Evaluating Face2Gene as a Tool to Identify Cornelia de Lange Syndrome by Facial Phenotypes

Computer-based facial recognition as an assisting diagnostic tool to identify children with Noonan syndrome

BACKGROUND

Noonan syndrome (NS) is a rare genetic disease, and patients who suffer from it exhibit a facial morphology that is characterized by a high forehead, hypertelorism, ptosis, inner epicanthal folds, down-slanting palpebral fissures, a highly arched palate, a round nasal tip, and posteriorly rotated ears. Facial analysis technology has recently been applied to identify many genetic syndromes (GSs). However, few studies have investigated the identification of NS based on the facial features of the subjects.

OBJECTIVES

This study develops advanced models to enhance the accuracy of diagnosis of NS.

METHODS

A total of 1,892 people were enrolled in this study, including 233 patients with NS, 863 patients with other GSs, and 796 healthy children. We took one to 10 frontal photos of each subject to build a dataset, and then applied the multi-task convolutional neural network (MTCNN) for data pre-processing to generate standardized outputs with five crucial facial landmarks. The ImageNet dataset was used to pre-train the network so that it could capture generalizable features and minimize data wastage. We subsequently constructed seven models for facial identification based on the VGG16, VGG19, VGG16-BN, VGG19-BN, ResNet50, MobileNet-V2, and squeeze-and-excitation network (SENet) architectures. The identification performance of seven models was evaluated and compared with that of six physicians.

RESULTS

All models exhibited a high accuracy, precision, and specificity in recognizing NS patients. The VGG19-BN model delivered the best overall performance, with an accuracy of 93.76%, precision of 91.40%, specificity of 98.73%, and F1 score of 78.34%. The VGG16-BN model achieved the highest AUC value of 0.9787, while all models based on VGG architectures were superior to the others on the whole. The highest scores of six physicians in terms of accuracy, precision, specificity, and the F1 score were 74.00%, 75.00%, 88.33%, and 61.76%, respectively. The performance of each model of facial recognition was superior to that of the best physician on all metrics.

CONCLUSION

Models of computer-assisted facial recognition can improve the rate of diagnosis of NS. The models based on VGG19-BN and VGG16-BN can play an important role in diagnosing NS in clinical practice.

The Approach to a Child with Dysmorphic Features: What the Pediatrician Should Know

The advancement of genetic knowledge and the discovery of an increasing number of genetic disorders has made the role of the geneticist progressively more complex and fundamental. However, most genetic disorders present during childhood; thus, their early recognition is a challenge for the pediatrician, who will be also involved in the follow-up of these children, often establishing a close relationship with them and their families and becoming a referral figure. In this review, we aim to provide the pediatrician with a general knowledge of the approach to treating a child with a genetic syndrome associated with dysmorphic features. We will discuss the red flags, the most common manifestations, the analytic collection of the family and personal medical history, and the signs that should alert the pediatrician during the physical examination. We will offer an overview of the physical malformations most commonly associated with genetic defects and the way to describe dysmorphic facial features. We will provide hints about some tools that can support the pediatrician in clinical practice and that also represent a useful educational resource, either online or through apps downloaded on a smartphone. Eventually, we will offer an overview of genetic testing, the ethical considerations, the consequences of incidental findings, and the main indications and limitations of the principal technologies.

Cornelia de Lange Spectrum

Cornelia de Lange syndrome (CdLS) is a rare congenital developmental disorder with multisystemic involvement. The clinical presentation is highly variable, but the classic phenotype, characterized by distinctive craniofacial features, pre- and postnatal growth retardation, extremity reduction defects, hirsutism and intellectual disability can be distinguished from the nonclassic phenotype, which is generally milder and more difficult to diagnose. In addition, the clinical features overlap with those of other neurodevelopmental disorders, so the use of consensus clinical criteria and artificial intelligence tools may be helpful in confirming the diagnosis. Pathogenic variants in NIPBL, which encodes a protein related to the cohesin complex, have been identified in more than 60% of patients, and pathogenic variants in other genes related to this complex in another 15%: SMC1A, SMC3, RAD21, and HDAC8. Technical advances in large-scale sequencing have allowed the description of additional genes (BRD4, ANKRD11, MAU2), but the lack of molecular diagnosis in 15% of individuals and the substantial clinical heterogeneity of the syndrome suggest that other genes and mechanisms may be involved. Although there is no curative treatment, there are symptomatic/palliative treatments that paediatricians should be aware of. The main medical complication in classic SCdL is gastro-esophageal reflux (GER), which should be treated early.

Artificial intelligence-based diagnosis in fetal pathology using external ear shapes

OBJECTIVE

Here we trained an automatic phenotype assessment tool to recognize syndromic ears in two syndromes in fetuses-=CHARGE and Mandibulo-Facial Dysostosis Guion Almeida type (MFDGA)-versus controls.

METHOD

We trained an automatic model on all profile pictures of children diagnosed with genetically confirmed MFDGA and CHARGE syndromes, and a cohort of control patients, collected from 1981 to 2023 in Necker Hospital (Paris) with a visible external ear. The model consisted in extracting landmarks from photographs of external ears, in applying geometric morphometry methods, and in a classification step using machine learning. The approach was then tested on photographs of two groups of fetuses: controls and fetuses with CHARGE and MFDGA syndromes.

RESULTS

The training set contained a total of 1489 ear photographs from 526 children. The validation set contained a total of 51 ear photographs from 51 fetuses. The overall accuracy was 72.6% (58.3%-84.1%, p < 0.001), and 76.4%, 74.9%, and 86.2% respectively for CHARGE, control and MFDGA fetuses. The area under the curves were 86.8%, 87.5%, and 90.3% respectively for CHARGE, controls, and MFDGA fetuses.

