Three-dimensional head shape quantification for infants with and without deformational plagiocephaly

Quantifying dysmorphologies of the neurocranium using artificial neural networks

BACKGROUND

Craniosynostosis, a congenital condition characterized by the premature fusion of cranial sutures, necessitates objective methods for evaluating cranial morphology to enhance patient treatment. Current subjective assessments often lead to inconsistent outcomes. This study introduces a novel, quantitative approach to classify craniosynostosis and measure its severity.

METHODS

An artificial neural network was trained to classify normocephalic, trigonocephalic, and scaphocephalic head shapes based on a publicly available dataset of synthetic 3D head models. Each 3D model was converted into a low-dimensional shape representation based on the distribution of normal vectors, which served as the input for the neural network, ensuring complete patient anonymity and invariance to geometric size and orientation. Explainable AI methods were utilized to highlight significant features when making predictions. Additionally, the Feature Prominence (FP) score was introduced, a novel metric that captures the prominence of distinct shape characteristics associated with a given class. Its relationship with clinical severity scores was examined using the Spearman Rank Correlation Coefficient.

RESULTS

The final model achieved excellent test accuracy in classifying the different cranial shapes from their low-dimensional representation. Attention maps indicated that the network's attention was predominantly directed toward the parietal and temporal regions, as well as toward the region signifying vertex depression in scaphocephaly. In trigonocephaly, features around the temples were most pronounced. The FP score showed a strong positive monotonic relationship with clinical severity scores in both scaphocephalic (ρ = 0.83, p < 0.001) and trigonocephalic (ρ = 0.64, p < 0.001) models. Visual assessments further confirmed that as FP values rose, phenotypic severity became increasingly evident.

CONCLUSION

This study presents an innovative and accessible AI-based method for quantifying cranial shape that mitigates the need for adjustments due to age-specific size variations or differences in the spatial orientation of the 3D images, while ensuring complete patient privacy. The proposed FP score strongly correlates with clinical severity scores and has the potential to aid in clinical decision-making and facilitate multi-center collaborations. Future work will focus on validating the model with larger patient datasets and exploring the potential of the FP score for broader applications. The publicly available source code facilitates easy implementation, aiming to advance craniofacial care and research.

Delayed Presentation of Unicoronal Craniosynostosis-Masked by Ipsilateral Posterior Deformational Plagiocephaly

Despite a greater awareness of the differential diagnosis of head shape abnormalities among pediatricians, the effect of deformational forces on calvarial morphology can complicate the diagnosis of craniosynostosis. In this report, we describe 2 patients diagnosed with unicoronal craniosynostosis (UCS) in a delayed fashion due to the presence of concomitant posterior deformational plagiocephaly (PDP). In both cases, the severity of each patients' PDP obscured changes typically associated with UCS. This unique presentation underscores the importance of having a high index of suspicion for possible premature suture fusion despite the presence of concomitant PDP.

Novel Screening and Monitoring Techniques for Deformational Plagiocephaly: A Systematic Review

This article summarizes the current state of diagnostic modalities for infant craniofacial deformities and highlights capable diagnostic tools available currently to pediatricians.

Automatic detection and monitoring of abnormal skull shape in children with deformational plagiocephaly using deep learning

Craniofacial anomaly including deformational plagiocephaly as a result of deformities in head and facial bones evolution is a serious health problem in newbies. The impact of such condition on the affected infants is profound from both medical and social viewpoint. Indeed, timely diagnosing through different medical examinations like anthropometric measurements of the skull or even Computer Tomography (CT) image modality followed by a periodical screening and monitoring plays a vital role in treatment phase. In this paper, a classification model for detecting and monitoring deformational plagiocephaly in affected infants is presented. The presented model is based on a deep learning network architecture. The given model achieves high accuracy of 99.01% with other classification parameters. The input to the model are the images captured by commonly used smartphone cameras which waives the requirement to sophisticated medical imaging modalities. The method is deployed into a mobile application which enables the parents/caregivers and non-clinical experts to monitor and report the treatment progress at home.

