediction of the adverse progression of the Class III skeletal imbalance.

CONCLUSIONS

Case-based reasoning methodology, which uses a personalized inference method, may bring additional information to approximate the skeletal progression of Class III malocclusion.

Sub-clustering in skeletal class III malocclusion phenotypes via principal component analysis in a southern European population

The main aim of this study was to generate an adequate sub-phenotypic clustering model of class III skeletal malocclusion in an adult population of southern European origin. The study design was conducted in two phases, a preliminary cross-sectional study and a subsequent discriminatory evaluation by main component and cluster analysis to identify differentiated skeletal sub-groups with differentiated phenotypic characteristics. Radiometric data from 699 adult patients of southern European origin were analyzed in 212 selected subjects affected by class III skeletal malocclusion. The varimax rotation was used with Kaiser normalization, to prevent variables with more explanatory capacity from affecting the rotation. A total of 21,624 radiographic measurements were obtained as part of the cluster model generation, using a total set of 55 skeletal variables for the subsequent analysis of the major component and cluster analyses. Ten main axes were generated representing 92.7% of the total variation. Three main components represented 58.5%, with particular sagittal and vertical variables acting as major descriptors. Post hoc phenotypic clustering retrieved six clusters: C1:9.9%, C2:18.9%, C3:33%, C4:3.77%, C5:16%, and C6:16%. In conclusion, phenotypic variation was found in the southern European skeletal class III population, demonstrating the existence of phenotypic variations between identified clusters in different ethnic groups.

Computer-aided heuristics in orthodontics

INTRODUCTION

During the decision-making process, physicians rely on heuristics that consist of simple, useful procedures for solving problems, intuitive shortcuts that produce reliable decisions based on limited information. In clinical situations characterized by a high degree of uncertainty such as those encountered in orthodontics, cognitive biases and judgment errors related to heuristics are not uncommon. This study aimed at promoting trust in the effective interface between the intuitive reasoning of the orthodontic practitioner and the computational heuristics emerging from simple statistical models.

METHODS

We propose an integrative model based on the interaction between clinical reasoning and 2 computational tools, cluster analysis and fast-and-frugal trees, to extract a structured craniofacial representation of untreated subjects with Class III malocclusion and to forecast the worsening of the malocclusion over time.

RESULTS

Cluster analysis of cephalometric values from 144 growing subjects with Class III malocclusion followed longitudinally (T1: mean age, 10.2 ± 1.9 years; T2: mean age, 13.8 ± 2.7 years) produced 3 morphologic subgroups with predominant sagittal, vertical, and slight maxillomandibular imbalances. Fast-and-frugal trees applied to different subgroups extracted heuristics that improved the prediction of key features associated with adverse craniofacial growth.

CONCLUSIONS

Provided that cephalometric values are placed in the appropriate framework, the matching between simple and fast computational approaches and clinical reasoning could help the intuitive logic, perception, and cognitive inferences of orthodontic practitioners on the outcome of patients affected by Class III disharmony, decreasing errors associated with flawed judgments and improving the accuracy of decision making.

Novel Sub-Clustering of Class III Skeletal Malocclusion Phenotypes in a Southern European Population Based on Proportional Measurements

Current phenotypic characterizations of Class III malocclusion are influenced more by gender or ethnic origin than by raw linear skeletal measurements. The aim of the present research is to develop a Class III skeletal malocclusion sub-phenotype characterization based on proportional cranial measurements using principal component analysis and cluster analysis. Radiometric data from 212 adult subjects (115 women and 96 men) of southern European origin affected by Class III skeletal malocclusion were analyzed. A total of 120 measurements were made, 26 were proportional skeletal measurements, which were used to perform principal component analysis and subsequent cluster analysis. The remaining 94 supplementary measurements were used for a greater description of the identified clusters. Principal component analysis established eight principal components that explained 85.1% of the total variance. The first three principal components explained 51.4% of the variance and described mandibular proportions, anterior facial height proportions, and posterior-anterior cranial proportions. Cluster analysis established four phenotypic subgroups, representing 18.4% (C1), 20.75% (C2), 38.68% (C3), and 22.17% (C4) of the sample. A new sub-clustering of skeletal Class III malocclusions that avoids gender influence is provided. Our results improve clinicians' resources for Class III malocclusion and could improve the diagnostic and treatment approaches for this malocclusion.