CONCLUSION

We report the first automatic fetal ear phenotyping model, with satisfactory classification performances. Further validations are required before using this approach as a diagnostic tool.

Application of the Face2Gene tool in an Italian dysmorphological pediatric clinic: Retrospective validation and future perspectives

Neurodevelopmental disorders exhibit recurrent facial features that can suggest the genetic diagnosis at a glance, but recognizing subtle dysmorphisms is a specialized skill that requires very long training. Face2Gene (FDNA Inc) is an innovative computer-aided phenotyping tool that analyses patient's portraits and suggests 30 candidate syndromes with similar morphology in a prioritized list. We hypothesized that the software could support even expert physicians in the diagnostic workup of genetic conditions. In this study, we assessed the performance of Face2Gene in an Italian dysmorphological pediatrics clinic. We uploaded two-dimensional face pictures of 145 children affected by genetic conditions with typical phenotypic traits. All diagnoses were previously confirmed by cytogenetic or molecular tests. Overall, the software's differential included the correct syndrome in most cases (98%). We evaluated the efficiency of the algorithm even considering the rareness of the genetic conditions. All "common" diagnoses were correctly identified, most of them with high diagnostic accuracy (93% in top-3 matches). Finally, the performance for the most common pediatric syndromes was calculated. Face2Gene performed well even for ultra-rare genetic conditions (75% within top-3 matches and 83% within top-10 matches). Expert geneticists maybe do not need computer support to recognize common syndromes, but our results prove that the tool can be useful not only for general pediatricians but also in dysmorphological clinics for ultra-rare genetic conditions.

Next generation phenotyping for diagnosis and phenotype-genotype correlations in Kabuki syndrome

The field of dysmorphology has been changed by the use Artificial Intelligence (AI) and the development of Next Generation Phenotyping (NGP). The aim of this study was to propose a new NGP model for predicting KS (Kabuki Syndrome) on 2D facial photographs and distinguish KS1 (KS type 1, KMT2D-related) from KS2 (KS type 2, KDM6A-related). We included retrospectively and prospectively, from 1998 to 2023, all frontal and lateral pictures of patients with a molecular confirmation of KS. After automatic preprocessing, we extracted geometric and textural features. After incorporation of age, gender, and ethnicity, we used XGboost (eXtreme Gradient Boosting), a supervised machine learning classifier. The model was tested on an independent validation set. Finally, we compared the performances of our model with DeepGestalt (Face2Gene). The study included 1448 frontal and lateral facial photographs from 6 centers, corresponding to 634 patients (527 controls, 107 KS); 82 (78%) of KS patients had a variation in the KMT2D gene (KS1) and 23 (22%) in the KDM6A gene (KS2). We were able to distinguish KS from controls in the independent validation group with an accuracy of 95.8% (78.9-99.9%, p < 0.001) and distinguish KS1 from KS2 with an empirical Area Under the Curve (AUC) of 0.805 (0.729-0.880, p < 0.001). We report an automatic detection model for KS with high performances (AUC 0.993 and accuracy 95.8%). We were able to distinguish patients with KS1 from KS2, with an AUC of 0.805. These results outperform the current commercial AI-based solutions and expert clinicians.

Cornelia de Lange Syndrome Caused by an Intragenic Heterozygous Deletion in RAD21 Detected through Very-High-Resolution Chromosomal Microarray Analysis

Cornelia de Lange syndrome is a genetic and clinically heterogeneous entity, caused by at least five genes. It is characterized by short stature, gestalt facies, microcephaly, neurodevelopmental disorders, and other anomalies. In this report, we present a 13-year-old female patient with microcephaly, cleft palate, polydactyly, short stature, triangular facies, frontal bossing, a bulbous nose, an overfolded helix, limited pronosupination, and an anomalous uterus. No neurodevelopmental disorders were reported. A chromosomal microarray analysis of 6.5 million markers was performed in the proband and her parents. The results showed a de novo heterozygous microdeletion of exons 9-14 within RAD21, which confirmed the diagnosis of Cornelia de Lange syndrome type 4. Our patient did not show any neurologic phenotype (until the time of diagnosis), although neurodevelopmental disorders are frequently present in patients with Cornelia de Lange syndrome type 4, and despite carrying a deletion that was larger than previously reported. Therefore, unknown genetic modifiers or intrinsic mechanisms of RAD21 variants may exist and should be studied.

AI-based diagnosis in mandibulofacial dysostosis with microcephaly using external ear shapes

Introduction

Mandibulo-Facial Dysostosis with Microcephaly (MFDM) is a rare disease with a broad spectrum of symptoms, characterized by zygomatic and mandibular hypoplasia, microcephaly, and ear abnormalities. Here, we aimed at describing the external ear phenotype of MFDM patients, and train an Artificial Intelligence (AI)-based model to differentiate MFDM ears from non-syndromic control ears (binary classification), and from ears of the main differential diagnoses of this condition (multi-class classification): Treacher Collins (TC), Nager (NAFD) and CHARGE syndromes.

Methods

The training set contained 1,592 ear photographs, corresponding to 550 patients. We extracted 48 patients completely independent of the training set, with only one photograph per ear per patient. After a CNN-(Convolutional Neural Network) based ear detection, the images were automatically landmarked. Generalized Procrustes Analysis was then performed, along with a dimension reduction using PCA (Principal Component Analysis). The principal components were used as inputs in an eXtreme Gradient Boosting (XGBoost) model, optimized using a 5-fold cross-validation. Finally, the model was tested on an independent validation set.

