Search

The most important factor that complicates the work of dysmorphologists is the significant phenotypic variability of the human face. Next-Generation Phenotyping (NGP) tools that assist clinicians with recognizing characteristic syndromic patterns are particularly challenged when confronted with patients from populations different from their training data. To that end, we systematically analyzed the impact of genetic ancestry on facial dysmorphism. For that purpose, we established the GestaltMatcher Database (GMDB) as a reference dataset for medical images of patients with rare genetic disorders from around the world. We collected 10,980 frontal facial images - more than a quarter previously unpublished - from 8,346 patients, representing 581 rare disorders. Although the predominant ancestry is still European (67%), data from underrepresented populations have been increased considerably via global collaborations (19% Asian and 7% African). This includes previously unpublished reports for more than 40% of the African patients. The NGP analysis on this diverse dataset revealed characteristic performance differences depending on the composition of training and test sets corresponding to genetic relatedness. For clinical use of NGP, incorporating non-European patients resulted in a profound enhancement of GestaltMatcher performance. The top-5 accuracy rate increased by +11.29%. Importantly, this improvement in delineating the correct disorder from a facial portrait was achieved without decreasing the performance on European patients. By design, GMDB complies with the FAIR principles by rendering the curated medical data findable, accessible, interoperable, and reusable. This means GMDB can also serve as data for training and benchmarking. In summary, our study on facial dysmorphism on a global sample revealed a considerable cross ancestral phenotypic variability confounding NGP that should be counteracted by international efforts for increasing data diversity. GMDB will serve as a vital reference database for clinicians and a transparent training set for advancing NGP technology.

Large language models (LLMs) are generating interest in medical settings. For example, LLMs can respond coherently to medical queries by providing plausible differential diagnoses based on clinical notes. However, there are many questions to explore, such as evaluating differences between open- and closed-source LLMs as well as LLM performance on queries from both medical and non-medical users. In this study, we assessed multiple LLMs, including Llama-2-chat, Vicuna, Medllama2, Bard/Gemini, Claude, ChatGPT3.5, and ChatGPT-4, as well as non-LLM approaches (Google search and Phenomizer) regarding their ability to identify genetic conditions from textbook-like clinician questions and their corresponding layperson translations related to 63 genetic conditions. For open-source LLMs, larger models were more accurate than smaller LLMs: 7b, 13b, and larger than 33b parameter models obtained accuracy ranges from 21%-49%, 41%-51%, and 54%-68%, respectively. Closed-source LLMs outperformed open-source LLMs, with ChatGPT-4 performing best (89%-90%). Three of 11 LLMs and Google search had significant performance gaps between clinician and layperson prompts. We also evaluated how in-context prompting and keyword removal affected open-source LLM performance. Models were provided with 2 types of in-context prompts: list-type prompts, which improved LLM performance, and definition-type prompts, which did not. We further analyzed removal of rare terms from descriptions, which decreased accuracy for 5 of 7 evaluated LLMs. Finally, we observed much lower performance with real individuals' descriptions; LLMs answered these questions with a maximum 21% accuracy., Competing Interests: Declaration of interests B.D.S. is the co-editor-in-chief of the American Journal of Medical Genetics and receives textbook royalties from Wiley., (Published by Elsevier Inc.)

Artificial intelligence (AI) is increasingly used in genomics research and practice, and generative AI has garnered significant recent attention. In clinical applications of generative AI, aspects of the underlying datasets can impact results, and confounders should be studied and mitigated. One example involves the facial expressions of people with genetic conditions. Stereotypically, Williams (WS) and Angelman (AS) syndromes are associated with a "happy" demeanor, including a smiling expression. Clinical geneticists may be more likely to identify these conditions in images of smiling individuals. To study the impact of facial expression, we analyzed publicly available facial images of approximately 3500 individuals with genetic conditions. Using a deep learning (DL) image classifier, we found that WS and AS images with non-smiling expressions had significantly lower prediction probabilities for the correct syndrome labels than those with smiling expressions. This was not seen for 22q11.2 deletion and Noonan syndromes, which are not associated with a smiling expression. To further explore the effect of facial expressions, we computationally altered the facial expressions for these images. We trained HyperStyle, a GAN-inversion technique compatible with StyleGAN2, to determine the vector representations of our images. Then, following the concept of InterfaceGAN, we edited these vectors to recreate the original images in a phenotypically accurate way but with a different facial expression. Through online surveys and an eye-tracking experiment, we examined how altered facial expressions affect the performance of human experts. We overall found that facial expression is associated with diagnostic accuracy variably in different genetic conditions., (© The Author(s) 2024. Published by Oxford University Press.)

