Three-dimensional photography for the evaluation of facial profiles in obstructive sleep apnoea

Correlation analysis of airway-facial phenotype in Crouzon syndrome by geometric morphometrics: A promising method for non-radiation airway evaluation

AIM

This study aimed to verify the correlation of the airway-facial phenotype and visualize the morphological variation in Crouzon syndrome patients. Additionally, to develop a non-radiation methodology for airway assessments.

METHOD

In this study, 22 patients diagnosed with Crouzon syndrome (Age: 7.80 ± 5.63 years; Gender distribution: 11 females and 11 males) were analysed. The soft tissue surface and airway were three-dimensionally reconstructed, and the entire facial phenotype was topologized and converted into spatial coordinates. Geometric morphometrics was employed to verify the correlation and visualize dynamic phenotypic variation associated with airway volume. A total of 276 linear variables were automatically derived from 24 anatomical landmarks, and principal component analysis (PCA) identified the 20 most significant parameters for airway evaluation. Correlation analyses between parameters and airway volume were performed. Then, patients were classified into three groups based on airway volume, and the differences among the groups were compared for evaluating the differentiating effectiveness of parameters.

RESULTS

The facial phenotype was strongly correlated with the airway (coefficient: 0.758). Morphological variation was characterized by (i) mandibular protrusion and anticlockwise rotation; (ii) midface retrusion; (iii) supraorbital frontward and (iv) lengthening of the facial height. All the anthropometric parameters were strongly associated with the airway, and the differences among the groups were statistically significant.

CONCLUSION

This study confirmed the strong correlation between facial phenotype and airway parameters in Crouzon syndrome patients. Despite the development of the airway, pathological midface retrusion was still aggravated, suggesting that surgical intervention was inevitable. Three-dimensional facial anthropometry has potential as a non-radiation examination for airway evaluation.

Predictive factors of therapeutic response according to craniofacial skeletal biotype in patients with sleep apnea syndrome using mandibular advancement devices: a pilot study

OBJECTIVES

To evaluate the influence of facial biotype in the therapeutic effect of mandibular advancement devices (MADs) according to polysomnographic records in patients diagnosed with sleep apnea-hypopnea syndrome (SAHS).

MATERIALS AND METHODS

A total of 46 patients were recruited. Patients were classified according to facial biotype (mesofacial, brachyfacial, or dolichofacial). The quantitative variables were described as the arithmetic mean and standard deviation or the median and interquartile range. Hypothesis tests used were Pearson's chi-square, paired-sample Student's t- test, the Wilcoxon test, one-way analysis of variance, Kruskal-Wallis test, and Mann-Whitney U-test. P < .05 was considered statistically significant.

RESULTS

A total of 46 patients were categorized into three facial biotype subgroups with no significant differences among them in age, body mass index, neck circumference, and sex. The respiratory disturbance index (RDI) results were as follows: brachyfacial patients had a reduction to 15 events/h (P < .001), the mesofacial patients had a reduction to 14 events/h (P < .001), and the dolichofacial patients did not show a significant reduction. The oxygen desaturation index (ODI) results were as follows: brachyfacial patients had a reduction in ODI episodes to 45 episodes/h (P = .001), mesofacial patients had a reduction to 18 episodes/h (P = .001). In the brachyfacial group, the number of awakenings with MAD therapy was reduced to 23 events/h (P = .003), while, in the mesofacial group, it was reduced to 37 episodes/h (P = .012).

CONCLUSIONS

The facial biotype influences the effectiveness of MAD therapy and is considered a good predictive factor.

The relationship between eye canting and vertical craniofacial skeletal asymmetry in adult patients with dento-maxillofacial deformities

BACKGROUND

Whether eye canting in patients with asymmetric dento-maxillofacial deformities is the result of anatomical asymmetry or a compensatory head position remains controversial.

OBJECTIVE

This cross-sectional study aims to verify whether eye canting is correlated with craniofacial skeletal asymmetry.

