The point-of-care use of a facial phenotyping tool in the genetics clinic: Enhancing diagnosis and education with machine learning

The Face and Features of RNU4-2: A New, Common, Recognizable, Yet Hidden Neurodevelopmental Disorder

BACKGROUND

RNU4-2 is a newly identified, noncoding gene responsible for a significant proportion of individuals with neurodevelopmental disorders (NDDs). Diagnosis is hampered by the inability of commonly employed clinical testing methods, including exome sequencing and currently formulated multigene panels, to detect variants in the noncoding region. The relatively high prevalence of this condition, predicted to affect thousands of undiagnosed children with NDDs, makes it even more relevant to have better tools to facilitate diagnosis. The initial report of the gene-disease association outlined aggregate phenotypic features but lacked detailed patient evaluations, potentially under-reporting phenotypic features and failing to highlight unique aspects. We aimed to identify individuals with RNU4-2 gene variants to deeply phenotype the clinical profile. We sought to define key features that may suggest the diagnosis, to highlight individuals for whom specialized testing, able to detect noncoding region variants, may be indicated.

METHODS

We reviewed genomic data from 6,734 individuals, identifying five with recurrent de novo RNU4-2 (n.64_65insT) variants. We clinically evaluated four. Findings were compared with those previously reported.

RESULTS

We identify common clinical features, a distinctive dysmorphic facial pattern, and shared imaging abnormalities. We describe novel aspects including longitudinal trajectory and treatment response.

CONCLUSIONS

Enhanced recognition of the RNU4-2 (n.64_65insT-common variant) phenotype, particularly the dysmorphic facial features, will facilitate earlier diagnosis. Distinctive characteristics will guide the selection of patients for testing able to detect RNU4-2 variants: genome sequencing or targeted gene testing. Furthermore, health and research systems may identify undiagnosed patients by querying databases for individuals exhibiting the traits described herein.

Health Care Professionals' Experience of Using AI: Systematic Review With Narrative Synthesis

BACKGROUND

There has been a substantial increase in the development of artificial intelligence (AI) tools for clinical decision support. Historically, these were mostly knowledge-based systems, but recent advances include non-knowledge-based systems using some form of machine learning. The ability of health care professionals to trust technology and understand how it benefits patients or improves care delivery is known to be important for their adoption of that technology. For non-knowledge-based AI tools for clinical decision support, these issues are poorly understood.

OBJECTIVE

The aim of this study is to qualitatively synthesize evidence on the experiences of health care professionals in routinely using non-knowledge-based AI tools to support their clinical decision-making.

METHODS

In June 2023, we searched 4 electronic databases, MEDLINE, Embase, CINAHL, and Web of Science, with no language or date limit. We also contacted relevant experts and searched reference lists of the included studies. We included studies of any design that reported the experiences of health care professionals using non-knowledge-based systems for clinical decision support in their work settings. We completed double independent quality assessment for all included studies using the Mixed Methods Appraisal Tool. We used a theoretically informed thematic approach to synthesize the findings.

RESULTS

After screening 7552 titles and 182 full-text articles, we included 25 studies conducted in 9 different countries. Most of the included studies were qualitative (n=13), and the remaining were quantitative (n=9) and mixed methods (n=3). Overall, we identified 7 themes: health care professionals' understanding of AI applications, level of trust and confidence in AI tools, judging the value added by AI, data availability and limitations of AI, time and competing priorities, concern about governance, and collaboration to facilitate the implementation and use of AI. The most frequently occurring are the first 3 themes. For example, many studies reported that health care professionals were concerned about not understanding the AI outputs or the rationale behind them. There were issues with confidence in the accuracy of the AI applications and their recommendations. Some health care professionals believed that AI provided added value and improved decision-making, and some reported that it only served as a confirmation of their clinical judgment, while others did not find it useful at all.

CONCLUSIONS

Our review identified several important issues documented in various studies on health care professionals' use of AI tools in real-world health care settings. Opinions of health care professionals regarding the added value of AI tools for supporting clinical decision-making varied widely, and many professionals had concerns about their understanding of and trust in this technology. The findings of this review emphasize the need for concerted efforts to optimize the integration of AI tools in real-world health care settings.

TRIAL REGISTRATION

PROSPERO CRD42022336359; https://tinyurl.com/2yunvkmb.

