Facial Recognition Software Succeeds in Diagnosing Rare Genetic Diseases
Many rare genetic diseases are difficult to identify, and their diagnostic process is long and tedious. However, a breakthrough allows doctors to use facial recognition software as part of the diagnostic process.
(Photo: shutterstock)Diagnosing many genetic diseases is a long and tedious process. Many of these diseases are extremely rare and require numerous tests and a trial-and-error process to finally determine the exact disease. The speed of the diagnostic process is crucial in order to begin proper treatment of the patient as soon as possible.
Starting today, researchers have a new tool in their arsenal to assist them in the diagnostic process: facial recognition software successfully diagnosed a rare genetic disease called DiGeorge syndrome in a conducted experiment. This disease affects one in 6000 children and includes defects such as heart problems and cleft palate.
The experiment was conducted by the Human Genome Research Institute of the British Health Services. Researchers fed the software 126 variations in facial features that aid in identification, and the software subsequently diagnosed the syndrome with 96.6% accuracy across populations of all ethnic backgrounds.
Medical geneticist Paul Kruschka explains: "Syndromes that distort human appearance look different in different parts of the world. Even experienced doctors can have difficulty diagnosing genetic syndromes in non-European populations." To bypass this issue, researchers enlisted 156 volunteers from all over the globe, created a 'dictionary' of all expressions of DiGeorge syndrome in every ethnic group, and input it into the software. As a result, this software can now assist any doctor in diagnosing the syndrome.
Researchers assert that this technological achievement is not limited to DiGeorge syndrome. The software has already demonstrated its capability in diagnosing Down syndrome, and it is expected to successfully diagnose at least two more genetic diseases soon.