Evaluation of the morphometric covariation between palatal and craniofacial skeletal morphology in class III malocclusion growing subjects

Background

To study the covariation between palatal and craniofacial skeletal morphology in Class III growing patients through geometric morphometric analysis (GMM).

Methods

In this retrospective study, 54 Class III subjects (24F,30M;7.6 ± 0.8yy) were enrolled following these inclusion criteria: European ancestry, Class III skeletal and dental relationship, early mixed dentition, prepubertal skeletal maturation, familiarity for Class III malocclusion, no pseudo Class III malocclusion. Each patient provided upper digital cast and cephalogram before starting the therapy. Landmarks and semilandmarks were digitized (239 on the casts;121 on the lateral radiographs) and GMM was used. Procrustes analysis and principal component analysis (PCA) were applied to show the principal components of palatal and craniofacial skeletal shape variation. Two-block partial least squares analysis (PLS) was used to assess pattern of covariation between palatal and craniofacial morphology.

Results

Regarding palatal shape variation, PC with largest variance (PC1) described morphological changes in the three space dimensions, while, concerning the craniofacial complex components, PC1 revealed morphological differences along the vertical plane. A significant covariation was found between palatal and craniofacial shape. PLS1 accounted for more than 61,7% of the whole covariation, correlating the craniofacial divergence to palatal height and width.

Conclusions

In Class III subjects increments of angle divergence are related to a narrow and high palate.

Exploiting the interplay between cross-sectional and longitudinal data in Class III malocclusion patients

The aim of the study was to investigate how to improve the forecasting of craniofacial unbalance risk during growth among patients affected by Class III malocclusion. To this purpose we used computational methodologies such as Transductive Learning (TL), Boosting (B), and Feature Engineering (FE) instead of the traditional statistical analysis based on Classification trees and logistic models. Such techniques have been applied to cephalometric data from 728 cross-sectional untreated Class III subjects (6-14 years of age) and from 91 untreated Class III subjects followed longitudinally during the growth process. A cephalometric analysis comprising 11 variables has also been performed. The subjects followed longitudinally were divided into two subgroups: favourable and unfavourable growth, in comparison with normal craniofacial growth. With respect to traditional statistical predictive analytics, TL increased the accuracy in identifying subjects at risk of unfavourable growth. TL algorithm was useful in diffusion of information from longitudinal to cross-sectional subjects. The accuracy in identifying high-risk subjects to growth worsening increased from 63% to 78%. Finally, a further increase in identification accuracy, up to 83%, was produced by FE. A ranking of important variables in identifying subjects at risk of growth worsening, therefore, has been obtained.

Subclustering in Skeletal Class III Phenotypes of Different Ethnic Origins: A Systematic Review

OBJECTIVE

We aimed to systematically review articles investigating the efficiency of the clustering of skeletal class III malocclusion phenotypic subtypes of different ethnic origins as a diagnostic tool.

METHODS

The review protocol was structured in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement and registered in Prospero (CRD42016053865). A survey of articles published up to March 2018 investigating the identification of different subgroups of skeletal class III malocclusion via cluster analysis was performed using 11 electronic databases. Any type of study design that addressed the classification of subclusters of class III malocclusion was considered. The Newcastle-Ottawa scale for cohort and cross-sectional (modified) studies was used for quality assessment.