Effects of artificial intelligence implementation on efficiency in medical imaging-a systematic literature review and meta-analysis

In healthcare, integration of artificial intelligence (AI) holds strong promise for facilitating clinicians' work, especially in clinical imaging. We aimed to assess the impact of AI implementation for medical imaging on efficiency in real-world clinical workflows and conducted a systematic review searching six medical databases. Two reviewers double-screened all records. Eligible records were evaluated for methodological quality. The outcomes of interest were workflow adaptation due to AI implementation, changes in time for tasks, and clinician workload. After screening 13,756 records, we identified 48 original studies to be incuded in the review. Thirty-three studies measured time for tasks, with 67% reporting reductions. Yet, three separate meta-analyses of 12 studies did not show significant effects after AI implementation. We identified five different workflows adapting to AI use. Most commonly, AI served as a secondary reader for detection tasks. Alternatively, AI was used as the primary reader for identifying positive cases, resulting in reorganizing worklists or issuing alerts. Only three studies scrutinized workload calculations based on the time saved through AI use. This systematic review and meta-analysis represents an assessment of the efficiency improvements offered by AI applications in real-world clinical imaging, predominantly revealing enhancements across the studies. However, considerable heterogeneity in available studies renders robust inferences regarding overall effectiveness in imaging tasks. Further work is needed on standardized reporting, evaluation of system integration, and real-world data collection to better understand the technological advances of AI in real-world healthcare workflows. Systematic review registration: Prospero ID CRD42022303439, International Registered Report Identifier (IRRID): RR2-10.2196/40485.

Automated fingerprint analysis as a diagnostic tool for the genetic disorder Kabuki syndrome

Purpose

Emerging therapeutic strategies for Kabuki syndrome (KS) make early diagnosis critical. Fingerprint analysis as a diagnostic aid for KS diagnosis could facilitate early diagnosis and expand the current patient base for clinical trials and natural history studies.

Method

Fingerprints of 74 individuals with KS, 1 individual with a KS-like phenotype, and 108 controls were collected through a mobile app. KS fingerprint patterns were studied using logistic regression and a convolutional neural network to differentiate KS individuals from controls.

Results

Our analysis identified 2 novel KS metrics (folding finger ridge count and simple pattern), which significantly differentiated KS fingerprints from controls, producing an area under the receiver operating characteristic curve value of 0.82 [0.75; 0.89] and a likelihood ratio of 9.0. This metric showed a sensitivity of 35.6% [23.73%; 47.46%] and a specificity of 96.04% [92.08%; 99.01%]. An independent artificial intelligence convolutional neural network classification-based method validated this finding and yielded comparable results, with a likelihood ratio of 8.7, sensitivity of 76.6%, and specificity of 91.2%.

Conclusion

Our findings suggest that automatic fingerprint analysis can have diagnostic use for KS and possible future utility for diagnosing other genetic disorders, enabling greater access to genetic diagnosis in areas with limited availability of genetic testing.

Genetically guided precision medicine clinical decision support tools: a systematic review

OBJECTIVES

Patient care using genetics presents complex challenges. Clinical decision support (CDS) tools are a potential solution because they provide patient-specific risk assessments and/or recommendations at the point of care. This systematic review evaluated the literature on CDS systems which have been implemented to support genetically guided precision medicine (GPM).

MATERIALS AND METHODS

A comprehensive search was conducted in MEDLINE and Embase, encompassing January 1, 2011-March 14, 2023. The review included primary English peer-reviewed research articles studying humans, focused on the use of computers to guide clinical decision-making and delivering genetically guided, patient-specific assessments, and/or recommendations to healthcare providers and/or patients.

RESULTS

The search yielded 3832 unique articles. After screening, 41 articles were identified that met the inclusion criteria. Alerts and reminders were the most common form of CDS used. About 27 systems were integrated with the electronic health record; 2 of those used standards-based approaches for genomic data transfer. Three studies used a framework to analyze the implementation strategy.

DISCUSSION

Findings include limited use of standards-based approaches for genomic data transfer, system evaluations that do not employ formal frameworks, and inconsistencies in the methodologies used to assess genetic CDS systems and their impact on patient outcomes.

CONCLUSION

We recommend that future research on CDS system implementation for genetically GPM should focus on implementing more CDS systems, utilization of standards-based approaches, user-centered design, exploration of alternative forms of CDS interventions, and use of formal frameworks to systematically evaluate genetic CDS systems and their effects on patient care.