Results

We trained the model on 1,592 ear photographs, corresponding to 1,296 control ears, 105 MFDM, 33 NAFD, 70 TC and 88 CHARGE syndrome ears. The model detected MFDM with an accuracy of 0.969 [0.838-0.999] (p < 0.001) and an AUC (Area Under the Curve) of 0.975 within controls (binary classification). Balanced accuracies were 0.811 [0.648-0.920] (p = 0.002) in a first multiclass design (MFDM vs. controls and differential diagnoses) and 0.813 [0.544-0.960] (p = 0.003) in a second multiclass design (MFDM vs. differential diagnoses).

Conclusion

This is the first AI-based syndrome detection model in dysmorphology based on the external ear, opening promising clinical applications both for local care and referral, and for expert centers.

Assessing facial weakness in myasthenia gravis with facial recognition software and deep learning

OBJECTIVE

Myasthenia gravis (MG) is an autoimmune disease leading to fatigable muscle weakness. Extra-ocular and bulbar muscles are most commonly affected. We aimed to investigate whether facial weakness can be quantified automatically and used for diagnosis and disease monitoring.

METHODS

In this cross-sectional study, we analyzed video recordings of 70 MG patients and 69 healthy controls (HC) with two different methods. Facial weakness was first quantified with facial expression recognition software. Subsequently, a deep learning (DL) computer model was trained for the classification of diagnosis and disease severity using multiple cross-validations on videos of 50 patients and 50 controls. Results were validated using unseen videos of 20 MG patients and 19 HC.

RESULTS

Expression of anger (p = 0.026), fear (p = 0.003), and happiness (p < 0.001) was significantly decreased in MG compared to HC. Specific patterns of decreased facial movement were detectable in each emotion. Results of the DL model for diagnosis were as follows: area under the curve (AUC) of the receiver operator curve 0.75 (95% CI 0.65-0.85), sensitivity 0.76, specificity 0.76, and accuracy 76%. For disease severity: AUC 0.75 (95% CI 0.60-0.90), sensitivity 0.93, specificity 0.63, and accuracy 80%. Results of validation, diagnosis: AUC 0.82 (95% CI: 0.67-0.97), sensitivity 1.0, specificity 0.74, and accuracy 87%. For disease severity: AUC 0.88 (95% CI: 0.67-1.0), sensitivity 1.0, specificity 0.86, and accuracy 94%.

INTERPRETATION

Patterns of facial weakness can be detected with facial recognition software. Second, this study delivers a 'proof of concept' for a DL model that can distinguish MG from HC and classifies disease severity.

A Deep-Learning-Based Method Can Detect Both Common and Rare Genetic Disorders in Fetal Ultrasound

A global survey indicates that genetic syndromes affect approximately 8% of the population, but most genetic diagnoses can only be performed after babies are born. Abnormal facial characteristics have been identified in various genetic diseases; however, current facial identification technologies cannot be applied to prenatal diagnosis. We developed Pgds-ResNet, a fully automated prenatal screening algorithm based on deep neural networks, to detect high-risk fetuses affected by a variety of genetic diseases. In screening for Trisomy 21, Trisomy 18, Trisomy 13, and rare genetic diseases, Pgds-ResNet achieved sensitivities of 0.83, 0.92, 0.75, and 0.96, and specificities of 0.94, 0.93, 0.95, and 0.92, respectively. As shown in heatmaps, the abnormalities detected by Pgds-ResNet are consistent with clinical reports. In a comparative experiment, the performance of Pgds-ResNet is comparable to that of experienced sonographers. This fetal genetic screening technology offers an opportunity for early risk assessment and presents a non-invasive, affordable, and complementary method to identify high-risk fetuses affected by genetic diseases. Additionally, it has the capability to screen for certain rare genetic conditions, thereby enhancing the clinic's detection rate.

Computer-aided facial analysis as a tool to identify patients with Silver-Russell syndrome and Prader-Willi syndrome

Genetic syndromes often show facial features that provide clues for the diagnosis. However, memorizing these features is a challenging task for clinicians. In the last years, the app Face2Gene proved to be a helpful support for the diagnosis of genetic diseases by analyzing features detected in one or more facial images of affected individuals. Our aim was to evaluate the performance of the app in patients with Silver-Russell syndrome (SRS) and Prader-Willi syndrome (PWS). We enrolled 23 pediatric patients with clinically or genetically diagnosed SRS and 29 pediatric patients with genetically confirmed PWS. One frontal photo of each patient was acquired. Top 1, top 5, and top 10 sensitivities were analyzed. Correlation with the specific genetic diagnosis was investigated. When available, photos of the same patient at different ages were compared. In the SRS group, Face2Gene showed top 1, top 5, and top 10 sensitivities of 39%, 65%, and 91%, respectively. In 41% of patients with genetically confirmed SRS, SRS was the first syndrome suggested, while in clinically diagnosed patients, SRS was suggested as top 1 in 33% of cases (p = 0.74). Face2Gene performed better in younger patients with SRS: in all patients in whom a photo taken at a younger age than the age of enrollment was available, SRS was suggested as top 1, albeit with variable degree of probability. In the PWS group, the top 1, top 5, and top 10 sensitivities were 76%, 97%, and 100%, respectively. PWS was suggested as top 1 in 83% of patients genetically diagnosed with paternal deletion of chromosome 15q11-13 and in 60% of patients presenting with maternal uniparental disomy of chromosome 15 (p = 0.17). The performance was uniform throughout the investigated age range (1-15 years).

CONCLUSION

In addition to a thorough medical history and detailed clinical examination, the Face2Gene app can be a useful tool to support clinicians in identifying children with a potential diagnosis of SRS or PWS.

WHAT IS KNOWN

• Several genetic syndromes present typical facial features that may provide clues for the diagnosis. • Memorizing all syndromic facial characteristics is a challenging task for clinicians.

