Fit prediction for filtering facepiece respirator using 3D face shape.

Interpretable facial landmark detection by multi-expert collaborative uncertainty-aware deep networks.

The role of sagittal maxillary-mandibular relationships on perceptions of facial shape esthetics: A three-dimensional morphometric analysis.

Backround: This cross-sectional study aimed to quantify the proportion of 3D tooth–face superimposition in young adults and examine age- and gender-related differences. Methods: In 98 dental students, intraoral MCI and facial scans were acquired under standardized protocols, processed in Mesh Mixer v. 3.5.474and Blender v. 4.3.2., and aligned to reference planes for superimposition. Residual tooth volume, reflecting tooth–face correspondence, was computed via Boolean subtraction. Statistical analyses were performed in IBM SPSS Statistics 25 at α = 0.05. Results: Total tooth volume (1,626,120.79 ± 210,659.56 × 103) exceeded the superimposed volume by 285,052.34 × 103 (17.53%; 95% CI: 15.84–19.22%; p < 0.001), giving a superimposition proportion of 82.6%. Positive correlations between total and remaining tooth volumes were observed overall (ρ = 0.448; p < 0.001), in females (ρ = 0.515; p < 0.001), and in participants < 21 years (ρ = 0.662; p < 0.001). Men had higher total tooth volume than women (1,706,232 ± 151,086 vs. 1,583,561 ± 225,978; p = 0.005). Conclusions: Volumetric analysis revealed high but incomplete dentofacial correspondence. Larger tooth volumes were associated with greater incongruence in females and younger participants, confirming volumetric size as a determinant of morphological congruence. Results highlight clinically meaningful superimposition, supporting the MCI as a reliable reference for restorative planning and esthetic reconstructions, and emphasize the value of 3D analysis for precise dentofacial evaluation and individualized esthetic planning.

The perception of facial beauty is shaped by a dynamic interplay of cultural, generational, and media-driven influences. Despite growing interest in aesthetic standards worldwide, there is a lack of data on facial beauty preferences within the Moroccan population. This study aims to identify the facial features considered most attractive by Moroccan individuals and to examine how these preferences vary across age, gender, and geographic region. A cross-sectional survey was conducted among 1000 Moroccan participants using digitally modified, hand-drawn anatomical portraits. Each image represented a variation within ten distinct facial units, including face shape, forehead, eyebrows, eyes, nose, cheeks, lips, chin, jawline, and neck. Participants selected their preferred option for each unit. Chi-square tests were used to analyze statistically significant associations between demographic variables and aesthetic choices (p < 0.05). The oval face shape, balanced vertical thirds, slightly prominent forehead, arched eyebrows, and positively tilted eyes were the most preferred features. Participants also favored a concave nasal dorsum with a rounded tip, moderately filled cheeks, well-defined and voluminous lips with symmetrical projection, a moderately projected chin, a well-defined jawline, and a long neck with an acute cervico-facial angle. While overarching trends were consistent, statistically significant variations were observed by age and region, underscoring Morocco's internal aesthetic diversity. Moroccan facial beauty preferences largely align with globalized aesthetic norms while preserving distinctive regional and generational nuances. These findings offer practical guidance for facial aesthetic practitioners and contribute to the broader understanding of beauty perception in North Africa. 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 .

Deep learning based face shape classification system with binary feature selection model

Background Head orientation during photography can distort facial shape, but this is rarely considered in cross-population studies focusing on facial morphology and its perception. Aim To assess how head positioning (Frankfurt Horizontal [FH] vs. natural position [NP]) affects facial shape and trait perception across two ethnic populations. Subjects and methods Facial photographs were collected from Buryats of Mongolian descent (N = 148) and Russians of European descent (N = 155). Geometric morphometrics were used to compare the facial shapes in the FH and NP positions. Independent rater groups with similar population origins, Kalmyks (N = 162) and Russians (N = 242), evaluated perceived aggressiveness and attractiveness based on these portraits. Results Head orientation significantly altered facial shape, particularly among the Buryat group. Despite the larger facial sexual dimorphism in the Buryat group, sex differences were more pronounced in FH across both samples. Correlations between facial shape and body metrics (height, Body Mass Index, and body fat mass) were stronger in FH. Ratings of male aggressiveness and female attractiveness were consistently influenced by head position across both rater populations and sexes. Conclusion Head positioning introduces systematic bias into facial shape analysis and perception. Failure to control for this factor may compromise study comparability and distort meta-analytic findings.