Development and validation of a point-of-care nursing mobile tool to guide the diagnosis of malnutrition in hospitalized adult patients: a multicenter, prospective cohort study

Malnutrition is a prevalent and severe issue in hospitalized patients with chronic diseases. However, malnutrition screening is often overlooked or inaccurate due to lack of awareness and experience among health care providers. This study aimed to develop and validate a novel digital smartphone-based self-administered tool that uses facial features, especially the ocular area, as indicators of malnutrition in inpatient patients with chronic diseases. Facial photographs and malnutrition screening scales were collected from 619 patients in four different hospitals. A machine learning model based on back propagation neural network was trained, validated, and tested using these data. The model showed a significant correlation (p < 0.05) and a high accuracy (area under the curve 0.834-0.927) in different patient groups. The point-of-care mobile tool can be used to screen malnutrition with good accuracy and accessibility, showing its potential for screening malnutrition in patients with chronic diseases.

Genetic counselors' utilization of ChatGPT in professional practice: A cross-sectional study

PURPOSE

The precision medicine era has seen increased utilization of artificial intelligence (AI) in the field of genetics. We sought to explore the ways that genetic counselors (GCs) currently use the publicly accessible AI tool Chat Generative Pre-trained Transformer (ChatGPT) in their work.

METHODS

GCs in North America were surveyed about how ChatGPT is used in different aspects of their work. Descriptive statistics were reported through frequencies and means.

RESULTS

Of 118 GCs who completed the survey, 33.8% (40) reported using ChatGPT in their work; 47.5% (19) use it in clinical practice, 35% (14) use it in education, and 32.5% (13) use it in research. Most GCs (62.7%; 74) felt that it saves time on administrative tasks but the majority (82.2%; 97) felt that a paramount challenge was the risk of obtaining incorrect information. The majority of GCs not using ChatGPT (58.9%; 46) felt it was not necessary for their work.

CONCLUSION

A considerable number of GCs in the field are using ChatGPT in different ways, but it is primarily helpful with tasks that involve writing. It has potential to streamline workflow issues encountered in clinical genetics, but practitioners need to be informed and uniformly trained about its limitations.

Validation of 3 Computer-Aided Facial Phenotyping Tools (DeepGestalt, GestaltMatcher, and D-Score): Comparative Diagnostic Accuracy Study

BACKGROUND

While characteristic facial features provide important clues for finding the correct diagnosis in genetic syndromes, valid assessment can be challenging. The next-generation phenotyping algorithm DeepGestalt analyzes patient images and provides syndrome suggestions. GestaltMatcher matches patient images with similar facial features. The new D-Score provides a score for the degree of facial dysmorphism.

OBJECTIVE

We aimed to test state-of-the-art facial phenotyping tools by benchmarking GestaltMatcher and D-Score and comparing them to DeepGestalt.

METHODS

Using a retrospective sample of 4796 images of patients with 486 different genetic syndromes (London Medical Database, GestaltMatcher Database, and literature images) and 323 inconspicuous control images, we determined the clinical use of D-Score, GestaltMatcher, and DeepGestalt, evaluating sensitivity; specificity; accuracy; the number of supported diagnoses; and potential biases such as age, sex, and ethnicity.

RESULTS

DeepGestalt suggested 340 distinct syndromes and GestaltMatcher suggested 1128 syndromes. The top-30 sensitivity was higher for DeepGestalt (88%, SD 18%) than for GestaltMatcher (76%, SD 26%). DeepGestalt generally assigned lower scores but provided higher scores for patient images than for inconspicuous control images, thus allowing the 2 cohorts to be separated with an area under the receiver operating characteristic curve (AUROC) of 0.73. GestaltMatcher could not separate the 2 classes (AUROC 0.55). Trained for this purpose, D-Score achieved the highest discriminatory power (AUROC 0.86). D-Score's levels increased with the age of the depicted individuals. Male individuals yielded higher D-scores than female individuals. Ethnicity did not appear to influence D-scores.

CONCLUSIONS

If used with caution, algorithms such as D-score could help clinicians with constrained resources or limited experience in syndromology to decide whether a patient needs further genetic evaluation. Algorithms such as DeepGestalt could support diagnosing rather common genetic syndromes with facial abnormalities, whereas algorithms such as GestaltMatcher could suggest rare diagnoses that are unknown to the clinician in patients with a characteristic, dysmorphic face.

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.