Demographics of Positional Plagiocephaly and Brachycephaly; Risk Factors and Treatment

ABSTRACT

In the last 3 decades, the incidence of positional cranial deformations in infants, such as positional plagiocephaly and positional brachycephaly, has increased. The deviating shape often causes parental concern for the later psychosocial wellbeing of the child. Treatment options are nonoperative, varying from positional change to helmet therapy, of which the effect has often been debated. Multiple risk factors have been associated with an increased risk on the development of these deformations. The goal of this study was to assess the impact of known risk factors on the type and severity of resulting positional cranial deformation.Parents were asked to fill out a questionnaire regarding the presence of risk factors, such as gender, age, pregnancy duration, method of delivery and reasons for atypical deliveries, breech position, birth weight, developmental status, positional preference, family history, number of siblings, and torticollis presence. Treatment methods were documented and plagiocephalometry was used to measure the cranial proportions. All children were invited to participate in short-term follow-up. The significance of the risk factors and the effect of different kinds of therapy are discussed. Although no significant correlation was found between severity and risk factors, some risk factors could be correlated with the kind of positional cranial deformation. Of the different therapies, helmet therapy had a significant impact on the reduction of positional plagiocephaly.

The Effect of Cranial Shape on Esthetic Self-Worth in Bald Men

ABSTRACT

During the last 3 decades, the incidence of positional plagiocephaly and brachycephaly has increased. Treatment options are conservative and can include physiotherapy and molding helmet therapy. The decision to start therapy is based on patient history and subjective assessment of cranial shape by the physician and the parents. To determine the impact of these deformations on esthetic self-worth, as well as cranial shape, this study was conducted on a population of adult bald men.Participants were asked to fill in a questionnaire about the amount of comments received on cranial shape, as well as grade their general esthetic and grade craniofacial features using a visual analogue scale. The cranial proportions and symmetry were measured using plagiocephalometry.50 adult bald men were included in this study. The prevalence of nonsynostotic plagiocephaly was 12%; the prevalence of nonsynostotic brachycephaly was 0%. No significant correlation was found between plagiocephaly or brachycephaly, cranial circumference, width or length and amount of comments on cranial shape, cosmetic assessment of appearance or cosmetic assessment of cranial shape. On average, teeth and eyes were given the highest cosmetic value, eyebrows and chin were given the lowest scores.Cranial shape does not have a significant impact on the amount of comments received on cranial shape, nor the cosmetic assessment of either overall appearance or cranial shape in bald men. For bald men, eyes and teeth are more important craniofacial features in cosmetic assessment. There is no clear esthetic need to treat deviating cranial shape with helmet therapy in male patients.

Molding Helmet Therapy for Deformational Brachycephaly

BACKGROUND

Molding helmet therapy is used for the treatment of infants with deformational brachycephaly (DB). There is a lack of rigorous outcome measures of helmet therapy in patients with DB using 3-dimensional (3D) imaging, with most reports rely on either subjective or 2-dimensional analyses. Furthermore, the longitudinal assessment of head shape improvement over the course of helmet therapy has not been documented. Our goal was to assess the outcome of molding helmet therapy using 3D surface imaging, and to document the pace of improvement during treatment.

METHODS

The head shape of 18 infants with DB who underwent orthotic molding helmet therapy was assessed. The 3D scans were obtained before treatment, during treatment, and at the end of treatment. First, we applied shape analysis techniques based on template deformation to obtain average (composite) heads of the 18 patients at the 3 time points of treatment (pretreatment, during, and posttreatment). In addition, we used 3D curvature analysis to quantify the degree of flatness at the same time points.

RESULTS

Molding helmet therapy started at 6.7 ± 0.9 months of age and lasted for 4.3 ± 0.8 months. The overall difference in the occipital contour between pretreatment and end of treatment was 6.3 ± 1.7 mm. Curvature analysis revealed that 15% of the back of the head had prehelmet marked flatness (mean curvature <5/m), which decreased to 9% at 2.5 months into treatment and 7% at the end of treatment.

CONCLUSION

Over 65% of the head shape improvement occurred during the 2.5 months of molding helmet therapy.

Head shape at age 36 months among children with and without a history of positional skull deformation