The goal this project is to create a face-shape classification and hairstyle recommendation system by combining Support Vector Machine (SVM) and Random Forest (RF) algorithms with Histogram of Oriented Gradients (HOG) feature extraction. This study is motivated by the growing demand for individualized appearance support, as many users find it difficult to find haircuts that complement their face features. The method first preprocesses facial photos, uses HOG to extract key geometric and texture-based features, and then uses SVM and RF models to categorize the images. For training, validation, and testing, a dataset of five different face shapes is utilized. According to experimental results, the Random Forest model has an accuracy of about 89%, while the SVM model achieves an accuracy of about 95%. These findings suggest that SVM is better suited for managing high-dimensional feature spaces generated by HOG extraction. A recommendation system that offers hairstyle recommendations based on the anticipated face shape is then integrated with the trained model. The system is useful for real-time use since it can process pictures taken with the camera or uploaded from the gallery. Overall, this study shows that integrating HOG with SVM offers a dependable basis for creating customized hairdo recommendations as well as an efficient method for face-shape classification.

The well-known topic of crisp graph planarity is contrasted with the more new and thoroughly studied field of planarity inside a fuzzy framework. In cubic fuzzy domain, cubic multisets with interval and fuzzy number to capture vagueness. Cubic fuzzy graphs (CuFGs) are structures which use cubic multisets to represent membership of vertices and edges. The interval represents a continuous process, whereas the point defines a specific process. Thus, cubic fuzzy graphs perform better than both interval valued graphs and fuzzy graphs as they indicate the level of participation of vertices and edges enabling management of uncertainty and ambiguity in interval valued fuzzy graphs (IvFG) and fuzzy graphs (FGs). The properties and characteristics of cubic fuzzy outerplanar graphs is investigated in this article, diving into a variety of fascinating elements of these topics. Cubic fuzzy graphs (CuFGs) can be created by removing individual vertices or edges from cubic fuzzy outerplanar subgraphs (CuFOSs). The study also includes examples of maximal and maximum cubic fuzzy outerplanar subgraphs obtained by removing both vertices and edges. Furthermore, the definition of cubic fuzzy dual graphs (CuFDGs), which are formed from cubic fuzzy outerplanar graphs (CuFGs) is presented and underlying relationship among these is explored. A practical application of this work is found in human face shape recognition, where cubic fuzzy graphs (CuFGs) offer a powerful framework for modeling facial geometry. It is observed that by accommodating uncertainty in facial feature positions and relationships, CuFGs enable more accurate recognition in complex biometric identification tasks.

ZBTB24 is a member of a protein family containing a Broad-Complex, Tramtrack, and Bric a Brac (BTB) domain, which functions in protein-protein interactions. ZBTB24, a transcription factor, binds its DNA targets through its C-terminal zinc finger (ZF) domain. Biallelic ZBTB24 pathogenic variants lead to the rare autosomal recessive Immunodeficiency, Centromeric instability and Facial anomalies type 2 (ICF2) syndrome. The majority of ICF2 patients carry biallelic loss-of-function variants in ZBTB24. The remaining patients harbor missense variants in the ZF domain that compromise the ability of ZBTB24 to transcriptionally activate CDCA7, the gene responsible for ICF subtype 3 syndrome. Although an ICF2 patient with compound heterozygous pathogenic variants, including a missense variant (p.Ser59Gly) in the BTB domain, has been reported, no ICF2 patients with biallelic missense variants in any ZBTB24 domains other than the zinc finger domain have been described. Similar to all subtypes of ICF syndrome, ZBTB24 pathogenic variants lead to significant DNA hypomethylation throughout the genome. Here we describe a patient with severe infections initiating during her first year of life, significant developmental delay and an abnormal facial shape, who carries a homozygous p.Val43Leu substitution in the BTB domain of ZBTB24. The patient's peripheral blood cells demonstrate whole genome DNA hypomethylation with patterns identical to those found in verified ICF2 patients. Both the p.Val43Leu and p.Ser59Gly variants cause significant ZBTB24 protein instability. Thus, we demonstrate that pathogenic missense variants in the BTB domain of ZBTB24 can functionally act as loss-of-function variants that result in ICF2 syndrome.