Trust and acceptability of data-driven clinical recommendations in everyday practice: A scoping review

BACKGROUND

Increasing attention is being given to the analysis of large health datasets to derive new clinical decision support systems (CDSS). However, few data-driven CDSS are being adopted into clinical practice. Trust in these tools is believed to be fundamental for acceptance and uptake but to date little attention has been given to defining or evaluating trust in clinical settings.

OBJECTIVES

A scoping review was conducted to explore how and where acceptability and trustworthiness of data-driven CDSS have been assessed from the health professional's perspective.

METHODS

Medline, Embase, PsycInfo, Web of Science, Scopus, ACM Digital, IEEE Xplore and Google Scholar were searched in March 2022 using terms expanded from: "data-driven" AND "clinical decision support" AND "acceptability". Included studies focused on healthcare practitioner-facing data-driven CDSS, relating directly to clinical care. They included trust or a proxy as an outcome, or in the discussion. The preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews (PRISMA-ScR) is followed in the reporting of this review.

RESULTS

3291 papers were screened, with 85 primary research studies eligible for inclusion. Studies covered a diverse range of clinical specialisms and intended contexts, but hypothetical systems (24) outnumbered those in clinical use (18). Twenty-five studies measured trust, via a wide variety of quantitative, qualitative and mixed methods. A further 24 discussed themes of trust without it being explicitly evaluated, and from these, themes of transparency, explainability, and supporting evidence were identified as factors influencing healthcare practitioner trust in data-driven CDSS.

CONCLUSION

There is a growing body of research on data-driven CDSS, but few studies have explored stakeholder perceptions in depth, with limited focused research on trustworthiness. Further research on healthcare practitioner acceptance, including requirements for transparency and explainability, should inform clinical implementation.

Krawitz PDMDPP. The future role of facial image analysis in ACMG classification guidelines

The use of next-generation sequencing (NGS) has dramatically improved the diagnosis of rare diseases. However, the analysis of genomic data has become complex with the increasing detection of variants by exome and genome sequencing. The American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) developed a 5-tier classification scheme in 2015 for variant interpretation, that has since been widely adopted. Despite efforts to minimise discrepancies in the application of these criteria, inconsistencies still occur. Further specifications for individual genes were developed by Variant Curation Expert Panels (VCEPs) of the Clinical Genome Resource (ClinGen) consortium, that also take into consideration gene or disease specific features. For instance, in disorders with a highly characerstic facial gestalt a "phenotypic match" (PP4) has higher pathogenic evidence than e.g. in a non-syndromic form of intellectual disability. With computational approaches for quantifying the similarity of dysmorphic features results of such analysis can now be used in a refined Bayesian framework for the ACMG/AMP criteria.

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.

Enhancing Molecular Testing for Effective Delivery of Actionable Gene Diagnostics

There is a deep need to navigate within our genomic data to find, understand and pave the way for disease-specific treatments, as the clinical diagnostic journey provides only limited guidance. The human genome is enclosed in every nucleated cell, and yet at the single-cell resolution many unanswered questions remain, as most of the sequencing techniques use a bulk approach. Therefore, heterogeneity, mosaicism and many complex structural variants remain partially uncovered. As a conceptual approach, nanopore-based sequencing holds the promise of being a single-molecule-based, long-read and high-resolution technique, with the ability of uncovering the nucleic acid sequence and methylation almost in real time. A key limiting factor of current clinical genetics is the deciphering of key disease-causing genomic sequences. As the technological revolution is expanding regarding genetic data, the interpretation of genotype-phenotype correlations should be made with fine caution, as more and more evidence points toward the presence of more than one pathogenic variant acting together as a result of intergenic interplay in the background of a certain phenotype observed in a patient. This is in conjunction with the observation that many inheritable disorders manifest in a phenotypic spectrum, even in an intra-familial way. In the present review, we summarized the relevant data on nanopore sequencing regarding clinical genomics as well as highlighted the importance and content of pre-test and post-test genetic counselling, yielding a complex approach to phenotype-driven molecular diagnosis. This should significantly lower the time-to-right diagnosis as well lower the time required to complete a currently incomplete genotype-phenotype axis, which will boost the chance of establishing a new actionable diagnosis followed by therapeutical approach.