OBJECTIVE In this study, the authors examined head shape through age 36 months for children with and without a history of positional plagiocephaly and/or brachycephaly (PPB). METHODS Infants with PPB (cases) were identified through a craniofacial clinic at the time of diagnosis. Infants without diagnosed PPB were identified through a participant registry. Clinician ratings of 3D cranial images were used to confirm the presence or absence of PPB. The cohort included 235 case infants (diagnosed PPB, confirmed with 3D imaging), 167 unaffected controls (no diagnosed PPB, no deformation detected), and 70 affected controls (no diagnosed PPB, discernible skull deformation). Participants were seen in infancy (age 7 months, on average) and again at ages 18 and 36 months. At each visit, automated 3D measures of skull deformation quantified posterior flattening and generated an absolute asymmetry score. The authors also used automated 2D measures to approximate overall asymmetry (approximate oblique cranial length ratio) and calculate the cephalic index. They used linear regression to compare cases to unaffected controls and to compare affected versus unaffected controls on all measures. They also calculated the proportion of children in each group with "persistent PPB," defined as one or more head shape measures above the 95th percentile relative to unaffected controls at 36 months. RESULTS Head shape became more rounded and symmetric for children with and without PPB, particularly between infancy and age 18 months. However, children with PPB continued to show greater skull deformation and asymmetry than unaffected controls at age 36 months. These differences were large in magnitude, ranging from 1 to 2 standard deviations (SDs), and in most (85.6%) of the cases, there was evidence of persistent PPB at 36 months. Similarly, although differences were more modest (i.e., 0.26-0.94 SD), affected controls continued to exhibit skull deformation on most measures relative to unaffected controls and approximately 30% had persistent PPB. Within the case group, head shape at 36 months was similar for untreated patients with PPB and for those who received helmet treatment and for patients with and without a history of torticollis. CONCLUSIONS Although head shape continues to improve, children with a history of skull deformation in infancy continue to exhibit measureable cranial flattening and asymmetry through age 36 months.

Accuracy of measurements used to quantify cranial asymmetry in deformational plagiocephaly

OBJECTIVE

Various measurements are used to quantify cranial asymmetry in deformational plagiocephaly (DP), but studies validating cut-off values and comparing the accuracy of such measurements are lacking. In this study, we compared the accuracy of four different measurements in classifying children with and without DP diagnosed by visual assessment, and sought to determine their optimal cut-off values.

STUDY DESIGN

Two experts rated 407 3D craniofacial images of children aged between 3 and 36 months old using the Argenta classification. We then measured the following asymmetry-related variables from the images: Oblique Cranial Length Ratio (OCLR), Diagonal Difference (DD), Posterior Cranial Asymmetry Index (PCAI), and weighted Asymmetry Score (wAS). We created receiver operating characteristic curves to evaluate the accuracy of these variables.

RESULTS

All variables performed well, but OCLR consistently provided the best discrimination in terms of area under the curve values. Subject's age had no clear effect on the cut-off values for OCLR, PCAI, and wAS; however, the cut-off for DD increased monotonically with age. When subjects with discrepant expert ratings were excluded, the optimal cut-off values for DP (Argenta class ≥ 1) across all age-groups were 104.0% for OCLR (83% sensitivity, 97% specificity), 10.5% for PCAI (90% sensitivity, 90% specificity), and 24.5 for wAS (88% sensitivity, 90% specificity).

CONCLUSION

We recommend using OCLR as the primary measurement, although PCAI and wAS may also be useful in monitoring cranial asymmetry. The threshold of relative asymmetry required for a deformation to appear clinically significant is not affected by the child's age, and DD has no additional utility in monitoring DP compared to using only OCLR.

Analyzing infant head flatness and asymmetry using kernel density estimation of directional surface data from a craniofacial 3D model

Infant skull deformation is analyzed using the distribution of head normal vector directions computed from a 3D image. Severity of flatness and asymmetry are quantified by functionals of the kernel estimate of the normal vector direction density. Using image data from 99 infants and clinical deformation ratings made by experts, our approach is compared with some recently suggested methods. The results show that the proposed method performs competitively. Copyright © 2016 John Wiley & Sons, Ltd.

Evidence for Use of Frontozygomaticus and Contralateral Eurion as Hand-Caliper Landmarks for Assessment of Deformational Plagiocephaly

The severity of deformational plagiocephaly is frequently measured by cranial vault asymmetry (CVA). Cranial vault asymmetry is a simple linear proxy for the three-dimensional deformity. Different anthropometric landmarks have been suggested as the endpoints of the cross-cranial diagonals that determine CVA. One promising albeit counterintuitive set of digital landmarks is the frontozygomaticus and contralateral eurion. The validity of nondigital caliper measures associated with the diagonals defined by the frontozygomaticus and contralateral eurions has not been tested. In this study, the authors compare caliper measures against stereophotogrammetric measures that have been documented to correlate strongly with overall skull asymmetry.Cranial vault asymmetry was assessed by direct anthropometry with 2 different measures on 36 patients. Frontozygomaticus and contralateral eurion (fz-eu) and 30 degrees off the anteroposterior diameter (30offAP). Three-dimensional photographs were obtained and also used to quantify CVA in these subjects; digital measures of fz-eu and an overall measure of plagiocephaly (Global) were calculated. Global and 30offAP obtained at 31 patient visits in 2011 were also included.The measure best-correlated with overall Global asymmetry was digital fz-eu (R = 0.80). Caliper fz-eu was not strongly correlated with Global asymmetry (R = 0.27) or with digital fz-eu (R = 0.34). Differences between the digital and caliper fz-eu measures were 5 ± 4 mm (mean ± st. dev.). Differences between the caliper fz-eu and 30offAP measures were 6 ± 4 mm.Digital fz-eu shows an excellent correlation to Global asymmetry. However, attempts to replicate this result in the clinical setting by measuring fz-eu with calipers were unsuccessful.