Previous research has shown substantial variation in sexual dimorphism of facial structure and skin color across human populations. This study investigates sexual dimorphism in both facial shape and skin color in understudied populations from the Horn of Africa - Somalis and Ethiopians - focusing on the relationship between facial morphology and pigmentation traits. Standardized frontal and profile photographs were collected from participants and analyzed using geometric morphometrics. Sexual shape dimorphism (SShD) was calculated by projecting each individual's facial shape onto a vector connecting average male and female shapes in Procrustes-aligned morphospace; higher values indicate more masculine morphology. Sexual color dimorphism (SCoD) was computed similarly, by projecting individual skin color values (from forehead and cheeks) onto a vector defined by average male-female difference in the CIE Lab* color space. Trajectory analysis and Bayesian hierarchical modelling were performed to examine associations between SShD and SCoD. Significant sexual dimorphism was detected in both facial shape and skin color across both populations. Male faces tended to be more robust and darker, while female faces were more gracile and lighter-skinned. However, despite group-level sex differences, individual-level associations between facial structure and skin pigmentation were weak or absent. Somali participants exhibited greater dimorphism in skin color compared to Ethiopians, while facial shape dimorphism remained consistent between groups. Both target populations revealed significant sexual dimorphism in both structural and pigmentation facial traits. However, in contrast to previous findings reported in certain West African populations, we found no statistically reliable association between these two traits at the individual level. This decoupling may indicate distinct evolutionary or ecological pressures shaping morphological and pigmentation traits. Our findings suggest that no universal trade-off or consistent relationship exists between morphological and pigmentation components of sexual dimorphism across human populations.

Abstract Previous studies show that people believe that a name belongs to a person if the person's face shape (FS) matches the lip shape formed when pronouncing the name. This represents a cross‐modal mapping effect between name pronunciation (NP) and FS. Considering that approximately 84.55% of Chinese people have a two‐character name, the present study specifically investigates which character of double‐character Chinese names plays a more critical role in the mapping relationship with the FS. We conducted four experiments that used disyllabic names with pronunciations involving contrasting lip shapes. The interval between the pronunciation of the name's first and final characters was set at either 0 or 1000 ms. Two presentation orders were used: presenting the name before the face image and the face image before the name. We found that the NP–FS mapping persists when the lip shapes of the initial and final syllables of a name differ, while the final syllable determinetavs the mapping. This underscores the significant role of the final syllable in driving NP–FS mapping, which supports the hypothesis that names must be completely encoded for recognition before being integrated with facial perception.

Poly-l-lactic acid (PLLA) is an injectable biostimulatory filler used for restoring facial fat volume loss. To evaluate the safety and efficacy of injectable PLLA injections for volume loss in the temples using dual plane injections. Assessments included live ratings and ratings of standardized pictures and ultrasound imaging at weeks 4, 8, 12, and 16 by a trained evaluator. At the 8-week follow-up, there was a statistically significant increase in temporal volume and improvement in skin elasticity in PLLA-treated subjects. Treatment was well-tolerated with minimal self-resolving adverse effects. Ultrasound imaging confirmed that the injection plane was free of vasculature, posing no safety risk. Repeated PLLA treatments in the temporal region elicited a global improvement in facial shape, particularly the mid and upper face region. The temple hollowness and facial laxity improved over time with no safety sequelae.