RESULTS

The final selection included 7 studies that met all the criteria for eligibility (% overall agreement 0.889, free marginal kappa 0.778). All studies identified at least 3 different types of class III clusters (ranging from 3 to 14 clusters; the total variation of the prevalence of each cluster ranged from 0.2% to 36.0%). The main shared variables used to describe the more prevalent clusters in the studies included were vertical measurements (Ar-Go-Me: 117.51°-135.8°); sagittal measurements: maxilla (SNA: 75.3°-82.95°), mandible (SNB: 77.03°-85.0°). With regard to ethnicity, a mean number of 8.5 and 3.5 clusters of class III were retrieved for Asian and Caucasian population, respectively.

CONCLUSIONS

The total number of clusters identified varied from 3 to 14 to explain all the variability in the phenotype class III malocclusions. Although each extreme may be too simple or complex to facilitate an exhaustive but useful classification for clinical use, a classification system including 4 to 7 clusters may prove to be efficient for clinical use in conjunction with complete and meticulous subgrouping.

CLINICAL SIGNIFICANCE

The identification and description of a subclustering classification system may constitute an additional step toward more precise orthodontic/orthopedic diagnosis and treatment of skeletal class III malocclusion.

Understanding interactions among cephalometrics variables during growth in untreated Class III subjects

Objective

The aim of the present study was to apply a computational method commonly used in data mining discipline, classification trees (CTs), to evaluate the growth features in untreated Class III subjects.

Materials and methods

CT was applied to data from 91 untreated Class III subjects (48 females and 43 males) and compared with the results of discriminant analysis (DA). For all subjects, lateral cephalograms were available at T1 (mean age 10.4 ± 2.0 years) and at T2 (mean age 15.4 ± 1.9 years). A cephalometric analysis comprising 11 variables was performed. The subjects were divided into two subgroups, unfavourable ('Bad') and favourable ('Good') growers, according to the quality of the skeletal growth rate in comparison with the normal craniofacial growth.

Results

CTs showed that the most informative attribute for the prediction of favourable/unfavourable skeletal growth was the SNA angle. Subjects with SNA values lower than 79.1 degrees showed a risk of 94 per cent of growing unfavourably. DA was able to select palatal plane to mandibular plane angle as predictors. DA, however, showed a statistically significant higher rate of misclassification when compared with CTs (40.7 per cent versus 12.1 per cent, binomial exact test: odds ratio = 6.20; P < 0.0001).

Conclusions

CTs provided a valid measure of elucidating the effective contribution of craniofacial characteristics in predicting favourable/unfavourable growth in untreated Class III subjects.

Prediction of Class III treatment outcomes through orthodontic data mining

OBJECTIVE

To determine whether it is possible to predict Class III treatment outcomes on the basis of a model derived from a combination of computational analyses derived from complexity science, such as fuzzy clustering repartition and network analysis.

METHODS

Cephalometric data of 54 Class III patients (32 females, 22 males) taken before (T1, mean age 8.2 ± 1.6 years) and after (T2, mean age 14.6 ± 1.8 years) early rapid maxillary expansion and facemask therapy followed by fixed appliances were analysed. Patients were classified at T1 on the basis of high membership grade into three main dentoskeletal fuzzy cluster phenotypes: hyperdivergent (HD), hypermandibular (HM), and balanced (Bal) phenotypes. The prevalence rate of successful and unsuccessful cases at T2 was calculated for the three clusters and compared by means of Fisher's exact test corrected for multiple testing (Holm-Bonferroni method).

RESULTS

Unsuccessful cases were 9 out of 54 patients (16.7%). Once patients were framed into their cluster membership, the individualized pre-treatment prediction of unsuccessful cases was largely differentiated: HD and HM patients showed a significantly greater prevalence rate of unsuccessful cases than Bal patients (0% in Bal cluster, 28.6% in HM cluster, and 33.3% in HD cluster). Network analysis captured some noticeable interdependencies of Class III patients, showing a more connected interactive structure of cephalometric data sets in HM and HD patients compared with Bal patients. The results were confirmed after minimizing the geometrical connections between cephalometric variables in the model.

CONCLUSIONS

Fuzzy clustering repartition can be usefully used to estimate an individualized risk of unsuccessful treatment outcome in Class III patients.