METHODS & MATERIALS

This study was performed by measuring the computed tomographic scans of 223 patients with dento-maxillofacial deformities in Mimics 21.0 software. First grouping was determined based on the intersection angle between the line passing through bilateral lateral canthus point and Frankfurt horizontal plane, and final grouping was based on measurements of the pregroups.

RESULTS

The patients were finally categorized into three groups: symmetry group (n = 163), asymmetry subgroups 1 (n = 33) and asymmetry subgroups 2 (n = 27). The results of multiple linear regression and comparisons among groups suggests the presence of orbital skeletal asymmetry in patients with eye canting and the eye canting is partly increment dependent of orbital skeletal asymmetry when using bilateral ears as the reference. The result also reveals that there is a greater angle between the line through bilateral ocular landmarks and the line through the bilateral ear landmarks in patients with ocular canting compared to patient without eye canting.

CONCLUSION

Patients with asymmetric dento-maxillofacial deformities and with eye canting have vertical asymmetry of the orbital and cranial skeletal landmarks. These symmetry differences between the eyes and ears may affect the overall craniofacial symmetry after orthognathic surgery.

Deep learning technique to detect craniofacial anatomical abnormalities concentrated on middle and anterior of face in patients with sleep apnea

OBJECTIVES

The aim of this study is to propose a deep learning-based model using craniofacial photographs for automatic obstructive sleep apnea (OSA) detection and to perform design explainability tests to investigate important craniofacial regions as well as the reliability of the method.

METHODS

Five hundred and thirty participants with suspected OSA are subjected to polysomnography. Front and profile craniofacial photographs are captured and randomly segregated into training, validation, and test sets for model development and evaluation. Photographic occlusion tests and visual observations are performed to determine regions at risk of OSA. The number of positive regions in each participant is identified and their associations with OSA is assessed.

RESULTS

The model using craniofacial photographs alone yields an accuracy of 0.884 and an area under the receiver operating characteristic curve of 0.881 (95% confidence interval, 0.839-0.922). Using the cutoff point with the maximum sum of sensitivity and specificity, the model exhibits a sensitivity of 0.905 and a specificity of 0.941. The bilateral eyes, nose, mouth and chin, pre-auricular area, and ears contribute the most to disease detection. When photographs that increase the weights of these regions are used, the performance of the model improved. Additionally, different severities of OSA become more prevalent as the number of positive craniofacial regions increases.

CONCLUSIONS

The results suggest that the deep learning-based model can extract meaningful features that are primarily concentrated in the middle and anterior regions of the face.

Self-helped detection of obstructive sleep apnea based on automated facial recognition and machine learning

PURPOSE

The diagnosis of obstructive sleep apnea (OSA) relies on time-consuming and complicated procedures which are not always readily available and may delay diagnosis. With the widespread use of artificial intelligence, we presumed that the combination of simple clinical information and imaging recognition based on facial photos may be a useful tool to screen for OSA.

METHODS

We recruited consecutive subjects suspected of OSA who had received sleep examination and photographing. Sixty-eight points from 2-dimensional facial photos were labelled by automated identification. An optimized model with facial features and basic clinical information was established and tenfold cross-validation was performed. Area under the receiver operating characteristic curve (AUC) indicated the model's performance using sleep monitoring as the reference standard.

RESULTS

A total of 653 subjects (77.2% males, 55.3% OSA) were analyzed. CATBOOST was the most suitable algorithm for OSA classification with a sensitivity, specificity, accuracy, and AUC of 0.75, 0.66, 0.71, and 0.76 respectively (P < 0.05), which was better than STOP-Bang questionnaire, NoSAS scores, and Epworth scale. Witnessed apnea by sleep partner was the most powerful variable, followed by body mass index, neck circumference, facial parameters, and hypertension. The model's performance became more robust with a sensitivity of 0.94, for patients with frequent supine sleep apnea.

CONCLUSION

The findings suggest that craniofacial features extracted from 2-dimensional frontal photos, especially in the mandibular segment, have the potential to become predictors of OSA in the Chinese population. Machine learning-derived automatic recognition may facilitate the self-help screening for OSA in a quick, radiation-free, and repeatable manner.