The most important factor that complicates the work of dysmorphologists is the significant phenotypic variability of the human face. Next-Generation Phenotyping (NGP) tools that assist clinicians with recognizing characteristic syndromic patterns are particularly challenged when confronted with patients from populations different from their training data. To that end, we systematically analyzed the impact of genetic ancestry on facial dysmorphism. For that purpose, we established the GestaltMatcher Database (GMDB) as a reference dataset for medical images of patients with rare genetic disorders from around the world. We collected 10,980 frontal facial images - more than a quarter previously unpublished - from 8,346 patients, representing 581 rare disorders. Although the predominant ancestry is still European (67%), data from underrepresented populations have been increased considerably via global collaborations (19% Asian and 7% African). This includes previously unpublished reports for more than 40% of the African patients. The NGP analysis on this diverse dataset revealed characteristic performance differences depending on the composition of training and test sets corresponding to genetic relatedness. For clinical use of NGP, incorporating non-European patients resulted in a profound enhancement of GestaltMatcher performance. The top-5 accuracy rate increased by +11.29%. Importantly, this improvement in delineating the correct disorder from a facial portrait was achieved without decreasing the performance on European patients. By design, GMDB complies with the FAIR principles by rendering the curated medical data findable, accessible, interoperable, and reusable. This means GMDB can also serve as data for training and benchmarking. In summary, our study on facial dysmorphism on a global sample revealed a considerable cross ancestral phenotypic variability confounding NGP that should be counteracted by international efforts for increasing data diversity. GMDB will serve as a vital reference database for clinicians and a transparent training set for advancing NGP technology.

Importance: The lack of standardized genetics training in pediatrics residencies, along with a shortage of medical geneticists, necessitates innovative educational approaches., Objective: To compare pediatric resident recognition of Kabuki syndrome (KS) and Noonan syndrome (NS) after 1 of 4 educational interventions, including generative artificial intelligence (AI) methods., Design, Setting, and Participants: This comparative effectiveness study used generative AI to create images of children with KS and NS. From October 1, 2022, to February 28, 2023, US pediatric residents were provided images through a web-based survey to assess whether these images helped them recognize genetic conditions., Interventions: Participants categorized 20 images after exposure to 1 of 4 educational interventions (text-only descriptions, real images, and 2 types of images created by generative AI)., Main Outcomes and Measures: Associations between educational interventions with accuracy and self-reported confidence., Results: Of 2515 contacted pediatric residents, 106 and 102 completed the KS and NS surveys, respectively. For KS, the sensitivity of text description was 48.5% (128 of 264), which was not significantly different from random guessing (odds ratio [OR], 0.94; 95% CI, 0.69-1.29; P = .71). Sensitivity was thus compared for real images vs random guessing (60.3% [188 of 312]; OR, 1.52; 95% CI, 1.15-2.00; P = .003) and 2 types of generative AI images vs random guessing (57.0% [212 of 372]; OR, 1.32; 95% CI, 1.04-1.69; P = .02 and 59.6% [193 of 324]; OR, 1.47; 95% CI, 1.12-1.94; P = .006) (denominators differ according to survey responses). The sensitivity of the NS text-only description was 65.3% (196 of 300). Compared with text-only, the sensitivity of the real images was 74.3% (205 of 276; OR, 1.53; 95% CI, 1.08-2.18; P = .02), and the sensitivity of the 2 types of images created by generative AI was 68.0% (204 of 300; OR, 1.13; 95% CI, 0.77-1.66; P = .54) and 71.0% (247 of 328; OR, 1.30; 95% CI, 0.92-1.83; P = .14). For specificity, no intervention was statistically different from text only. After the interventions, the number of participants who reported being unsure about important diagnostic facial features decreased from 56 (52.8%) to 5 (7.6%) for KS (P < .001) and 25 (24.5%) to 4 (4.7%) for NS (P < .001). There was a significant association between confidence level and sensitivity for real and generated images., Conclusions and Relevance: In this study, real and generated images helped participants recognize KS and NS; real images appeared most helpful. Generated images were noninferior to real images and could serve an adjunctive role, particularly for rare conditions.