WHAT IS NEW

• Face2Gene may represent a useful support for pediatricians for the diagnosis of genetic syndromes. • Face2Gene app can be a useful tool to integrate in the diagnostic path of patients with SRS and PWS.

Expansion of the genotypic and phenotypic spectrum of CTCF-related disorder guides clinical management: 43 new subjects and a comprehensive literature review

Monoallelic variants of CTCF cause an autosomal dominant neurodevelopmental disorder with a wide range of features, including impacts on the brain, growth, and craniofacial development. A growing number of subjects with CTCF-related disorder (CRD) have been identified due to the increased application of exome sequencing, and further delineation of the clinical spectrum of CRD is needed. Here, we examined the clinical features, including facial profiles, and genotypic spectrum of 107 subjects with identified CTCF variants, including 43 new and 64 previously described subjects. Among the 43 new subjects, 23 novel variants were reported. The cardinal clinical features in subjects with CRD included intellectual disability/developmental delay (91%) with speech delay (65%), motor delay (53%), feeding difficulties/failure to thrive (66%), ocular abnormalities (56%), musculoskeletal anomalies (53%), and behavioral problems (52%). Other congenital anomalies were also reported, but none of them were common. Our findings expanded the genotypic and phenotypic spectrum of CRD that will guide genetic counseling, management, and surveillance care for patients with CRD. Additionally, a newly built facial gestalt on the Face2Gene tool will facilitate prompt recognition of CRD by physicians and shorten a patient's diagnostic odyssey.

Chung-Jansen syndrome can mimic Cornelia de Lange syndrome: Another player among chromatinopathies?

Cornelia de Lange syndrome (CdLS) is a rare multisystem congenital neurodevelopmental disorder (NDD) characterized by distinctive facial anomalies, short stature, developmental delay, hirsutism, gastrointestinal abnormalities and upper limb reduction defects. CdLS syndrome is associated with causative variants in genes encoding for the cohesin complex, a cellular machinery involved in chromatid pairing, DNA repair and gene-expression regulation. In this report, we describe a familial case of a syndromic presentation in a 4-year-old patient (P1) and in his mother (P2). Trio-based Whole Exome Sequencing (WES) performed on P1 was first negative. Since his phenotypic evolution during the follow-up was reminiscent of the CdLS spectrum, a reanalysis of WES data, focused on CdLS-related genes, was requested. Although no alterations in those genes was detected, we identified the likely pathogenetic variant c.40G > A (p.Glu14Lys) in the PHIP gene, in the meanwhile associated with Chung-Jansen syndrome. Reverse phenotyping carried out in both patients confirmed the molecular diagnosis. CHUJANS belongs to NDDs, featuring developmental delay, mild-to-moderate intellectual disability, behavioral problems, obesity and facial dysmorphisms. Moreover, as here described, CHUJANS shows a significant overlap with the CdLS spectrum, with specific regard to facial gestalt. On the basis of our findings, we suggest to include PHIP among genes routinely analyzed in patients belonging to the CdLS spectrum.

An automatic facial landmarking for children with rare diseases

Two to three thousand syndromes modify facial features: their screening requires the eye of an expert in dysmorphology. A widely used tool in shape characterization is geometric morphometrics based on landmarks, which are precise and reproducible anatomical points. Landmark positioning is user dependent and time consuming. Many automatic landmarking tools are currently available but do not work for children, because they have mainly been trained using photographic databases of healthy adults. Here, we developed a method for building an automatic landmarking pipeline for frontal and lateral facial photographs as well as photographs of external ears. We evaluated the algorithm on patients diagnosed with Treacher Collins (TC) syndrome as it is the most frequent mandibulofacial dysostosis in humans and is clinically recognizable although highly variable in severity. We extracted photographs from the photographic database of the maxillofacial surgery and plastic surgery department of Hôpital Necker-Enfants Malades in Paris, France with the diagnosis of TC syndrome. The control group was built from children admitted for craniofacial trauma or skin lesions. After testing two methods of object detection by bounding boxes, a Haar Cascade-based tool and a Faster Region-based Convolutional Neural Network (Faster R-CNN)-based tool, we evaluated three different automatic annotation algorithms: the patch-based active appearance model (AAM), the holistic AAM, and the constrained local model (CLM). The final error corresponding to the distance between the points placed by automatic annotation and those placed by manual annotation was reported. We included, respectively, 1664, 2044, and 1375 manually annotated frontal, profile, and ear photographs. Object recognition was optimized with the Faster R-CNN-based detector. The best annotation model was the patch-based AAM (p < 0.001 for frontal faces, p = 0.082 for profile faces and p < 0.001 for ears). This automatic annotation model resulted in the same classification performance as manually annotated data. Pretraining on public photographs did not improve the performance of the model. We defined a pipeline to create automatic annotation models adapted to faces with congenital anomalies, an essential prerequisite for research in dysmorphology.

Smartphone-Based Facial Scanning as a Viable Tool for Facially Driven Orthodontics?

The current paradigm shift in orthodontic treatment planning is based on facially driven diagnostics. This requires an affordable, convenient, and non-invasive solution for face scanning. Therefore, utilization of smartphones' TrueDepth sensors is very tempting. TrueDepth refers to front-facing cameras with a dot projector in Apple devices that provide real-time depth data in addition to visual information. There are several applications that tout themselves as accurate solutions for 3D scanning of the face in dentistry. Their clinical accuracy has been uncertain. This study focuses on evaluating the accuracy of the Bellus3D Dental Pro app, which uses Apple's TrueDepth sensor. The app reconstructs a virtual, high-resolution version of the face, which is available for download as a 3D object. In this paper, sixty TrueDepth scans of the face were compared to sixty corresponding facial surfaces segmented from CBCT. Difference maps were created for each pair and evaluated in specific facial regions. The results confirmed statistically significant differences in some facial regions with amplitudes greater than 3 mm, suggesting that current technology has limited applicability for clinical use. The clinical utilization of facial scanning for orthodontic evaluation, which does not require accuracy in the lip region below 3 mm, can be considered.