Characterization of craniofacial-based clinical phenotypes in children with suspected obstructive sleep apnea

STUDY OBJECTIVES

We conducted this study to identify phenotypes of obstructive sleep apnea (OSA) in children based on lifestyle, sleep habits, age, obesity, sex, soft tissue facial features, and specific craniofacial abnormalities.

METHODS

Seventy-three children with symptoms of pediatric OSA who underwent overnight observed polysomnography participated in this study. Soft tissue facial features were assessed using a 3-dimensional stereophotogrammetric system. Craniofacial abnormalities were evaluated based on the most common facial features associated with orthodontic treatment needs. Data regarding lifestyle, sleep habits, age, obesity, and sex were also collected. To identify phenotypes of OSA, a sequential analysis was then performed on categories of variables using fuzzy clustering with medoids.

RESULTS

Craniofacial abnormalities and soft tissue facial features defined clusters. Three clusters were identified. Cluster 1 comprised a group of younger children (5.9 ± 3.8 years) without obesity, without craniofacial abnormalities, and with smaller soft tissue facial features dimensions. Cluster 2 comprised a group of older children (9.6 ± 3.9 years) without obesity and with larger mandibular dimensions and mildly arched palates (71.4%). Cluster 3 comprised a group of older children (9.2 ± 3.9 years) with obesity and a history of health issues (68.4%), excessive lower facial height (63.2%), and midface deficiency (73.7%). No differences were observed across clusters regarding sleep features. A moderate severity of obstructive and mixed respiratory events was observed in all 3 clusters.

CONCLUSIONS

The study results did not identify distinct phenotypes of pediatric OSA based on soft tissue facial features or craniofacial abnormalities alone. Age and body mass index likely modify the effect of soft tissue facial features and craniofacial abnormalities as risk factors for OSA in children.

CITATION

Fernandes Fagundes NC, Loliencar P, MacLean JE, Flores-Mir C, Heo G. Characterization of craniofacial-based clinical phenotypes in children with suspected obstructive sleep apnea. J Clin Sleep Med. 2023;19(11):1857-1865.

Geometric morphometrics and machine learning from three-dimensional facial scans for difficult mask ventilation prediction

Background

Unanticipated difficult mask ventilation (DMV) is a potentially life-threatening event in anesthesia. Nevertheless, predicting DMV currently remains a challenge. This study aimed to verify whether three dimensional (3D) facial scans could predict DMV in patients scheduled for general anesthesia.

Methods

The 3D facial scans were taken on 669 adult patients scheduled for elective surgery under general anesthesia. Clinical variables currently used as predictors of DMV were also collected. The DMV was defined as the inability to provide adequate and stable ventilation. Spatially dense landmarks were digitized on 3D scans to describe sufficient details for facial features and then processed by 3D geometric morphometrics. Ten different machine learning (ML) algorithms, varying from simple to more advanced, were introduced. The performance of ML models for DMV prediction was compared with that of the DIFFMASK score. The area under the receiver operating characteristic curves (AUC) with its 95% confidence interval (95% CI) as well as the specificity and sensitivity were used to evaluate the predictive value of the model.

Results

The incidence of DMV was 35/669 (5.23%). The logistic regression (LR) model performed best among the 10 ML models. The AUC of the LR model was 0.825 (95% CI, 0.765-0.885). The sensitivity and specificity of the model were 0.829 (95% CI, 0.629-0.914) and 0.733 (95% CI, 0.532-0.819), respectively. The LR model demonstrated better predictive performance than the DIFFMASK score, which obtained an AUC of 0.785 (95% CI, 0.710-0.860) and a sensitivity of 0.686 (95% CI, 0.578-0.847). Notably, we identified a significant morphological difference in the mandibular region between the DMV group and the easy mask ventilation group.

Conclusion

Our study indicated a distinct morphological difference in the mandibular region between the DMV group and the easy mask ventilation group. 3D geometric morphometrics with ML could be a rapid, efficient, and non-invasive tool for DMV prediction to improve anesthesia safety.