Artificial intelligence (AI) for facial diagnostics is increasingly used in the genetics clinic to evaluate patients with potential genetic conditions. Current approaches focus on one type of AI called Deep Learning (DL). While DL- based facial diagnostic platforms have a high accuracy rate for many conditions, less is understood about how this technology assesses and classifies (categorizes) images, and how this compares to humans. To compare human and computer attention, we performed eye-tracking analyses of geneticist clinicians (n = 22) and non-clinicians (n = 22) who viewed images of people with 10 different genetic conditions, as well as images of unaffected individuals. We calculated the Intersection-over-Union (IoU) and Kullback-Leibler divergence (KL) to compare the visual attentions of the two participant groups, and then the clinician group against the saliency maps of our deep learning classifier. We found that human visual attention differs greatly from DL model's saliency results. Averaging over all the test images, IoU and KL metric for the successful (accurate) clinician visual attentions versus the saliency maps were 0.15 and 11.15, respectively. Individuals also tend to have a specific pattern of image inspection, and clinicians demonstrate different visual attention patterns than non-clinicians (IoU and KL of clinicians versus non-clinicians were 0.47 and 2.73, respectively). This study shows that humans (at different levels of expertise) and a computer vision model examine images differently. Understanding these differences can improve the design and use of AI tools, and lead to more meaningful interactions between clinicians and AI technologies., Competing Interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: BDS is the Co-Editor-in-Chief of the American Journal of Medical Genetics, and receives textbook royalties from Wiley., (Copyright: This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.)

Artificial intelligence (AI) is used in an increasing number of areas, with recent interest in generative AI, such as using ChatGPT to generate programming code or DALL-E to make illustrations. We describe the use of generative AI in medical education. Specifically, we sought to determine whether generative AI could help train pediatric residents to better recognize genetic conditions. From publicly available images of individuals with genetic conditions, we used generative AI methods to create new images, which were checked for accuracy with an external classifier. We selected two conditions for study, Kabuki (KS) and Noonan (NS) syndromes, which are clinically important conditions that pediatricians may encounter. In this study, pediatric residents completed 208 surveys, where they each classified 20 images following exposure to one of 4 possible educational interventions, including with and without generative AI methods. Overall, we find that generative images perform similarly but appear to be slightly less helpful than real images. Most participants reported that images were useful, although real images were felt to be more helpful. We conclude that generative AI images may serve as an adjunctive educational tool, particularly for less familiar conditions, such as KS.

Deep learning (DL) and other types of artificial intelligence (AI) are increasingly used in many biomedical areas, including genetics. One frequent use in medical genetics involves evaluating images of people with potential genetic conditions to help with diagnosis. A central question involves better understanding how AI classifiers assess images compared to humans. To explore this, we performed eye-tracking analyses of geneticist clinicians and non-clinicians. We compared results to DL-based saliency maps. We found that human visual attention when assessing images differs greatly from the parts of images weighted by the DL model. Further, individuals tend to have a specific pattern of image inspection, and clinicians demonstrate different visual attention patterns than non-clinicians.