Case report: A novel case of parental mosaicism in SMC1A gene causes inherited Cornelia de Lange syndrome

Ultimate advances in genetic technologies have permitted the detection of transmitted cases of congenital diseases due to parental gonadosomatic mosaicism. Regarding Cornelia de Lange syndrome (CdLS), up to date, only a few cases are known to follow this inheritance pattern. However, the high prevalence of somatic mosaicism recently reported in this syndrome (∼13%), together with the disparity observed in tissue distribution of the causal variant, suggests that its prevalence in this disorder could be underestimated. Here, we report a new case of parental gonadosomatic mosaicism in SMC1A gene that causes inherited CdLS, in which the mother of the patient carries the causative variant in very low allele frequencies in buccal swab and blood. While the affected child presents with typical CdLS phenotype, his mother does not show any clinical manifestations. As regards SMC1A, the difficulty of clinical identification of carrier females has been already recognized, as well as the gender differences observed in CdLS expressivity when the causal variant is found in this gene. Currently, the use of DNA deep-sequencing techniques is highly recommended when it comes to molecular diagnosis of patients, as well as in co-segregation studies. These enable us to uncover gonadosomatic mosaic events in asymptomatic or oligosymptomatic parents that had been overlooked so far, which might have great implications regarding genetic counseling for recurrence risk.

A Novel Intragenic Duplication in the HDAC8 Gene Underlying a Case of Cornelia de Lange Syndrome

Cornelia de Lange syndrome (CdLS) is a multisystemic genetic disorder characterized by distinctive facial features, growth retardation, and intellectual disability, as well as various systemic conditions. It is caused by genetic variants in genes related to the cohesin complex. Single-nucleotide variations are the best-known genetic cause of CdLS; however, copy number variants (CNVs) clearly underlie a substantial proportion of cases of the syndrome. The NIPBL gene was thought to be the locus within which clinically relevant CNVs contributed to CdLS. However, in the last few years, pathogenic CNVs have been identified in other genes such as HDAC8, RAD21, and SMC1A. Here, we studied an affected girl presenting with a classic CdLS phenotype heterozygous for a de novo ~32 kbp intragenic duplication affecting exon 10 of HDAC8. Molecular analyses revealed an alteration in the physiological splicing that included a 96 bp insertion between exons 9 and 10 of the main transcript of HDAC8. The aberrant transcript was predicted to generate a truncated protein whose accessibility to the active center was restricted, showing reduced ease of substrate entry into the mutated enzyme. Lastly, we conclude that the duplication is responsible for the patient’s phenotype, highlighting the contribution of CNVs as a molecular cause underlying CdLS.

Case Report: The success of face analysis technology in extremely rare genetic diseases in Korea: Tatton–Brown–Rahman syndrome and Say–Barber –Biesecker–Young–Simpson variant of ohdo syndrome

Tatton–Brown–Rahman syndrome (TBRS) and Say–Barber–Biesecker– Young–Simpson variant of Ohdo syndrome (SBBYSS) are extremely rare genetic disorders with less than 100 reported cases. Patients with these disorders exhibit a characteristic facial dysmorphism: TBRS is characterized by a round face, a straight and thick eyebrow, and prominent maxillary incisors, whereas SBBYSS is characterized by mask-like facies, blepharophimosis, and ptosis. The usefulness of Face2Gene as a tool for the identification of dysmorphology syndromes is discussed, because, in these patients, it suggested TBRS and SBBYSS within the top five candidate disorders. Face2Gene is useful for the diagnosis of extremely rare diseases in Korean patients, suggesting the possibility of expanding its clinical applications.

Refining nosology by modelling variation among facial phenotypes: the RASopathies

BACKGROUND

In clinical genetics, establishing an accurate nosology requires analysis of variations in both aetiology and the resulting phenotypes. At the phenotypic level, recognising typical facial gestalts has long supported clinical and molecular diagnosis; however, the objective analysis of facial phenotypic variation remains underdeveloped. In this work, we propose exploratory strategies for assessing facial phenotypic variation within and among clinical and molecular disease entities and deploy these techniques on cross-sectional samples of four RASopathies: Costello syndrome (CS), Noonan syndrome (NS), cardiofaciocutaneous syndrome (CFC) and neurofibromatosis type 1 (NF1).

METHODS

From three-dimensional dense surface scans, we model the typical phenotypes of the four RASopathies as average 'facial signatures' and assess individual variation in terms of direction (what parts of the face are affected and in what ways) and severity of the facial effects. We also derive a metric of phenotypic agreement between the syndromes and a metric of differences in severity along similar phenotypes.

RESULTS

CFC shows a relatively consistent facial phenotype in terms of both direction and severity that is similar to CS and NS, consistent with the known difficulty in discriminating CFC from NS based on the face. CS shows a consistent directional phenotype that varies in severity. Although NF1 is highly variable, on average, it shows a similar phenotype to CS.

CONCLUSIONS

We established an approach that can be used in the future to quantify variations in facial phenotypes between and within clinical and molecular diagnoses to objectively define and support clinical nosologies.

Review on Facial-Recognition-Based Applications in Disease Diagnosis

Diseases not only manifest as internal structural and functional abnormalities, but also have facial characteristics and appearance deformities. Specific facial phenotypes are potential diagnostic markers, especially for endocrine and metabolic syndromes, genetic disorders, facial neuromuscular diseases, etc. The technology of facial recognition (FR) has been developed for more than a half century, but research in automated identification applied in clinical medicine has exploded only in the last decade. Artificial-intelligence-based FR has been found to have superior performance in diagnosis of diseases. This interdisciplinary field is promising for the optimization of the screening and diagnosis process and assisting in clinical evaluation and decision-making. However, only a few instances have been translated to practical use, and there is need of an overview for integration and future perspectives. This review mainly focuses on the leading edge of technology and applications in varieties of disease, and discusses implications for further exploration.