Accuracy of three-dimensional optical devices for facial soft-tissue measurement in clinical practice of stomatology: A PRISMA systematic review

BACKGROUND

The accuracy of 3-dimensional (3D) optical devices for facial soft-tissue measurement is essential to the success of clinical treatment in stomatology. The aim of the present systematic review was to summarize the accuracy of 3D optical devices used for facial soft-tissue assessment in stomatology.

METHODS

An extensive systematic literature search was performed in the PubMed/MEDLINE, Embase, Scopus and Cochrane Library databases for studies published in the English language up to May 2022 in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Peer-reviewed journal articles evaluating the facial soft-tissue morphology by 3D optical devices were included. The risk of bias was performed using the Quality Assessment Tool for Diagnostic Accuracy Studies-2 guidelines by the 2 reviewers. The potential publication bias was analyzed using the Review Manager software.

RESULTS

The query returned 1853 results. A total of 38 studies were included in this review. Articles were categorized based on the principle of devices: laser-based scanning, structured-light scanning, stereophotogrammetry and red, green, blue-depth camera.

CONCLUSION

Overall, the 3D optical devices demonstrated excellent accuracy and reliability for facial soft-tissue measurement in stomatology. red, green, blue-depth camera can collect accurate static and dynamic 3D facial scans with low cost and high measurement accuracy. Practical needs and availability of resources should be considered when these devices are used in clinical settings.

Classification of Alar Dynamic Aesthetic in an Asian Female Population: Experts or Automatic Algorithms?

AIM

To provide referenced classifications of alar dynamic aesthetics from both subjective and objective perspectives for determining proper surgical strategies in alarplasty.

METHODS

A total of 150 healthy Asian female participants were instructed to perform two standardized facial movements including a resting pose and a maximum smile while taking care not to show their teeth. The participants were recorded using a dynamic three-dimensional surface imaging system. Frames depicting the resting position and the alar maximum enlargement during the smile were exported separately for anthropometric analysis and classification. The alar dynamic aesthetic was assessed through measurement of the anthropomorphic changes comparing the resting and maximum smile statuses and then transformed into quantitative analysis through the algorithm [Formula: see text]. Subjective classification and evaluation of the subject cosmetic deficiencies and proposals for therapeutic interventions to improve the subjects' alar dynamic aesthetic were performed by three senior plastic surgeons through visualization of the resting and smiling images. The surgeons were asked to divide and classify the subjects into three groups (Class I, Class II and Class III) according to the surgeons' perceptions of degree of the subjects' deficiencies in alar dynamic aesthetic. The more deficiency there was in the aesthetic, the higher the class that the subject was assigned into. The surgeons were presented with the full set of images of the patients on two separate occasions each three months apart, to assess interobserver reliability. Clustering analysis, which is based on machine learning, was applied for objective classification of the images.

RESULTS

According to the senior plastic surgeon experts' subjective classification, the subjects' alar flaring mobility was judged as follows: Class I (6.78 ± 3.84%), Class II (10.35 ± 4.18%), and Class III (18.68 ± 4.15%), while alar base mobility was judged as Class I (12.71 ± 7.57%), Class II (20.06 ± 10.06%), and Class III (30.86 ± 13.20%). By clustering analysis, alar flaring mobility was determined to be Class I (7.01 ± 3.51%), Class II (11.18 ± 4.76%), and Class III (12.72 ± 5.66%), while alar base mobility was Class I (9.07 ± 4.23%), Class II (21.88 ± 4.25%), and Class III (38.59 ± 7.08%). No statistical significance was found in the distribution and assignment of classes between the two methodologies.

CONCLUSION

Classifications of alar dynamic aesthetics could arouse attention to facial dynamic aesthetics and provide referenced quantitative parameters for plastic surgeons to determine appropriate treatments for alarplasty. For patients with Class I mobility, treatments are not recommended, while minimally invasive treatments can be deemed to be optional for patients with Class II alar mobility to potentially improve alar dynamic aesthetics. For patients with Class III alar mobility, surgical treatments are strongly recommended as options. Combing subjective classification with automated algorithms can provide a novel perspective and improve reliability for facial aesthetic classification analysis.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors   www.springer.com/00266 .