Automated face extraction and normalization of 3D Mesh Data

3D stereophotography is rapidly being adopted by medical researchers for analysis of facial forms and features. An essential step for many applications using 3D face data is to first crop the head and face from the raw images. The goal of this paper is to develop a reliable automatic methodology for extracting the face from raw data with texture acquired from a stereo imaging system, based on the medical researchers' specific requirements. We present an automated process, including eye and nose estimation, face detection, Procrustes analysis and final noise removal to crop out the faces and normalize them. The proposed method shows very reliable results on several datasets, including a normal adult dataset and a very challenging dataset consisting of infants with cleft lip and palate.

Digital capture, design, and manufacturing of a facial prosthesis: Clinical report on a pediatric patient

A digitally captured, designed, and fabricated facial prosthesis is presented as an alternative to customary maxillofacial prosthodontics fabrication techniques, where a facial moulage and patient cooperation may be difficult.

Learning to Rank the Severity of Unrepaired Cleft Lip Nasal Deformity on 3D Mesh Data

Cleft lip is a birth defect that results in deformity of the upper lip and nose. Its severity is widely variable and the results of treatment are influenced by the initial deformity. Objective assessment of severity would help to guide prognosis and treatment. However, most assessments are subjective. The purpose of this study is to develop and test quantitative computer-based methods of measuring cleft lip severity. In this paper, a grid-patch based measurement of symmetry is introduced, with which a computer program learns to rank the severity of cleft lip on 3D meshes of human infant faces. Three computer-based methods to define the midfacial reference plane were compared to two manual methods. Four different symmetry features were calculated based upon these reference planes, and evaluated. The result shows that the rankings predicted by the proposed features were highly correlated with the ranking orders provided by experts that were used as the ground truth.

Classifying Craniosynostosis with a 3D Projection-Based Feature Extraction System

Craniosynostosis, a disorder in which one or more fibrous joints of the skull fuse prematurely, causes skull deformity and is associated with increased intracranial pressure and developmental delays. Although clinicians can easily diagnose craniosynostosis and can classify its type, being able to quantify the condition is an important problem in craniofacial research. While several papers have attempted this quantification through statistical models, the methods have not been intuitive to biomedical researchers and clinicians who want to use them. The goal of this work was to develop a general platform upon which new quantification measures could be developed and tested. The features reported in this paper were developed as basic shape measures, both single-valued and vector-valued, that are extracted from a single plane projection of the 3D skull. This technique allows us to process images that would otherwise be eliminated in previous systems due to poor resolution, noise or imperfections on their CT scans. We test our new features on classification tasks and also compare their performance to previous research. In spite of its simplicity, the classification accuracy of our new features is significantly higher than previous results on head CT scan data from the same research studies.

Quantification of skull deformity for craniofacial research

Craniosynostosis, a disorder in which one or more fibrous joints of the skull fuse prematurely, causes skull malformation and may be associated with increased intracranial pressure and developmental delays. In order to perform medical research studies that relate phenotypic abnormalities to outcomes such as cognitive ability or results of surgery, biomedical researchers need an automated methodology for quantifying the degree of abnormality of the disorder. This paper addresses that need by proposing a set of features derived from CT scans of the skull that can be used for this purpose. A thorough set of experiments is used to evaluate the features as compared to two human craniofacial experts in a ranking evaluation.

Plagiocephaly and brachycephaly treatment with cranial orthosis: a case report

The number of cranial deformities has increased considerably since international efforts of pediatricians to recommend parents putting their babies to sleep in the supine position as a strategy to reduce sudden death syndrome of the newborn. On the one hand, this program has demonstrated very efficient results at reducing deaths and, on the other hand, such recommendation has increased the incidence of cranial asymmetries. In addition, infants are kept too long in one position, much of this due to abusive use of strollers, baby carriers, car seats, swings and other devices. Among resulting asymmetries, the most frequently found are plagiocephaly (parallelogram shaped skull, with posterior unilateral flattening with the opposite frontal area also flattened) and brachycephaly (occipital bilateral flattening). The present study is a case report of a patient with brachycephaly associated with deformational plagiocephaly treated with cranial orthosis. The same physician clinically evaluated the patient before and after treatment using photographic recording and a laser scanning device, which allows the accurate measurement of variables determining asymmetries. It became clear during treatment that there was significant improvement in cranial symmetry documented by decrease in the cephalic index, diagonal difference and volume gain in the quadrant that was flattened. The authors conclude that orthotic therapy is a safe and effective therapeutic modality for position cranial asymmetries.

