There are more than 7000 known genetic disorders, and the number of affected is estimated to be about 6-10% of the population. Around 30 to 40% of genetic disorders have physical changes in the face and skull such as Down's syndrome or Fragile X syndrome. Therefore, the known facial phenotype of many genetic disorders is highly informative to clinical diagnosis.
Since a large number of genetic diseases are associated with special facial phenotypes that are difficult to remember, automated facial analysis such as Face2Gene and GestaltMatcher can assist in the identification and diagnosis of facial phenotypes related to various genetic diseases. Although the current advances in whole exome sequencing (whole exome sequencing) or whole genome sequencing (whole genome sequencing) have greatly improved the diagnostic rate of genetic diseases, about half of the patients are still undiagnosed.
For patients with special facial phenotypes, the investigators believe that by combining automated facial analysis and whole exome sequencing data, it should be possible to provide a fast and accurate diagnostic model of genetic mutations for genetic diseases. GestaltMatcher Database is a medical imaging database of rare diseases developed by Professor Peter Krawitz of the University of Bonn, Germany. The database's artificial intelligence module will infer a patient's possible diagnosis based on the patient's photo, age, gender, race, and clinical description. The database will be open to medical researchers in related fields to improve the diagnosis of rare diseases.
The investigators will use GestaltMatcher to assist in the diagnosis of patients, and compare the accuracy and significant differences in facial deformities between Taiwanese patients and patients from different countries. And use Eye Tracker to analyze how doctors diagnose patients through facial photos, and compare whether there are significant differences between foreign patients and Taiwanese patients in the diagnosis literature of Taiwanese doctors. The project will also analyze how genetic doctors at the University of Bonn in Germany diagnose patients, and compare it with Taiwanese doctors to better understand the differences in the process of doctors diagnosing patients and ethnic backgrounds.
Cases with abnormal appearance of clinical symptoms and suspected genetic diseases
Unable to cooperate with the examiner
Cases with abnormal appearance of clinical symptoms and suspected genetic diseases at National Taiwan University Hospital
Case-only
Prospective
Take photo
DeviceWe will use the camera to take 2 front and side images of the participant, and select one of the better images to upload to the facial analysis system (GestaltMatcher Database) for analysis.
The diagnosis accuracy rate of use GestaltMatcher to analysis participant facial features in rare diseases
GestaltMatcher is a automated facial analysis software, which utilized deep convolutional neural networks trained on patients' photos as an encoder to convert facial photos into feature vectors to form a Clinical Face Phenotype Space (CFPS). And quantified the similarity among patients by the cosine distance of two vectors in CFPS. With this approach, the investigators can support the ultra-rare syndromes that lack images to be trained and push the supported syndromes. Base on on this technology, the investigators use GestaltMatcher to analysis participant facial features and compare those vectors which are similar to patients to find possible syndromes. The investigators will verify more clinical phenotype and genetic data of the participants for verification.
The significant difference in facial deformities between Taiwan participant and participant from different countries by GestaltMatcher
The investigators will ues those facial feature vectors from GestaltMatcher Database to compare with Taiwan participants in the research platform within the GestaltMatcher Database to find the facial differences between Taiwan and other different countries participants by pairwise matrix. To find out where is the difference between Taiwan and different countries participants, improving the diagnostic perspectives in Taiwanese groups.
Criteria for doctors to diagnose participants from facial features assessed by eye trackers
Use Eye Tracker to mark out where and when are the doctors pays attention to when observing the participant dysmorphic features. To analyze how doctors diagnose patients through facial photos.
Diagnosis differences between Taiwan and Bonn University genetics doctors
Compare whether there are significant differences between Taiwanese doctors using literature data and phenotypic analysis to diagnose foreign patients and Taiwanese patients, and analyzing how geneticists at Bonn University diagnose patients
Click on the lead sponsor to view all of their trials.
National Taiwan University Hospital
Lead sponsor
University of Bonn
Collaborator
