Accuracy and precision of a 3D anthropometric facial analysis with and without landmark labeling before image acquisition

The 3D Tele Motion Tracking for the Orthodontic Facial Analysis

Aim. This study aimed to evaluate the reliability of 3D-TMT, previously used only for dynamic testing, in a static cephalometric evaluation. Material and Method. A group of 40 patients (20 males and 20 females; mean age 14.2 ± 1.2 years; 12–18 years old) was included in the study. The measurements obtained by the 3D-TMT cephalometric analysis with a conventional frontal cephalometric analysis were compared for each subject. Nine passive markers reflectors were positioned on the face skin for the detection of the profile of the patient. Through the acquisition of these points, corresponding plans for three-dimensional posterior-anterior cephalometric analysis were found. Results. The cephalometric results carried out with 3D-TMT and with traditional posterior-anterior cephalometric analysis showed the 3D-TMT system values are slightly higher than the values measured on radiographs but statistically significant; nevertheless their correlation is very high. Conclusion. The recorded values obtained using the 3D-TMT analysis were correlated to cephalometric analysis, with small but statistically significant differences. The Dahlberg errors resulted to be always lower than the mean difference between the 2D and 3D measurements. A clinician should use, during the clinical monitoring of a patient, always the same method, to avoid comparing different millimeter magnitudes.

Are Three-Dimensional Monitors More Appropriate Than Two-Dimensional Monitors in the Three-Dimensional Analysis?

In orthognathic surgery, three-dimensional (3D) program-based analysis of 3D reconstructions of computed tomography (CT) images is commonly used, and images viewed on a monitor. The authors compared the coordinates of facial landmarks on images in a 3D program displayed on a two-dimensional (2D) (standard) or 3D monitor. Facial bone CT data from 30 patients were reconstructed in 3D. Four researchers identified 33 facial landmarks, 3 times each on 2D and 3D monitors, for each patient, by their x-, y-, and z-coordinates. The time taken to complete these identifications was measured.For each set of coordinates, the average intraclass coefficient was >0.8 for 2D and 3D analyses, as well as among 4 readers. It took on average of 2 minutes 46 seconds to identify the landmarks on the 2D monitor, compared with 2 minutes 25 seconds on the 3D monitor. The variance of individual coordinates differed when measured on 2D or 3D monitor. The landmarks affected were located near the median region of the facial area, and are important for setting the reference sagittal plane during diagnosis for orthognathic surgery. Therefore, identifying facial landmarks using 3D monitors may be helpful for conducting accurate facial diagnoses.

The Accuracy of Conformation of a Generic Surface Mesh for the Analysis of Facial Soft Tissue Changes

PURPOSE

Three dimensional analysis of the face is required for the assessment of complex changes following surgery, pathological conditions and to monitor facial growth. The most suitable method may be "dense surface correspondence".

MATERIALS AND METHODS

This method utilizes a generic facial mesh and "conformation process" to establish anatomical correspondences between two facial images. The aim of this study was to validate the use of conformed meshes to measure simulated maxillary and mandibular surgical movements. The "simulation" was performed by deforming the actual soft tissues of the participant during image acquisition. The study was conducted on 20 volunteers and used 77 facial landmarks pre-marked over six anatomical regions; left cheek, right cheek, left upper lip, philtrum, right upper lip and chin region. Each volunteer was imaged at rest and after performing 5 different simulated surgical procedures using 3D stereophotogrammetry. The simulated surgical movement was determined by measuring the Euclidean distances and the mean absolute x, y and z distances of the landmarks making up the six regions following digitization. A generic mesh was then conformed to each of the aligned six facial 3D images. The same six regions were selected on the aligned conformed simulated meshes and the surgical movement determined by determining the Euclidean distances and the mean absolute x, y and z distances of the mesh points making up the six regions were determined.

RESULTS

In all cases the mean Euclidian distance between the simulated movement and conformed region was less than 0.7 mm. For the x, y and z directions the majority of differences in the mean absolute distances were less than 1.0mm except in the x-direction for the left and right cheek regions, which was above 2.0 mm.

CONCLUSIONS

This concludes that the conformation process has an acceptable level of accuracy and is a valid method of measuring facial change between two images i.e. pre- and post-surgery. The conformation accuracy is higher toward the center of the face than the peripheral regions.

Evaluation of a Three-Dimensional Stereophotogrammetric Method to Identify and Measure the Palatal Surface Area in Children with Unilateral Cleft Lip and Palate

Objective

To assess a three-dimensional (3D) stereophotogrammetric method for area delimitation and evaluation of the dental arches of children with unilateral cleft lip and palate (UCLP). Obtained data were also used to assess the 3D changes occurring in the maxillary arch with the use of orthopedic therapy prior to rhinocheiloplasty and before the first year of life.