Face recognition system is the most crucial work in today's world. It plays an essential role for a variety of reasons, including 1) searching for a person, 2) matching a person who committed a crime, 3) various cosmetologically processes, and many others. Throughout all of humanity's organs, a person's face is particularly important. Human personalities can be altered by even minor changes in facial shape. As a result of genetic mutations, severe injuries, or strong medication, a human face can develop a variety of deformities, which may have the largest influence on individuals. As a result, the rapid change may damage the recognition process in the future. In the present environment, where an epidemic-like state has already materialized, grave health issues are persistently impacting the entire population. Thus, using substantial quantities of medication is solely one method of treating these medical conditions. For this reason, taking large amounts of medication may be a contributing factor to a number of facial abnormalities. In order to evaluate these human facial imperfections computer vision with human interaction was chosen for this investigation. This research examines an input image of a human's frontal face and employs a segregation approach to distinguish the deformed faces. To identify those facial images, this strategy first implemented the Viola-Jones face detection algorithm, then Fast Fourier Transformation (FFT) and Discrete Cosine Transformation analysis for feature extraction, and finally Euclidean distance measurement techniques to analyze the input faces.

Population-level morphological differences in human cranio-skeletal morphology are reflective of a multitude of genetic, epigenetic, functional and environmental factors. While such differences have been described in postnatal and adult skeletons, how early they originate during cranio-skeletal ontogeny is unknown. Here we compare early human facial morphogenesis utilizing a collection of fetal specimens of Japanese (JP) and North American - European ancestry (NAP). Utilizing 3D micro-CT images and geometric morphometric analyses, we have identified subtle, yet significant differences in the morphology of the upper and midfacial regions in fetuses in the second trimester of gestation. Notably, the orbital rims and lateral areas of the zygomatic bones in the JP specimens have more prominent superior and inferior orbital rims, with a more prominent lateral project of the zygomatic bone, compared to the NAP. These data suggest that the cranioskeletal structures display early facial patterning unique to genealogically distinct populations. Since these specimens are derived from very early developmental stages, the impacts of functional factors are expected to be minor, while inherited genetic and epigenetic programming predominate in driving the observed morphological differences.

Facial dysmorphism is a feature of many monogenic disorders and is important in diagnostics, variant interpretation and nosology. Nevertheless, comprehensively assessing the complex facial shape changes associated with specific syndromes remains challenging. Here, we present three-dimensional (3D) morphometric approaches to overcome these limitations, using Cri-du-Chat syndrome (CdCS) as a model. We analysed 3D facial images from 24 participants with CdCS, 4540 unaffected controls and five participants with rare 5p15.33-15.32 deletions, incorporating two methods to account for age- and sex-related facial variation. We quantified phenotypic variation within and between groups and explored genotype-phenotype correlations in CdCS. We identified changes in the characteristic facial features of CdCS with age and found that facial shape in CdCS differed from controls in highly consistent directions, but with varying magnitudes of effect. 5p15.33-15.32 heterozygotes had non-specific dysmorphic features that were objectively different from those in CdCS, delineating multiple critical regions for facial dysmorphism on chromosome 5p. This work explores 3D facial morphometry to complement the standard clinical assessment of facial dysmorphism. It provides insights into the genetic basis of facial shape in CdCS and highlights the potential of 3D morphometric techniques to facilitate clinical diagnostics, variant interpretation and delineation of syndrome nosology.

The face is a complex, variable structure shaped by environmental and functional adaptations. In humans, the remarkable diversity of facial shapes underpins identity and mutual recognition. The developmental process from cranial neural crest cell migration to facial prominence fusion is conserved and essential for determining facial shape. However, the molecular and cellular underpinnings are not fully understood. We reconstruct facial development in the mouse model at the single-cell level, and show that the facial mesenchyme exhibits a remarkable molecular heterogeneity predominantly driven by positional programs. We then explore the role of these spatially defined murine mesenchymal populations in the extraordinary diversity of human facial shapes. By integrating molecular and spatial coordinates with human genome-wide association studies and genes linked to abnormal human facial shapes, we link genetic variants associated with facial features to individual cell populations and transcriptional signatures. This integrative approach provides a framework for exploring evolutionary processes behind facial variation and offers new insights into congenital facial syndromes.