Workflow to develop 3D designed personalized neonatal CPAP masks using iPhone structured light facial scanning

Background

Continuous positive airway pressure (CPAP) is a common mode of respiratory support used in neonatal intensive care units. In preterm infants, nasal CPAP (nCPAP) therapy is often delivered via soft, biocompatible nasal mask suitable for long-term direct skin contact and held firmly against the face. Limited sizes of nCPAP mask contribute to mal-fitting related complications and adverse outcomes in this fragile population. We hypothesized that custom-fit nCPAP masks will improve the fit with less skin pressure and strap tension improving efficacy and reducing complications associated with nCPAP therapy in neonates.

Methods

After IRB approval and informed consent, we evaluated several methods to develop 3D facial models to test custom 3D nCPAP masks. These methods included camera-based photogrammetry, laser scanning and structured light scanning using a Bellus3D Face Camera Pro and iPhone X running either Bellus3D FaceApp for iPhone, or Heges application. This data was used to provide accurate 3D neonatal facial models. Using CAD software nCPAP inserts were designed to be placed between proprietary nCPAP mask and the model infant’s face. The resulted 3D designed nCPAP mask was form fitted to the model face. Subsequently, nCPAP masks were connected to a ventilator to provide CPAP and calibrated pressure sensors and co-linear tension sensors were placed to measures skin pressure and nCPAP mask strap tension.

Results

Photogrammetry and laser scanning were not suited to the neonatal face. However, structured light scanning techniques produced accurate 3D neonatal facial models. Individualized nCPAP mask inserts manufactured using 3D printed molds and silicon injection were effective at decreasing surface pressure and mask strap pressure in some cases by more than 50% compared to CPAP masks without inserts.

Conclusions

We found that readily available structured light scanning devices such as the iPhone X are a low cost, safe, rapid, and accurate tool to develop accurate models of preterm infant facial topography. Structured light scanning developed 3D nCPAP inserts applied to commercially available CPAP masks significantly reduced skin pressure and strap tension at clinically relevant CPAP pressures when utilized on model neonatal faces. This workflow maybe useful at producing individualized nCPAP masks for neonates reducing complications due to misfit.

Predicting the presence and severity of obstructive sleep apnea based on mandibular measurements using quantitative analysis of facial profiles via three-dimensional photogrammetry

BACKGROUND

In obstructive sleep apnea (OSA), the upper airway is obstructed during sleep due to obesity and/or posterior collapse of the tongue root. Maxillofacial morphological abnormalities increase the risk of OSA in the Asian population. This study sought to elucidate whether three-dimensional (3D) photogrammetry measurements correlate with the severity of OSA irrespective of sex and degree of obesity.

METHODS

A prospective pilot study was performed, in which 37 consecutive adult patients (M/F = 28/9) underwent polysomnography and 3D photogrammetry in the supine position for the diagnosis of OSA. Measurements obtained from 3D photogrammetry included mandibular width (Mw), mandibular length (Ml), mandibular depth (Md), mandibular width-length angle (Mwla), and mandibular area (Ma). The effects of sex and body mass index (BMI) on the measurements and their association with the apnea-hypopnea index (AHI) were statistically analyzed. The inter-rater reliability of the measurements was evaluated using intraclass correlation coefficients (ICC).

RESULTS

Mwla (R = 0.73, p < 0.01), Mw (R = 0.39, p < 0.05), and Md (R = -0.34, p < 0.05) were significantly correlated with the severity of OSA. On multivariate analysis, Mwla (p < 0.01) and Md (p < 0.05) remained independent factors for AHI after adjusting for sex, age, BMI, and neck circumference. In addition, diagnosability analysis revealed that Mwla was useful for identifying the presence of OSA (AHI ≥5) (cutoff: 78.6°, sensitivity: 0.938, specificity: 0.800, area under the curve: 0.931). The ICC was >0.9, showing high reliability.

CONCLUSIONS

This study suggests that Mwla measured using 3D photogrammetry can predict the presence of OSA and correlates with the severity of OSA, independent of obesity and sex.