Deep learning (DL) is applied in many biomedical areas. We performed a scoping review on DL in medical genetics. We first assessed 14,002 articles, of which 133 involved DL in medical genetics. DL in medical genetics increased rapidly during the studied period. In medical genetics, DL has largely been applied to small data sets of affected individuals (mean = 95, median = 29) with genetic conditions (71 different genetic conditions were studied; 24 articles studied multiple conditions). A variety of data types have been used in medical genetics, including radiologic (20%), ophthalmologic (14%), microscopy (8%), and text-based data (4%); the most common data type was patient facial photographs (46%). DL authors and research subjects overrepresent certain geographic areas (United States, Asia, and Europe). Convolutional neural networks (89%) were the most common method. Results were compared with human performance in 31% of studies. In total, 51% of articles provided data access; 16% released source code. To further explore DL in genomics, we conducted an additional analysis, the results of which highlight future opportunities for DL in medical genetics. Finally, we expect DL applications to increase in the future. To aid data curation, we evaluated a DL, random forest, and rule-based classifier at categorizing article abstracts., Competing Interests: Conflict of Interest The authors declare no conflicts of interest., (Published by Elsevier Inc.)

Background: In medical genetics, one application of neural networks is the diagnosis of genetic diseases based on images of patient faces. While these applications have been validated in the literature with primarily pediatric subjects, it is not known whether these applications can accurately diagnose patients across a lifespan. We aimed to extend previous works to determine whether age plays a factor in facial diagnosis as well as to explore other factors that may contribute to the overall diagnostic accuracy. Methods: To investigate this, we chose two relatively common conditions, Williams syndrome and 22q11.2 deletion syndrome. We built a neural network classifier trained on images of affected and unaffected individuals of different ages and compared classifier accuracy to clinical geneticists. We analyzed the results of saliency maps and the use of generative adversarial networks to boost accuracy. Results: Our classifier outperformed clinical geneticists at recognizing face images of these two conditions within each of the age groups (the performance varied between the age groups): 1) under 2 years old, 2) 2-9 years old, 3) 10-19 years old, 4) 20-34 years old, and 5) ≥35 years old. The overall accuracy improvement by our classifier over the clinical geneticists was 15.5 and 22.7% for Williams syndrome and 22q11.2 deletion syndrome, respectively. Additionally, comparison of saliency maps revealed that key facial features learned by the neural network differed with respect to age. Finally, joint training real images with multiple different types of fake images created by a generative adversarial network showed up to 3.25% accuracy gain in classification accuracy. Conclusion: The ability of clinical geneticists to diagnose these conditions is influenced by the age of the patient. Deep learning technologies such as our classifier can more accurately identify patients across the lifespan based on facial features. Saliency maps of computer vision reveal that the syndromic facial feature attributes change with the age of the patient. Modest improvements in the classifier accuracy were observed when joint training was carried out with both real and fake images. Our findings highlight the need for a greater focus on age as a confounder in facial diagnosis., Competing Interests: BS is the editor-in-chief of the American Journal of Medical Genetics. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Duong, Hu, Tekendo-Ngongang, Hanchard, Liu, Solomon and Waikel.)

Neural networks have shown strong potential in research and in healthcare. Mainly due to the need for large datasets, these applications have focused on common medical conditions, where more data are typically available. Leveraging publicly available data, we trained a neural network classifier on images of rare genetic conditions with skin findings. We used approximately 100 images per condition to classify 6 different genetic conditions. We analyzed both preprocessed images that were cropped to show only the skin lesions as well as more complex images showing features such as the entire body segment, the person, and/or the background. The classifier construction process included attribution methods to visualize which pixels were most important for computer-based classification. Our classifier was significantly more accurate than pediatricians or medical geneticists for both types of images and suggests steps for further research involving clinical scenarios and other applications., Competing Interests: B.D.S.: Editor-in-Chief of the American Journal of Medical Genetics; previously (until 2019) employee of GeneDx, Inc., a genetic/genomic testing laboratory that is a subsidiary of Opko Health, and previous holder of Opko Health stock options; previously (until 2019) member of Scientific Advisory Board for FDNA. All other authors declare no competing interests.