Disease-associated c-MYC downregulation in human disorders of transcriptional regulation

Cornelia de Lange syndrome (CdLS) is a rare multiorgan developmental disorder caused by pathogenic variants in cohesin genes. It is a genetically and clinically heterogeneous dominant (both autosomal and X-linked) rare disease. Increasing experimental evidence indicates that CdLS is caused by a combination of factors, such as gene expression dysregulation, accumulation of cellular damage and cellular aging, which collectively contribute to the CdLS phenotype. The CdLS phenotype overlaps with a number of related diagnoses such as KBG syndrome and Rubinstein-Taybi syndrome both caused by variants in chromatin-associated factors other than cohesin. The molecular basis underlying these overlapping phenotypes is not clearly defined. Here, we found that cells from individuals with CdLS and CdLS-related diagnoses are characterized by global transcription disturbance and share common dysregulated pathways. Intriguingly, c-MYC (subsequently referred to as MYC) is downregulated in all cell lines and represents a convergent hub lying at the center of dysregulated pathways. Subsequent treatment with estradiol restores MYC expression by modulating cohesin occupancy at its promoter region. In addition, MYC activation leads to modification in expression in hundreds of genes, which in turn reduce the oxidative stress level and genome instability. Together, these results show that MYC plays a pivotal role in the etiopathogenesis of CdLS and CdLS-related diagnoses and represents a potential therapeutic target for these conditions.

Best practices for the interpretation and reporting of clinical whole genome sequencing

Whole genome sequencing (WGS) shows promise as a first-tier diagnostic test for patients with rare genetic disorders. However, standards addressing the definition and deployment practice of a best-in-class test are lacking. To address these gaps, the Medical Genome Initiative, a consortium of leading health care and research organizations in the US and Canada, was formed to expand access to high quality clinical WGS by convening experts and publishing best practices. Here, we present best practice recommendations for the interpretation and reporting of clinical diagnostic WGS, including discussion of challenges and emerging approaches that will be critical to harness the full potential of this comprehensive test.

Subclinical myocardial dysfunction is revealed by speckle tracking echocardiography in patients with Cornelia de Lange syndrome

This study assesses a possible cardiac dysfunction in individuals with Cornelia de Lange syndrome (CdLS) without diagnosed congenital heart disease (CHD) and its association with other factors. Twenty patients and 20 controls were included in the study divided into three age-dependent groups (A: < 10 yrs, B: 10-20 yrs, C: > 20 yrs), and were evaluated using conventional echocardiography, tissue doppler imaging (TDI), two-dimensional speckle tracking and genetic and biochemical analyses. The left ventricular global longitudinal strain (GLS) was altered (< 15.9%) in 55% of patients, being pathological in the older group (A: 19.7 ± 6.6; B: -17.2 ± 4.7; C: -13.6 ± 2.9). The speckle tracking technique revealed a downward trend in the values of strain, strain rate and velocity, especially in the oldest group. Likewise, the ejection fraction (LVEF) and shortening fraction (LVFS) values, although preserved, also showed a decreased with age (p < 0.05). The analytical markers of cardiovascular risk and cardiac function showed no alterations. The molecular analyses revealed 16 individuals carrying pathogenic variants in NIPBL, two with variants in SMC1A, one with a variant in RAD21 and one with a HDAC8 variant. This is the first systematic approach that demonstrates that individuals with CdLS may present early cardiomyopathy, which can be detected by speckle tracking technique even before the appearance of clinical symptoms and the alteration of other echocardiographic or analytical parameters. For all these reasons, cardiological followup is suggested even in the absence of CHD, especially from adolescence onwards.

Deep learning-based facial image analysis in medical research: a systematic review protocol

INTRODUCTION

Deep learning techniques are gaining momentum in medical research. Evidence shows that deep learning has advantages over humans in image identification and classification, such as facial image analysis in detecting people's medical conditions. While positive findings are available, little is known about the state-of-the-art of deep learning-based facial image analysis in the medical context. For the consideration of patients' welfare and the development of the practice, a timely understanding of the challenges and opportunities faced by research on deep-learning-based facial image analysis is needed. To address this gap, we aim to conduct a systematic review to identify the characteristics and effects of deep learning-based facial image analysis in medical research. Insights gained from this systematic review will provide a much-needed understanding of the characteristics, challenges, as well as opportunities in deep learning-based facial image analysis applied in the contexts of disease detection, diagnosis and prognosis.

METHODS

Databases including PubMed, PsycINFO, CINAHL, IEEEXplore and Scopus will be searched for relevant studies published in English in September, 2021. Titles, abstracts and full-text articles will be screened to identify eligible articles. A manual search of the reference lists of the included articles will also be conducted. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses framework was adopted to guide the systematic review process. Two reviewers will independently examine the citations and select studies for inclusion. Discrepancies will be resolved by group discussions till a consensus is reached. Data will be extracted based on the research objective and selection criteria adopted in this study.

ETHICS AND DISSEMINATION

As the study is a protocol for a systematic review, ethical approval is not required. The study findings will be disseminated via peer-reviewed publications and conference presentations.

PROSPERO REGISTRATION NUMBER

CRD42020196473.