Design

Within the collaboration between the Università degli Studi di Milano (Italy) and the University CES of Medellin (Colombia), 96 palatal cast models obtained from neonatal patients with UCLP were analyzed using a 3D stereophotogrammetric imaging system.

Main Outcome Measures

The area of the minor and greater cleft segments on the digital dental cast surface were delineated by the visualization tool of the stereophotogrammetric software and then examined. “Trueness” of the measurements, as well as systematic and random errors between operators' tracings (“precision”) were calculated.

Results

The method gave area measurements close to true values (errors lower than 2%), without systematic measurement errors for tracings by both interoperators and intraoperators ( P > .05). Statistically significant differences ( P < .05) were noted for alveolar segment and time.

Conclusions

Maxillary segments have the potential for growth during presurgical orthopedic treatment in the early neonatal period. The cleft segment delimitation on digital dental casts and area measurements by the 3D stereophotogrammetric system revealed an accurate (true and precise) method for evaluating the stone casts of newborn patients with UCLP.

3-D Facial Landmark Localization With Asymmetry Patterns and Shape Regression from Incomplete Local Features

We present a method for the automatic localization of facial landmarks that integrates nonrigid deformation with the ability to handle missing points. The algorithm generates sets of candidate locations from feature detectors and performs combinatorial search constrained by a flexible shape model. A key assumption of our approach is that for some landmarks there might not be an accurate candidate in the input set. This is tackled by detecting partial subsets of landmarks and inferring those that are missing, so that the probability of the flexible model is maximized. The ability of the model to work with incomplete information makes it possible to limit the number of candidates that need to be retained, drastically reducing the number of combinations to be tested with respect to the alternative of trying to always detect the complete set of landmarks. We demonstrate the accuracy of the proposed method in the face recognition grand challenge database, where we obtain average errors of approximately 3.5 mm when targeting 14 prominent facial landmarks. For the majority of these our method produces the most accurate results reported to date in this database. Handling of occlusions and surfaces with missing parts is demonstrated with tests on the Bosphorus database, where we achieve an overall error of 4.81 and 4.25 mm for data with and without occlusions, respectively. To investigate potential limits in the accuracy that could be reached, we also report experiments on a database of 144 facial scans acquired in the context of clinical research, with manual annotations performed by experts, where we obtain an overall error of 2.3 mm, with averages per landmark below 3.4 mm for all 14 targeted points and within 2 mm for half of them. The coordinates of automatically located landmarks are made available on-line.

Three-dimensional human facial morphologies as robust aging markers

Aging is associated with many complex diseases. Reliable prediction of the aging process is important for assessing the risks of aging-associated diseases. However, despite intense research, so far there is no reliable aging marker. Here we addressed this problem by examining whether human 3D facial imaging features could be used as reliable aging markers. We collected > 300 3D human facial images and blood profiles well-distributed across ages of 17 to 77 years. By analyzing the morphological profiles, we generated the first comprehensive map of the aging human facial phenome. We identified quantitative facial features, such as eye slopes, highly associated with age. We constructed a robust age predictor and found that on average people of the same chronological age differ by ± 6 years in facial age, with the deviations increasing after age 40. Using this predictor, we identified slow and fast agers that are significantly supported by levels of health indicators. Despite a close relationship between facial morphological features and health indicators in the blood, facial features are more reliable aging biomarkers than blood profiles and can better reflect the general health status than chronological age.

Metric precision via soft-tissue landmarks in three-dimensional structured-light scans of human faces

OBJECTIVES

The goal of the current work was to investigate the suitability of ten facial soft-tissue landmarks (trichion, glabella, nasion, left and right orbital, subnasal, left and right porion, pogonion, gnathion) as reference points for metric facial analysis by analyzing their intra- and interserial precision in all three dimensions.

METHODS

The faces of 32 volunteers were scanned using a standardized protocol with a structured-light scanner (FaceSCAN(3D) Scientific Photolab 60 Hz; 3D Shape, Erlangen, Germany). Three examiners placed the landmarks twice within a 2-week interval. Image processing (Onyx Ceph(3); Image Instruments; Chemnitz, Germany) and statistical (SAS 9.2; SAS Institute, Cary, NC, USA) software was used for analysis. All measurements were corrected for mean values per patient and analyzed for intraserial and interserial error by model II ANOVA for a simple classification with random effects.

