Identifying facial phenotypes of genetic disorders using deep learning

@article{Gurovich2019IdentifyingFP,
  title={Identifying facial phenotypes of genetic disorders using deep learning},
  author={Yaron Gurovich and Yair Hanani and Omri Bar and Guy Nadav and Nicole Fleischer and Dekel Gelbman and Lina Basel-Salmon and Peter M. Krawitz and Susanne B. Kamphausen and Martin Zenker and Lynne M. Bird and Karen W. Gripp},
  journal={Nature Medicine},
  year={2019},
  volume={25},
  pages={60-64}
}
Syndromic genetic conditions, in aggregate, affect 8% of the population1. Many syndromes have recognizable facial features2 that are highly informative to clinical geneticists3–5. Recent studies show that facial analysis technologies measured up to the capabilities of expert clinicians in syndrome identification6–9. However, these technologies identified only a few disease phenotypes, limiting their role in clinical settings, where hundreds of diagnoses must be considered. Here we present a… 
Deep Learning-Based Analysis of Face Images as a Screening Tool for Genetic Syndromes
TLDR
A screening tool based on deep learning and facial recognition could not only detect known diseases, but also detect patients with diseases that were not previously known, indicating that this tool could help in screening patients before they are referred to the genetic unit.
Evaluation of Face2Gene using facial images of patients with congenital dysmorphic syndromes recruited in Japan
TLDR
An evaluation of Face2Gene using patients with congenital dysmorphic syndromes recruited in Japan suggests that the deep learning-based diagnosis assistance system in its current format is already useful in suggesting candidate Syndromes to clinical geneticists.
Automated syndrome diagnosis by three-dimensional facial imaging
TLDR
Deep phenotyping by quantitative 3D facial imaging has considerable potential to facilitate syndrome diagnosis and “unaffected” relatives may identify unrecognized cases or may reveal novel examples of semidominant inheritance.
Efficiency of Computer-Aided Facial Phenotyping (DeepGestalt) in Individuals With and Without a Genetic Syndrome: Diagnostic Accuracy Study
TLDR
A machine learning–based framework for the automated differentiation of DeepGestalt’s output on photos of individuals with and without a genetic syndrome is proposed and may help enhance it and similar computer-aided facial phenotyping tools.
Efficiency of Computer-Aided Facial Phenotyping (DeepGestalt) in Individuals With and Without a Genetic Syndrome: Diagnostic Accuracy Study (Preprint)
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
GestaltMatcher: Overcoming the limits of rare disease matching using facial phenotypic descriptors
TLDR
To improve recognition of ultra-rare diseases, GestaltMatcher is created, which uses a deep convolutional neural network based on the DeepGestalt framework and could accelerate the clinical diagnosis of patients with ultra-RAre disorders and facial dysmorphism.
Neural networks for classification and image generation of aging in genetic syndromes
TLDR
Comparison of saliency maps revealed that key facial features learned by the neural network differed slightly with respect to age, and 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.
GestaltMatcher: Overcoming the limits of rare disease matching using facial phenotypic descriptors
TLDR
GestaltMatcher is a deep convolutional neural network that uses photographs of 21,558 patients with 1,395 rare disorders to define a "Clinical Face Phenotype Space", allowing test patients to be matched to a molecular diagnosis even when the disorder was not part of the training set.
Differentiating molecular etiologies of Angelman syndrome through facial phenotyping using deep learning.
TLDR
It is determined that DeepGestalt demonstrated a high degree of discrimination between the deletion subtype and UPD or imprinting defects, and a lower degree ofdiscrimination with the UBE3A pathogenic variants subtype.
Evaluating Face2Gene as a Tool to Identify Cornelia de Lange Syndrome by Facial Phenotypes
TLDR
Overall, it is concluded that computer-assisted image analysis based on deep learning could support the clinical diagnosis of CdLS.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 49 REFERENCES
Diagnostically relevant facial gestalt information from ordinary photos
TLDR
An automatic approach that implements recent developments in computer vision extracts phenotypic information from ordinary non-clinical photographs and models human facial dysmorphisms in a multidimensional 'Clinical Face Phenotype Space' that provides a novel method for inferring causative genetic variants from clinical sequencing data through functional genetic pathway comparisons.
The face of Noonan syndrome: Does phenotype predict genotype
TLDR
It is determined that some individuals with mutations in the most commonly affected gene, PTPN11, which is correlated with the cardinal physical features, may have a quite atypical face, while some others with KRAS mutations can have a very typical face.
Recognition of the Cornelia de Lange syndrome phenotype with facial dysmorphology novel analysis
TLDR
The results from both studies indicated that the FDNA technology's detection rate was comparable to that of dysmorphology experts, indicating that utilizing such technologies may be a useful tool in a clinical setting.
Automatic recognition of the XLHED phenotype from facial images
TLDR
The automated facial recognition system represents a promising non‐invasive technology to screen patients at all ages for a possible diagnosis of ectodermal dysplasia, with greatest sensitivity and specificity for males affected with XLHED.
Down syndrome in diverse populations
TLDR
Evaluation using a digital facial analysis technology of a larger diverse cohort of newborns to adults was able to diagnose Down syndrome with a sensitivity of 0.961, specificity of0.924, and accuracy of 1.065, demonstrating the accuracy and promise of digital facialAnalysis technology in the diagnosis of Down syndrome internationally.
Computer-based recognition of dysmorphic faces
TLDR
The results prove that certain syndromes are associated with a specific facial pattern and that this pattern can be described in mathematical terms.
A Deep Learning Frame-Work for Recognizing Developmental Disorders
TLDR
A novel framework to detect developmental disorders from facial images based on Deep Convolutional Neural Networks for feature extraction and results indicate that the model performs better than average human intelligence in terms of differentiating amongst different disabilities.
Identification of dysmorphic syndromes using landmark-specific local texture descriptors
TLDR
This work presents a general framework for the detection of genetic disorders from facial pictures, combining geometrical and texture features, and computes optimal local image filters and soft neighborhood weighting matrices that enhance the discriminative ability of the system.
Digital facial dysmorphology for genetic screening: Hierarchical constrained local model using ICA
TLDR
The promising results indicate that the proposed simple, non-invasive and automated framework for Down syndrome detection based on disease-specific facial patterns could assist in Down syndrome screening effectively and extensible to detection of other genetic syndromes.
22q11.2 deletion syndrome in diverse populations
TLDR
It is demonstrated how facial analysis technology can assist clinicians in making accurate 22q11.2 DS diagnoses and will assist in earlier detection and in increasing recognition of 22q12.2 deletion syndrome throughout the world.
...
1
2
3
4
5
...