Predicting sleep apnea from three-dimensional face photography

STUDY OBJECTIVES

Craniofacial anatomy is recognized as an important predisposing factor in the pathogenesis of obstructive sleep apnea (OSA). This study used three-dimensional (3D) facial surface analysis of linear and geodesic (shortest line between points over a curved surface) distances to determine the combination of measurements that best predicts presence and severity of OSA.

METHODS

3D face photographs were obtained in 100 adults without OSA (apnea-hypopnea index [AHI] < 5 events/h), 100 with mild OSA (AHI 5 to < 15 events/h), 100 with moderate OSA (AHI 15 to < 30 events/h), and 100 with severe OSA (AHI ≥ 30 events/h). Measurements of linear distances and angles, and geodesic distances were obtained between 24 anatomical landmarks from the 3D photographs. The accuracy with which different combinations of measurements could classify an individual as having OSA or not was assessed using linear discriminant analyses and receiver operating characteristic analyses. These analyses were repeated using different AHI thresholds to define presence of OSA.

RESULTS

Relative to linear measurements, geodesic measurements of craniofacial anatomy improved the ability to identify individuals with and without OSA (classification accuracy 86% and 89% respectively, P < .01). A maximum classification accuracy of 91% was achieved when linear and geodesic measurements were combined into a single predictive algorithm. Accuracy decreased when using AHI thresholds ≥ 10 events/h and ≥ 15 events/h to define OSA although greatest accuracy was always achieved using a combination of linear and geodesic distances.

CONCLUSIONS

This study suggests that 3D photographs of the face have predictive value for OSA and that geodesic measurements enhance this capacity.

Development of a smart-fit system for CPAP interface selection

Continuous Positive Airway Pressure (CPAP) therapy is commonly prescribed for longstanding, acute cases of Obstructive Sleep Apnea (OSA) during which patients must wear a tight-fitting breathing mask overnight for the duration of the treatment. Because this condition frequently leads to the permanent use of CPAP masks, interface selection is a crucial factor influencing the treatment quality and effectiveness. Masks/interface selection is normally performed on a trial an error basis with clinicians informing their selection based on OSA-related factors with basic fitting feedback from patients. However, it is not uncommon for patients to abandon the treatment or request additional consultations due to ill-fitting CPAP mask with the main sources of discomfort being perceived air leakage and mask/strap overtightening leading to skin damage. This work introduces a novel system (Smart-Fit), for CPAP interface selection using advanced digital technologies, such as Reverse Engineering and Computational Modeling (Finite Element Analysis) which are paired to evaluate and determine the best fitting interface for each clinical case. The model simplifies the number of 3D facial landmarks to 12 and established that a 2 mm scan resolution is enough for accurate scans. The Von Mises stress map in ANSYS serves as an indicator of potential high-pressure areas, triggering the need for a chance of mask size. Current results indicate the Smart Fit System can enable a "best fit CPAP interface" to be selected considering individual's physical characteristics and existing CPAP interface configurations. The development of the Smart Fit System is an evolution compared to traditional CPAP interface selection approach, which optimizes the CPAP interface selection process.

[What prospects for the sleep apnea syndrome and connected health?]

Connected health is a growing field and can be viewed from different perspectives, particularly in sleep apnea syndrome. The purpose of this review is to show how all these aspects of connected health are already used in the management of sleep apnea syndrome (SAS) and its comorbidities. First, it can give patients a better understanding and a better assessment of their health. It also facilitates their healthcare by allowing them a greater role in their care pathway. For healthcare providers, connected health tools make it possible to set up new procedures for diagnosing and monitoring ambulatory patients, and for the making of joint decisions by health professionals and patients. Finally, for researchers, e-health generates massive amounts of data, thus facilitating the acquisition of knowledge in real life situations and the development of new methodologies for clinical studies that are faster, less expensive and just as reliable. All these considerations are already applicable in the field of sleep apnea, both for proposed treatments and for comorbidities management and for the patient's involvement in his/her care pathway.