RESULTS

Median intraserial precision was 0.40 mm (range 0.05-1.01 mm) overall, including 0.40 mm (0.33-0.85 mm) on the x-axis, 0.64 mm (0.36-0.87 mm) on the y-axis, and 0.27 mm (0.05-1.01 mm) on the z-axis. Interserial precision was substantially lower at a median of 0.05 mm (0-0.22 mm), often not statistically assessable with intraserial precision. We observed no landmark-associated differences; in particular, the medians of the bilateral landmarks orbital (intraserial: 0.40 mm; interserial: 0.02 mm) and porion (intraserial: 0.36 mm; interserial: small and not assessable) were in the middle of the range of our results. Trichion (intraserial: 0.73 mm; interserial: 0.05 mm) and gnathion (intraserial: 0.87 mm; interserial: 0.20 mm) revealed the highest degrees of intraserial measurement imprecision. Outliers were identified in 1.2% (64 of 5400) of measurements.

CONCLUSION

All selected landmarks offer suitably high levels of intra- and interserial precision for the three-dimensional (3D) metric assessment of facial soft-tissue parameters. No difference between (bi)lateral and facial midline landmarks was noted.

Three-dimensional evaluation of the relationship between jaw divergence and facial soft tissue dimensions

OBJECTIVE

To evaluate the relationship between mandibular divergence and vertical and transverse dimensions of the face.

MATERIALS AND METHODS

A sample was recruited from the orthodontic clinic of the University of Otago, New Zealand. The recruited participants (N  =  60) were assigned to three different groups based on the mandibular plane angle (hyperdivergent, n  =  20; normodivergent, n  =  20; and hypodivergent, n  =  20). The sample consisted of 31 females and 29 males, with a mean age of 21.1 years (SD ± 5.0). Facial scans were recorded for each participant using a three-dimensional (3D) white-light scanner and then merged to form a single 3D image of the face. Vertical and transverse measurements of the face were assessed from the 3D facial image.

RESULTS

The hyperdivergent sample had a significantly larger total and lower anterior facial height than the other two groups (P < .05), although no difference was found for the middle facial height (P > .05). Similarly, there were no significant differences in the transverse measurements of the three study groups (P > .05). Both gender and body mass index (BMI) had a greater influence on the transverse dimension.

CONCLUSIONS

Hyperdivergent facial types are associated with a long face but not necessarily a narrow face. Variations in facial soft tissue vertical and transversal dimensions are more likely to be due to gender. Body mass index has a role in mandibular width (GoGo) assessment.

Stereophotogrammetric evaluation of tooth-induced labial protrusion

PURPOSE

To better manage dental treatment outcome, a previsualization of desired appearances can be used to understand patients' wishes. A deeper comprehension of labial modifications related to hard-tissue movements is advantageous. The purpose of the study was to evaluate tooth restoration-induced labial displacements in three dimensions.

MATERIALS AND METHODS

In a group of 20 healthy Caucasian individuals, simulations of vestibular translations of maxillary anterior crowns were obtained by placing an acrylic resin veneer on the labial surfaces of maxillary incisors and canines. Three-dimensional stereophotogrammetric acquisitions were made to evaluate soft-tissue changes induced by the simulations. Linear dislocation of selected landmarks and labial surfaces were quantified using dedicated software.

RESULTS

All paired and two midline labial landmarks had significant displacements, ranging from 13% (Subnasale landmark) to 103% (left Cheilion landmark) of veneer thickness (2 mm thick). A significant positive correlation was obtained between the lower lip displacement and overjet values.

CONCLUSIONS

The vestibular shift of maxillary incisors and canines affect both upper and lower vermilion areas, without involving cutaneous perilabial landmarks.

Orthodontic soft-tissue parameters: a comparison of cone-beam computed tomography and the 3dMD imaging system

INTRODUCTION

Orthodontists rely heavily on soft-tissue analysis to determine facial esthetics and treatment stability. The aim of this retrospective study was to determine the equivalence of soft-tissue measurements between the 3dMD imaging system (3dMD, Atlanta, Ga) and the segmented skin surface images derived from cone-beam computed tomography.

METHODS

Seventy preexisting 3dMD facial photographs and cone-beam computed tomography scans taken within minutes of each other for the same subjects were registered in 3 dimensions and superimposed using Vultus (3dMD) software. After reliability studies, 28 soft-tissue measurements were recorded with both imaging modalities and compared to analyze their equivalence. Intraclass correlation coefficients and Bland-Altman plots were used to assess interexaminer and intraexaminer repeatability and agreement. Summary statistics were calculated for all measurements. To demonstrate equivalence of the 2 methods, the difference needed a 95% confidence interval contained entirely within the equivalence limits defined by the repeatability results.

RESULTS

Statistically significant differences were reported for the vermilion height, mouth width, total facial width, mouth symmetry, soft-tissue lip thickness, and eye symmetry.