Clinical relevance of postzygotic mosaicism in Cornelia de Lange syndrome and purifying selection of NIPBL variants in blood

Postzygotic mosaicism (PZM) in NIPBL is a strong source of causality for Cornelia de Lange syndrome (CdLS) that can have major clinical implications. Here, we further delineate the role of somatic mosaicism in CdLS by describing a series of 11 unreported patients with mosaic disease-causing variants in NIPBL and performing a retrospective cohort study from a Spanish CdLS diagnostic center. By reviewing the literature and combining our findings with previously published data, we demonstrate a negative selection against somatic deleterious NIPBL variants in blood. Furthermore, the analysis of all reported cases indicates an unusual high prevalence of mosaicism in CdLS, occurring in 13.1% of patients with a positive molecular diagnosis. It is worth noting that most of the affected individuals with mosaicism have a clinical phenotype at least as severe as those with constitutive pathogenic variants. However, the type of genetic change does not vary between germline and somatic events and, even in the presence of mosaicism, missense substitutions are located preferentially within the HEAT repeat domain of NIPBL. In conclusion, the high prevalence of mosaicism in CdLS as well as the disparity in tissue distribution provide a novel orientation for the clinical management and genetic counselling of families.

Cornelia de Lange Syndrome: From a Disease to a Broader Spectrum

Cornelia de Lange syndrome (CdLS) is a genetic disease that exemplifies the evolution of knowledge in the field of rare genetic disorders. Originally described as a unique pattern of major and minor anomalies, over time this syndrome has been shown to be characterized by a significant variability of clinical expression. By increasing the number of patients described, knowledge of the natural history of the condition has been enriched with the demonstration of the relative frequency of various potential comorbidities. Since 2006, the discovery of CdLS's molecular basis has shown an equally vast genetic heterogeneity linked to the presence of variants in genes encoding for the cohesin complex pathway. The most recent clinical-genetic data led to the classification of the "original syndrome" into a "clinical spectrum" that foresees the presence of classic patients, of non-classic forms, and of conditions that show a modest phenotypic overlapping with the original disease. Finally, the knowledge of the molecular basis of the disease has allowed the development of basic research projects that could lay the foundations for the development of possible innovative pharmacological treatments.

ANKRD11 variants: KBG syndrome and beyond

Mutations affecting the transcriptional regulator Ankyrin Repeat Domain 11 (ANKRD11) are mainly associated with the multisystem developmental disorder known as KBG syndrome, but have also been identified in individuals with Cornelia de Lange syndrome (CdLS) and other developmental disorders caused by variants affecting different chromatin regulators. The extensive functional overlap of these proteins results in shared phenotypical features, which complicate the assessment of the clinical diagnosis. Additionally, re-evaluation of individuals at a later age occasionally reveals that the initial phenotype has evolved toward clinical features more reminiscent of a developmental disorder different from the one that was initially diagnosed. For this reason, variants in ANKRD11 can be ascribed to a broader class of disorders that fall within the category of the so-called chromatinopathies. In this work, we report on the clinical characterization of 23 individuals with variants in ANKRD11. The subjects present primarily with developmental delay, intellectual disability and dysmorphic features, and all but two received an initial clinical diagnosis of either KBG syndrome or CdLS. The number and the severity of the clinical signs are overlapping but variable and result in a broad spectrum of phenotypes, which could be partially accounted for by the presence of additional molecular diagnoses and distinct pathogenic mechanisms.

Craniofacial features of 3q29 deletion syndrome: Application of next-generation phenotyping technology

3q29 deletion syndrome (3q29del) is a recurrent deletion syndrome associated with neuropsychiatric disorders and congenital anomalies. Dysmorphic facial features have been described but not systematically characterized. This study aims to detail the 3q29del craniofacial phenotype and use a machine learning approach to categorize individuals with 3q29del through analysis of 2D photos. Detailed dysmorphology exam and 2D facial photos were ascertained from 31 individuals with 3q29del. Photos were used to train the next-generation phenotyping algorithm DeepGestalt (Face2Gene by FDNA, Inc, Boston, MA) to distinguish 3q29del cases from controls and all other recognized syndromes. Area under the curve of receiver operating characteristic curves (AUC-ROC) was used to determine the capacity of Face2Gene to identify 3q29del cases against controls. In this cohort, the most common observed craniofacial features were prominent forehead (48.4%), prominent nose tip (35.5%), and thin upper lip vermillion (25.8%). The FDNA technology showed an ability to distinguish cases from controls with an AUC-ROC value of 0.873 (p = 0.006) and led to the inclusion of 3q29del as one of the supported syndromes. This study found a recognizable facial pattern in 3q29del, as observed by trained clinical geneticists and next-generation phenotyping technology. These results expand the potential application of automated technology such as FDNA in identifying rare genetic syndromes, even when facial dysmorphology is subtle.

Targeted Gene Sequencing, Bone Health, and Body Composition in Cornelia de Lange Syndrome

The aim of this study was to evaluate bone health and body composition by dual-energy X-ray absorptiometry (DXA) in individuals with Cornelia de Lange Syndrome (CdLS). Overall, nine individuals with CdLS (five females, all Caucasian, aged 5–38 years) were assessed. Total body less head (TBLH) and lumbar spine (LS) scans were performed, and bone serum biomarkers were determined. Molecular analyses were carried out and clinical scores and skeletal features were assessed. Based on deep sequencing of a custom target gene panel, it was discovered that eight of the nine CdLS patients had potentially causative genetic variants in NIPBL. Fat and lean mass indices (FMI and LMI) were 3.4–11.1 and 8.4–17.0 kg/m2, respectively. For TBLH areal bone mineral density (aBMD), after adjusting for height for age Z-score of children and adolescents, two individuals (an adolescent and an adult) had low BMD (aBMD Z-scores less than –2.0 SD). Calcium, phosphorus, 25-OH-vitamin D, parathyroid hormone, and alkaline phosphatase levels were 2.08–2.49 nmol/L, 2.10–3.75 nmol/L, 39.94–78.37 nmol/L, 23.4–80.3 pg/mL, and 43–203 IU/L, respectively. Individuals with CdLS might have normal adiposity and low levels of lean mass measured with DXA. Bone health in this population seems to be less of a concern during childhood and adolescence. However, they might be at risk for impaired bone health due to low aBMD in adulthood.

