Collaboration between the FDNA, the Cure Sanfilippo Foundation, and the Jonah’s Just Begun Foundation has led to technology that successfully recognises the facial phenotype (observable characteristics) of patients with mucopolysaccharidosis IIIB, also known as Sanfilippo syndrome type-B.
The study was part of the Genomics Collaborative, which is a project led by the FDNA to use deep learning and artificial intelligence in medicine. The full article can be read here, at MySanAntonio.
Sanfilippo syndrome is a life-shortening genetic condition that leads to brain damage. There are four subtypes of the disease known as A, B, C, and D. A different missing enzyme causes each subtype. Those with type B have a variation in the NAGLU gene that results in the body being unable to break down large sugar molecules. This can lead to neurological symptoms including dementia, aggressive behaviour, seizures, deafness, hyperactivity, poor sleep, and vision loss. Those with the disease also tend to have certain physical features, such as coarse hair, prominent foreheads, large lips, and a low nasal bridge, amongst others.
The scientists working on the Genomics Collaborative study used these facial features to create an algorithm that can be used to diagnose people with Sanfilippo syndrome type-B. The researchers used photos from 614 patients to train a facial analysis program to recognise common features. The algorithm is now believed to be able to recognise the disease in undiagnosed patients. Ilana Jacqueline, the Coordinator of the Genomics Collaborative, thanked the families involved in the research, and said,
“their contributions will help advance this technology with the potential to change lives for the better.”