Nonmelanoma skin cancers (NMSCs) consist of a variety of tumor types including basal cell carcinoma, squamous cell carcinoma, a variety of hair follicle tumors, and sebaceous gland tumors. Genetic alterations that alter the fate of multipotent stem cells are believed to influence NMSC phenotype. We previously generated a transgenic mouse line which constitutively expressed c-myc under the control of the K14 promoter (K14.MYC2). These mice exhibited an increase in size and number of sebaceous glands, suggesting that c-myc diverted multipotential stem cells to a sebaceous lineage. Our goal in the current study was to determine if alterations in the commitment of multipotent stem cells to different cell fates would influence tumor phenotype. To this end, we exposed K14.MYC2 mice to a chemical carcinogenesis protocol and discovered that these mice were predisposed to develop sebaceous adenomas. Our data demonstrate that genetic alterations that alter the fate of multipotent stem cells during embryonic development can markedly influence the phenotype of NMSC that develop following exposure to carcinogens.

The homing properties of subsets of lymphocytes and dendritic cells (DC) are regulated in part by the profile of chemokine receptors expressed. To determine how CCR6 influences cell trafficking, a mutant allele of the mouse CCR6 gene was produced that includes an enhanced green fluorescent protein (EGFP) reporter under the control of the CCR6 promoter. In mice heterozygous for the EGFP/CCR6 knock-in, CCR6 expression was detected on all mature B cells, subpopulations of splenic CD4(+) and CD8(+) T cells, and on some CD11c(+) DC. Most CD11b(+) myeloid DC expressed CCR6, but CD8alpha(+) lymphoid DC were negative for CCR6. Among myeloid DC, the CD4(+) subset was uniformly positive for CCR6 expression and the CD4(-) subset was mostly CCR6 positive. Epidermal Langerhans cells (LC) also expressed CCR6, but at lower levels than splenic myeloid DC. Culture of bone marrow precursors from the knock-in mice with GM-CSF for 4 to 6 days led to the appearance of a subset of CD11c(+) DC expressing CCR6. The differences in CCR6 expression among the major DC subsets indicate that CCR6 and its chemokine ligand MIP-3alpha participate in determining the positioning of DC subsets in epithelial and lymphoid tissues.

Background: The phase II enzyme NAD(P)H :quinone oxidoreductase 1 (NQO1) catalyzes quinone detoxification, protecting cells from redox cycling, oxidative stress, mutagenicity, and cytotoxicity induced by quinones and its precursors. We have used NQO1(-/-) C57BL/6 mice to show that NQO1 protects them from skin cancer induced by the polycyclic aromatic hydrocarbon benzo[a]pyrene. Herein, we used NQO1(-/-) mice to investigate whether NQO1 also protects them against 7,12-dimethylbenz[a]anthracene (DMBA), where methyl substituents diminish primary quinone formation., Methods: Dorsal skin of NQO1(-/-) or wild-type C57BL/6 mice was shaved. When tested as a complete carcinogen, DMBA (500 or 750 microg in 100 microL of acetone) alone was applied to the shaved area. When tested as a tumor initiator, DMBA (200 or 400 nmol in 100 microL of acetone) was applied to the shaved area; 1 week later, twice-weekly applications of phorbol 12-myristate 13-acetate (PMA)-10 microg dissolved in 200 microL of acetone-to the same area began and were continued for 20 weeks. Tumor development was monitored in all mice (12-15 per group). All statistical tests were two-sided., Results: When DMBA (750 microg) was tested as a complete carcinogen, about 50% of the DMBA-treated NQO1(-/-) mice but no DMBA-treated wild-type mouse developed skin tumors. When DMBA (both concentrations) was used as a tumor initiator, NQO1(-/-) mice developed larger tumors at a greater frequency than their wild-type littermates. Twenty-three weeks after the first PMA treatment in the tumor initiator test, all 30 NQO1(-/-) mice given 400 nmol of DMBA had developed skin tumors, compared with 33% (10 of 30) of treated wild-type mice (P<.001)., Conclusions: NQO1(-/-) mice are more susceptible to DMBA-induced skin cancer than are their wild-type littermates, suggesting that NQO1 may protect cells from DMBA carcinogenesis.