Facial features of lysosomal storage disorders

INTRODUCTION

The use of facial recognition technology has diversified the diagnostic toolbelt for clinicians and researchers for the accurate diagnoses of patients with rare and challenging disorders. Specific identifiers in patient images can be grouped using artificial intelligence to allow the recognition of diseases and syndromes with similar features. Lysosomal storage disorders are rare, and some have prominent and unique features that may be used to train the accuracy of facial recognition software algorithms. Noteworthy features of lysosomal storage disorders (LSDs) include facial features such as prominent brows, wide noses, thickened lips, mouth, and chin, resulting in coarse and rounded facial features.

AREAS COVERED

We evaluated and report the prevalence of facial phenotypes in patients with different LSDs, noting two current examples when artificial intelligence strategies have been utilized to identify distinctive facies.

EXPERT OPINION

Specific LSDs, including Gaucher disease, Mucolipidosis IV and Fabry disease have recently been distinguished using facial recognition software. Additional lysosomal disorders LSDs lysosomal storage disorders with unique and distinguishable facial features also merit evaluation using this technology. These tools may ultimately aid in the identification of specific LSDs and shorten the diagnostic odyssey for patients with these rare and under-recognized disorders.

First Italian experience using the automated craniofacial gestalt analysis on a cohort of pediatric patients with multiple anomaly syndromes

BACKGROUND

In this study, we used the novel DeepGestalt technology powered by Face2Gene (FDNA Inc., MA, USA) in suggesting a correct diagnosis based on the facial gestalt of well-known multiple anomaly syndromes. Only molecularly characterized pediatric patients were considered in the present research.

SUBJECTS AND METHODS

A total of 19 two-dimensional (2D) images of patients affected by several molecularly confirmed craniofacial syndromes (14 monogenic disorders and 5 chromosome diseases) and evaluated at the main involved Institution were analyzed using the Face2Gene CLINIC application (vs.19.1.3). Patients were cataloged into two main analysis groups (A, B) according to the number of clinical evaluations. Specifically, group A contained the patients evaluated more than one time, while in group B were comprised the subjects with a single clinical assesment. The algorithm's reliability was measured based on its capacity to identify the correct diagnosis as top-1 match, within the top-10 match and top-30 matches, only based on the uploaded image and not any other clinical finding or HPO terms. Failure was represented by the top-0 match.

RESULTS

The correct diagnosis was suggested respectively in 100% (8/8) and 81% (9/11) of cases of group A and B, globally failing in 16% (3/19).

CONCLUSION

The tested tool resulted to be useful in identifying the facial gestalt of a heterogeneous group of syndromic disorders. This study illustrates the first Italian experience with the next generation phenotyping technology, following previous works and providing additional observations.

Relationships between Nursing Students' Skill Mastery, Test Anxiety, Self-Efficacy, and Facial Expressions: A Preliminary Observational Study

Test anxiety and self-efficacy significantly influence the mastery of nursing skills. Facial expression recognition tools are central components to recognising these elements. This study investigated the frequent facial expressions conveyed by nursing students and examined the relationships between nursing skill mastery, test anxiety, self-efficacy, and facial expressions in a test-taking situation. Thirty-three second-year nursing students who were attending a university in a Korean metropolitan city participated. Test anxiety, self-efficacy, and facial expressions were collected while the students inserted indwelling catheters. Using Microsoft Azure software, the researchers examined the students' facial expressions. Negative facial expressions, such as anger, disgust, sadness, and surprise, were more common during the test-taking situation than the practice trial. Fear was positively correlated with anxiety. None of the facial expressions had significant relationships with self-efficacy; however, disgust was positively associated with nursing skill mastery. The facial expressions during the practice and test-taking situations were similar; however, fear and disgust may have been indicators of test anxiety and skill mastery. To create a screening tool for detecting and caring for students' emotions, further studies should explore students' facial expressions that were not evaluated in this study.

The utility of DNA methylation signatures in directing genome sequencing workflow: Kabuki syndrome and CDK13-related disorder

Kabuki syndrome (KS) is a neurodevelopmental disorder characterized by hypotonia, intellectual disability, skeletal anomalies, and postnatal growth restriction. The characteristic facial appearance is not pathognomonic for KS as several other conditions demonstrate overlapping features. For 20‐30% of children with a clinical diagnosis of KS, no causal variant is identified by conventional genetic testing of the two associated genes, KMT2D and KDM6A. Here, we describe two cases of suspected KS that met clinical diagnostic criteria and had a high gestalt match on the artificial intelligence platform Face2Gene. Although initial KS testing was negative, genome‐wide DNA methylation (DNAm) was instrumental in guiding genome sequencing workflow to establish definitive molecular diagnoses. In one case, a positive DNAm signature for KMT2D led to the identification of a cryptic variant in KDM6A by genome sequencing; for the other case, a DNAm signature different from KS led to the detection of another diagnosis in the KS differential, CDK13‐related disorder. This approach illustrates the clinical utility of DNAm signatures in the diagnostic workflow for the genome analyst or clinical geneticist—especially for disorders with overlapping clinical phenotypes.

Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition

Artificial intelligence (AI) technology is widely applied in different medical fields, including the diagnosis of various diseases on the basis of facial phenotypes, but there is no evaluation or quantitative synthesis regarding the performance of artificial intelligence. Here, for the first time, we summarized and quantitatively analyzed studies on the diagnosis of heterogeneous diseases on the basis on facial features. In pooled data from 20 systematically identified studies involving 7 single diseases and 12,557 subjects, quantitative random-effects models revealed a pooled sensitivity of 89% (95% CI 82% to 93%) and a pooled specificity of 92% (95% CI 87% to 95%). A new index, the facial recognition intensity (FRI), was established to describe the complexity of the association of diseases with facial phenotypes. Meta-regression revealed the important contribution of FRI to heterogeneous diagnostic accuracy (p = 0.021), and a similar result was found in subgroup analyses (p = 0.003). An appropriate increase in the training size and the use of deep learning models helped to improve the diagnostic accuracy for diseases with low FRI, although no statistically significant association was found between accuracy and photographic resolution, training size, AI architecture, and number of diseases. In addition, a novel hypothesis is proposed for universal rules in AI performance, providing a new idea that could be explored in other AI applications.

Development and evaluation of a machine learning-based point-of-care screening tool for genetic syndromes in children: a multinational retrospective study

BACKGROUND

Delays in the diagnosis of genetic syndromes are common, particularly in low and middle-income countries with limited access to genetic screening services. We, therefore, aimed to develop and evaluate a machine learning-based screening technology using facial photographs to evaluate a child's risk of presenting with a genetic syndrome for use at the point of care.

METHODS

In this retrospective study, we developed a facial deep phenotyping technology based on deep neural networks and facial statistical shape models to screen children for genetic syndromes. We trained the machine learning models on facial photographs from children (aged <21 years) with a clinical or molecular diagnosis of a genetic syndrome and controls without a genetic syndrome matched for age, sex, and race or ethnicity. Images were obtained from three publicly available databases (the Atlas of Human Malformations in Diverse Populations of the National Human Genome Research Institute, Face2Gene, and the dataset available from Ferry and colleagues) and the archives of the Children's National Hospital (Washington, DC, USA), in addition to photographs taken on a standard smartphone at the Children's National Hospital. We designed a deep learning architecture structured into three neural networks, which performed image standardisation (Network A), facial morphology detection (Network B), and genetic syndrome risk estimation, accounting for phenotypic variations due to age, sex, and race or ethnicity (Network C). Data were divided randomly into 40 groups for cross validation, and the performance of the model was evaluated in terms of accuracy, sensitivity, and specificity in both the total population and stratified by race or ethnicity, age, and sex.

FINDINGS

Our dataset included 2800 facial photographs of children (1318 [47%] female and 1482 [53%] male; 1576 [56%] White, 432 [15%] African, 430 [15%] Hispanic, and 362 [13%] Asian). 1400 children with 128 genetic conditions were included (the most prevalent being Williams-Beuren syndrome [19%], Cornelia de Lange syndrome [17%], Down syndrome [16%], 22q11.2 deletion [13%], and Noonan syndrome [12%] syndrome) in addition to 1400 photographs of matched controls. In the total population, our deep learning-based model had an accuracy of 88% (95% CI 87-89) for the detection of a genetic syndrome, with 90% sensitivity (95% CI 88-92) and 86% specificity (95% CI 84-88). Accuracy was greater in White (90%, 89-91) and Hispanic populations (91%, 88-94) than in African (84%, 81-87) and Asian populations (82%, 78-86). Accuracy was also similar in male (89%, 87-91) and female children (87%, 85-89), and similar in children younger than 2 years (86%, 84-88) and children aged 2 years or older (eg, 89% [87-91] for those aged 2 years to <5 years).

INTERPRETATION

This genetic screening technology could support early risk stratification at the point of care in global populations, which has the potential accelerate diagnosis and reduce mortality and morbidity through preventive care.

FUNDING

Children's National Hospital and Government of Abu Dhabi.