Recently, posters or images with designs similar to Disney Pixar have been circulating on various social media platforms, such as Instagram and TikTok. With a design that is quite adorable, it is no wonder that the image design created using artificial intelligence (AI) has attracted the attention of people. Seeing the popularity of posters via Bing Image Creator, people can try to create character descriptions according to their favourite cartoons/ animations, such as Disney, via Bing Image Creator. This research simulated the use of Bing Image Creator to teach writing an English descriptive text to be illustrated into an animation or a cartoon. The analysis shows that the Bing Image Creator AI application is currently entertaining to create/produce the images they want without having design or drawing skills. This research is descriptive qualitative research. Teachers/lecturers can ask students to visit or access Bing Image Creator at https://www.bing.com/create. Students need to have a Microsoft account, but if they do not have one, they can create an account first. They sign in using a Microsoft account. After logging in, type the character description into the column provided. They can enter character descriptions as desired. To create a Disney AI cartoon, students can add the words "Disney Pixar style" to the beginning of the character description. They can create more specific commands to turn photos into Disney cartoons. The order may include descriptions such as 1) gender, male or female. 2) character shape, such as face shape, eyes, jaw, nose, cheeks, hair, clothes, position of body parts, and others. 2) certain poses/styles and accessories used. 3) objects worn on body parts, for example, types of clothing, colors, motifs/patterns. 4) objects used or around 5) expression, facial expression, or feeling. 6) activities carried out. 7) shooting angle. 8) surrounding background. To get accurate animated image results, students can write detailed and specific descriptions. Finally, wait for the image to finish and download it. Further research still needs to be carried out to determine students' perceptions of the use of Binh Image Creator in illustrating their written character descriptions of a desired cartoon. Keywords: AI-based teaching, Bing Image Creator, teaching writing, descriptive text,

Age-related changes in facial ligaments contribute to altered facial shape and soft tissue descent. Radiofrequency (RF) has been utilized for skin rejuvenation by promoting collagen fiber contraction and synthesis through increased expression of heat shock proteins (HSPs). The primary component of ligamentous collagen fibers undergoes structural modifications with age, exhibiting increased fragmentation and a reduced collagen type I/III ratio. This study aimed to investigate whether RF irradiation alleviates senescence-related changes in facial ligaments through HSP70-mediated molecular remodeling using a UV-induced photoaging rat model. In senescent fibroblasts, RF enhanced the interaction between HSP70 and IκBα kinase (IKK)γ while reducing IκBα phosphorylation, which was associated with decreased nuclear factor-kappa B (NF-κB) activation. These RF-mediated changes were attenuated by an HSP70 inhibitor, suggesting that RF reduces NF-κB activity via HSP70 modulation. RF also suppressed expression levels of matrix metalloproteinases and SMAD7 in senescent fibroblasts. Consistent with in vitro findings, RF increased the interaction between HSP70 and IKKγ while decreasing IκBα phosphorylation and NF-κB activity in the UV-induced photoaging (senescent) facial ligaments of rat models. Furthermore, RF enhanced the collagen type I/III ratio and increased collagen fiber density within the ligaments. Scanning electron microscopy revealed that RF irradiation increased collagen fiber bundle diameter and enhanced the helical structure of those fibers. Overall, RF mitigates senescence-related changes in facial ligaments through HSP70 modulation. Considering that facial ligament laxity contributes to soft tissue descent, facial ligament-targeting approaches may promote a more youthful facial structure. RF demonstrates the possibility in reducing senescence-associated changes within facial ligaments.

Typically feminine morphological traits in women include a neotenous facial structure with large eyes, full lips, and an oval face shape, and a curvaceous body with large breasts, a narrow waist, and full hips and buttocks. Compared to men, women also show higher second-to-fourth finger (2D:4D) ratios as well as less muscle mass, lower physical strength, and a higher voice pitch. Due to a putative association with oestrogen levels, feminine traits are often claimed to cue women’s reproductive potential. However, the evidence for this is scarce and inconsistent, typically measuring proxies rather than actual fertility outcomes. Here, we report a systematic review of direct fertility measures as a function of morphological traits in women, including breast size, waist-to-hip ratio (WHR), voice pitch, strength, and 2D:4D; no articles were found measuring facial femininity. The review included 19 articles comprising 68 effect sizes (31 samples from 16 countries; total N = 125,062). Our review showed that a less feminine WHR may cue past fertility, and a more feminine 2D:4D may be, at best, weakly associated with fertility. Overall, we conclude that the current evidence base is too weak to support the claim that women’s feminine morphological traits are associated with reproductive potential.