The beta-catenin/TCF signaling pathway is essential for the maintenance of epithelial stem cells in the small intestine. c-Myc a downstream target of beta-catenin/TCF (ref. 2), can induce differentiation of epidermal stem cells in vitro. To determine the role of c-Myc in epidermal stem cells in vivo, we have targeted expression of human MYC2 to the hair follicles and the basal layer of mouse epidermis using a keratin 14 vector (K14.MYC2). Adult K14.MYC2 mice gradually lose their hair and develop spontaneous ulcerated lesions due to a severe impairment in wound healing; their keratinocytes show impaired migration in response to wounding. The expression of beta1 integrin, which is preferentially expressed in epidermal stem cells is unusually low in the epidermis of K14.MYC2 mice. Label-retaining analysis to identify epidermal stem cells reveals a 75% reduction in the number of stem cells in 3-month-old K14.MYC2 mice, compared with wildtype mice. We conclude that deregulated expression of c-Myc in stem cells reduces beta1 integrin expression, which is essential to both keratinocyte migration and stem cell maintenance.

NAD(P)H:quinone oxidoreductase 1 (NQO1) is a flavoprotein that catalyzes the metabolic detoxification of quinones and their derivatives. This protects cells against quinone-induced oxidative stress, cytotoxicity, and mutagenicity. C57BL6 NQO1-/- mice, deficient in NQO1 RNA and protein, were generated in our laboratory. To investigate the role of NQO1 in chemical carcinogenesis, the dorsal skin of NQO1-deficient (NQO1-/-) and wild-type (NQO1+/+) mice were treated with a single dose of benzo(a)pyrene, followed by twice weekly applications of phorbol-12-myristate-13-acetate. The NQO1-/- mice showed a much higher frequency of skin tumor development when compared with their wild-type littermates. Interestingly, the male NQO1-/- mice were slower to develop skin tumors than their NQO1-/- female littermates. Histological analysis of the NQO1-/- tumors showed proliferative activity. These results demonstrate that NQO1 acts as an endogenous factor in protection against benzo(a)pyrene carcinogenicity.

c-Myc overexpression has been associated with several types of human cancers. To study the role of c-myc in epidermal differentiation and carcinogenesis, a transgenic mouse model was created to overexpress c-Myc in the epidermis. Human c-myc 2 cDNA was subcloned into a 6.5 kb mouse loricrin expression vector, ML.myc2. This loricrin promoter primarily directs expression in the epidermis in both proliferating and differentiated keratinocytes. On day 4, ML.myc2 transgenic pups develop a hyperkeratotic phenotype, which progressively worsens until day 7. Upon histological analysis, both hyperplasia and hyperkeratosis were evident. Bromodeoxyuridine (BrdU) incorporation revealed that transgenic mice had a threefold increase in the number of proliferating cells as compared with a normal littermate. Proliferative cells in the ML.myc2 epidermis were also found to be suprabasal, suggesting an inhibition of terminal differentiation in keratinocytes. Inhibition of terminal differentiation by c-Myc overexpression was further suggested by aberrant expression of differentiation markers, keratin 1, keratin 6, loricrin, and filaggrin in ML.myc2 transgenic mice. Interestingly, ML.myc2 keratinocytes exhibit a reduced sensitivity to UV-B induced apoptosis, in vivo. In vitro studies reveal the reduced sensitivity of ML.myc2 keratinocytes to UV-B irradiation is growth factor dependent. These findings provide evidence that overexpression of c-Myc in the epidermis induces proliferation, inhibits terminal differentiation and decreases the sensitivity of keratinocytes to UV-B induced apoptosis.