Efficiency of Computer-Aided Facial Phenotyping (DeepGestalt) in Individuals With and Without a Genetic Syndrome: Diagnostic Accuracy Study

Background

Collectively, an estimated 5% of the population have a genetic disease. Many of them feature characteristics that can be detected by facial phenotyping. Face2Gene CLINIC is an online app for facial phenotyping of patients with genetic syndromes. DeepGestalt, the neural network driving Face2Gene, automatically prioritizes syndrome suggestions based on ordinary patient photographs, potentially improving the diagnostic process. Hitherto, studies on DeepGestalt’s quality highlighted its sensitivity in syndromic patients. However, determining the accuracy of a diagnostic methodology also requires testing of negative controls.

Objective

The aim of this study was to evaluate DeepGestalt's accuracy with photos of individuals with and without a genetic syndrome. Moreover, we aimed to propose a machine learning–based framework for the automated differentiation of DeepGestalt’s output on such images.

Methods

Frontal facial images of individuals with a diagnosis of a genetic syndrome (established clinically or molecularly) from a convenience sample were reanalyzed. Each photo was matched by age, sex, and ethnicity to a picture featuring an individual without a genetic syndrome. Absence of a facial gestalt suggestive of a genetic syndrome was determined by physicians working in medical genetics. Photos were selected from online reports or were taken by us for the purpose of this study. Facial phenotype was analyzed by DeepGestalt version 19.1.7, accessed via Face2Gene CLINIC. Furthermore, we designed linear support vector machines (SVMs) using Python 3.7 to automatically differentiate between the 2 classes of photographs based on DeepGestalt's result lists.

Results

We included photos of 323 patients diagnosed with 17 different genetic syndromes and matched those with an equal number of facial images without a genetic syndrome, analyzing a total of 646 pictures. We confirm DeepGestalt’s high sensitivity (top 10 sensitivity: 295/323, 91%). DeepGestalt’s syndrome suggestions in individuals without a craniofacially dysmorphic syndrome followed a nonrandom distribution. A total of 17 syndromes appeared in the top 30 suggestions of more than 50% of nondysmorphic images. DeepGestalt’s top scores differed between the syndromic and control images (area under the receiver operating characteristic [AUROC] curve 0.72, 95% CI 0.68-0.76; P<.001). A linear SVM running on DeepGestalt’s result vectors showed stronger differences (AUROC 0.89, 95% CI 0.87-0.92; P<.001).

Conclusions

DeepGestalt fairly separates images of individuals with and without a genetic syndrome. This separation can be significantly improved by SVMs running on top of DeepGestalt, thus supporting the diagnostic process of patients with a genetic syndrome. Our findings facilitate the critical interpretation of DeepGestalt’s results and may help enhance it and similar computer-aided facial phenotyping tools.

A novel mosaic variant on SMC1A reported in buccal mucosa cells, albeit not in blood, of a patient with Cornelia de Lange-like presentation

Mosaicism in Cornelia de Lange syndrome (CdLS) has been reported in clinically diagnosed CdLS patients with negative molecular testing using blood as the specimen, particularly in the NIPBL gene. Here we report a novel mosaic variant in SMC1A identified in the buccal swab DNA of a patient with a mild CdLS phenotype. Our patient presented with global developmental delay, dysmorphic features, microcephaly, and short stature, with no limb defect. Face2Gene, a digital tool that analyzes facial morphology, demonstrated a 97% match between our patient and the CdLS gestalt. An initial next-generation sequencing (NGS)-based CdLS panel test, including NIPBL, HDAC8, RAD21, SMC1A, and SMC3, completed using DNA isolated from leukocytes, was negative, and subsequent trio exome sequencing was nondiagnostic. The exome identified biallelic variants of uncertain significance in a candidate gene, NSMCE2. In the pursuit of a molecular diagnosis, a second NGS-based CdLS panel test was ordered on a buccal swab specimen and a novel variant, c.793_795delGAG (p.Glu265del) in SMC1A, was detected at 60% mosaicism. Retrospective analysis of the former panel and exome data revealed the SMC1A variant at 4% and 2%, respectively, both far below standard reporting thresholds. Given that mosaicism has been frequently reported in CdLS, we suggest selecting a different tissue for testing in clinically suspected CdLS cases, even after negative molecular results via blood specimen.

Pitt-Hopkins Syndrome: Clinical and Molecular Findings of a 5-Year-Old Patient

Pitt Hopkins syndrome (PTHS) is a very rare condition and until now, approximately 500 patients were reported worldwide, of which not all are genetically confirmed. Usually, individuals with variants affecting exons 1 to 5 in the TCF4 gene associate mild intellectual disability (ID), between exons 5 to 8, moderate to severe ID and sometimes have some of the characteristics of PTHS, and variants starting from exon 9 to exon 20 associate a typical PTHS phenotype. In this report, we describe the clinical and molecular findings of a Caucasian boy diagnosed with PTHS. PTHS phenotype is described including craniofacial dysmorphism with brachycephaly, biparietal narrowing, wide nasal bridge, thin and linear lateral eyebrows, palpebral edema, full cheeks, short philtrum, wide mouth with prominent and everted lips, prominent Cupid’s bow, downturned corners of the mouth, microdontia and also the clinical management of the patient. The previously and the current diagnosis scores are described in this report and also the challenges and their benefits for an accurate and early diagnosis.