Shape-based classification of 3D facial data to support 22q11.2DS craniofacial research

3D imaging systems are used to construct high-resolution meshes of patient's heads that can be analyzed by computer algorithms. Our work starts with such 3D head meshes and produces both global and local descriptors of 3D shape. Since these descriptors are numeric feature vectors, they can be used in both classification and quantification of various different abnormalities. In this paper, we define these descriptors, describe our methodology for constructing them from 3D head meshes, and show through a set of classification experiments involving cases and controls for a genetic disorder called 22q11.2 deletion syndrome that they are suitable for use in craniofacial research studies. The main contributions of this work include: automatic generation of novel global and local data representations, robust automatic placement of anthropometric landmarks, generation of local descriptors for nasal and oral facial features from landmarks, use of local descriptors for predicting various local facial features, and use of global features for 22q11.2DS classification, showing their potential use as descriptors in craniofacial research.

Brain volume and shape in infants with deformational plagiocephaly

PURPOSE

Infants with deformational plagiocephaly (DP) have been shown to exhibit developmental delays relative to unaffected infants. Although the mechanisms accounting for these delays are unknown, one hypothesis focuses on underlying differences in brain development. In this study, we used MRI to examine brain volume and shape in infants with and without DP.

METHODS

Participants included 20 infants with DP (mean age = 7.9 months, SD = 1.2; n = 12 male) and 21 controls (mean age = 7.9 months, SD = 1.3; n = 11 male). Measures included volumes of the total brain and cerebellum; midsagittal areas of the corpus callosum and cerebellar vermis; and linear distance measures used to quantify the shape of selected brain structures. We also evaluated the association between shape measures and developmental scores on the Bayley Scales of Infant and Toddler Development-III (BSID-III).

RESULTS

Brain volume did not distinguish cases and controls (p = .214-.976). However, cases exhibited greater asymmetry and flattening of the posterior brain (p < .001-.002) and cerebellar vermis (p = .035), shortening of the corpus callosum (p = .012), and differences in the orientation of the corpus callosum (p = .005). Asymmetry and flattening of brain structures were associated with worse developmental outcomes on the BSID-III.

CONCLUSIONS

Infants with DP show differences in brain shape, consistent with the skull deformity characteristic of this condition, and shape measures were associated with infant development. Longitudinal studies, beginning in the neonatal period, are needed to clarify whether developmental effects precede or follow brain deformation.

Longitudinal, three-dimensional analysis of head shape in children with and without deformational plagiocephaly or brachycephaly

OBJECTIVES

To assess 3-dimensional (3D) changes in head shape in infancy and at age 18 months in children with and without plagiocephaly or brachycephaly.

STUDY DESIGN

Using a longitudinal design, we evaluated head shape using 3D surface imaging. We compared the head shapes of children with (1) diagnosed deformational plagiocephaly or brachycephaly (cases; n=233); (2) unaffected controls, with no evidence of dysmorphology (n=167); and (3) affected controls, who despite having no previous diagnosis demonstrated skull dysmorphology on 3D surface imaging (n=70).

RESULTS

Cases had greater skull flattening and asymmetry than unaffected controls at both time points, as did controls with skull dysmorphology. In all groups, head shapes became less flat and more symmetric over time. Among cases, symmetry improved slightly more for those who received orthotic treatment.

CONCLUSIONS

Although head shape improves over time for children with deformational plagiocephaly or brachycephaly, skull dysmorphology persists relative to unaffected controls. Further research is needed to clarify the extent to which these differences are detectable to clinicians and lay observers.

The Use of Genetic Programming for Learning 3D Craniofacial Shape Quantifications

Craniofacial disorders commonly result in various head shape dysmorphologies. The goal of this work is to quantify the various 3D shape variations that manifest in the different facial abnormalities in individuals with a craniofacial disorder called 22q11.2 Deletion Syndrome. Genetic programming (GP) is used to learn the different 3D shape quantifications. Experimental results show that the GP method achieves a higher classification rate than those of human experts and existing computer algorithms [1], [2].