CONCLUSIONS

There are areas of nonequivalence between the 2 imaging methods; however, the differences are clinically acceptable from the orthodontic point of view.

Precise Evaluation of Anthropometric 2D Software Processing of Hand in Comparison with Direct Method

Various studies carried out on different photo anthropometry, but each one had some deficiencies which during the years they have been resolved. The objective of this paper is to test the efficiency of two-dimensional image processing software in photo anthropometry of hand. In this applied research, 204 office workers and industrial workers were selected. Their hands were measured by manual with photo anthropometric methods. In this study, designing the "Hand Photo Anthropometry Set," we tried to fix the angle and distance of the camera in all of the photos. Thus, some of the common mistakes in photo anthropometric method got controlled. The taken photos were analyzed by Digimizer software, version 4.1.1.0 and Digital Caliper (Model: Mitutoyo Corp., Tokyo, Japan) was used via manual method. t-test statistical test on data revealed that there is no significant difference between the manual and photo anthropometric results (P > 0.05) and the correlation coefficients for hand dimensions are similar in both methods illustrated in the range of 0.71-0.95. The statistical analyses showed that photo anthropometry can be replaced with manual methods. Furthermore, it can provide a great help to develop an anthropometric database for work gloves manufacturers. Since the hand anthropometry is a necessary input for tool design, this survey can be used to determine the percentiles of workers' hands.

Three-dimensional evaluation of changes in lip position from before to after orthodontic appliance removal

INTRODUCTION

Our objectives were to develop a reproducible method of superimposing 3-dimensional images for measuring soft-tissue changes over time and to use this method to document changes in lip position after the removal of orthodontic appliances.

METHODS

Three-dimensional photographs of 50 subjects were made in repose and maximum intercuspation before and after orthodontic appliance removal with a stereo camera. For reliability assessment, 2 photographs were repeated for 15 patients. The images were registered on stable areas, and surface-to-surface measurements were made for defined landmarks.

RESULTS

Mean changes were below the level of clinical significance (set at 1.5 mm). However, 51% and 18% of the subjects experienced changes greater than 1.5 mm at the commissures and lower lips, respectively.

CONCLUSIONS

The use of serial 3-dimensional photographs is a reliable method of documenting soft-tissue changes. Soft-tissue changes after appliance removal are not clinically significant; however, there is great individual variability.

Methods to quantify soft-tissue based facial growth and treatment outcomes in children: a systematic review

Context

Technological advancements have led craniofacial researchers and clinicians into the era of three-dimensional digital imaging for quantitative evaluation of craniofacial growth and treatment outcomes.

Objective

To give an overview of soft-tissue based methods for quantitative longitudinal assessment of facial dimensions in children until six years of age and to assess the reliability of these methods in studies with good methodological quality.

Data Source

PubMed, EMBASE, Cochrane Library, Web of Science, Scopus and CINAHL were searched. A hand search was performed to check for additional relevant studies.

Study Selection

Primary publications on facial growth and treatment outcomes in children younger than six years of age were included.

Data Extraction

Independent data extraction by two observers. A quality assessment instrument was used to determine the methodological quality. Methods, used in studies with good methodological quality, were assessed for reliability expressed as the magnitude of the measurement error and the correlation coefficient between repeated measurements.

Results

In total, 47 studies were included describing 4 methods: 2D x-ray cephalometry; 2D photography; anthropometry; 3D imaging techniques (surface laser scanning, stereophotogrammetry and cone beam computed tomography). In general the measurement error was below 1 mm and 1° and correlation coefficients range from 0.65 to 1.0.

Conclusion

Various methods have shown to be reliable. However, at present stereophotogrammetry seems to be the best 3D method for quantitative longitudinal assessment of facial dimensions in children until six years of age due to its millisecond fast image capture, archival capabilities, high resolution and no exposure to ionizing radiation.

Large-scale objective phenotyping of 3D facial morphology

Abnormal phenotypes have played significant roles in the discovery of gene function, but organized collection of phenotype data has been overshadowed by developments in sequencing technology. In order to study phenotypes systematically, large-scale projects with standardized objective assessment across populations are considered necessary. The report of the 2006 Human Variome Project meeting (Cotton et al, 2007) recommended documentation of phenotypes through electronic means by collaborative groups of computational scientists and clinicians using standard, structured descriptions of disease-specific phenotypes. In this report, we describe progress over the past decade in three-dimensional (3D) digital imaging and shape analysis of the face, and future prospects for large-scale facial phenotyping. Illustrative examples are given throughout using a collection of 1,107 3D face images of healthy controls and individuals with a range of genetic conditions involving facial dysmorphism.