Researchers develop machine-learning screening tool for genetic syndromes in children


With an average accuracy of 88%, deep learning technology offers rapid genetic testing that could speed up the diagnosis of genetic syndromes, recommending further investigation or referral to a specialist within seconds, according to a study published in The Lancet Digital Health. Formed with data from 2,800 pediatric patients from 28 countries, the technology also takes into account facial variability related to gender, age, race and ethnicity, researcher study finds of the Children’s National Hospital.

We have built a software device to increase access to care and machine learning technology to identify disease patterns that are not immediately obvious to the human eye or intuition, and to help lay physicians in genetics. This technological innovation can help children without access to specialized clinics, which are not available in most countries. Ultimately, this can help reduce health inequalities in underfunded societies. “

Marius George Linguraru, D.Phil., MA, M.Sc., principal investigator at the Sheikh Zayed Institute for Pediatric Surgical Innovation at the National Children’s Hospital and lead author of the study

This machine learning technology indicates the presence of a genetic syndrome from a photograph of the face captured at the point of care, such as in pediatricians’ offices, maternity wards and general practitioner clinics.

“Unlike other technologies, the strength of this program is to distinguish ‘normal’ from ‘non-normal’, making it an effective screening tool in the hands of community caregivers,” said Marshall L. Summar, MD. , director of the Rare Disease Institute for National Children. “This can dramatically speed up the time to diagnosis by providing a robust indicator for patients who need further investigation. This first step is often the biggest obstacle to progressing to a diagnosis. Once a patient is in the examination system, the likelihood of diagnosis (in many ways) is greatly increased.

Every year, millions of children are born with genetic disorders -; including Down syndrome, a condition in which a child is born with an extra copy of its 21st chromosome causing developmental delays and disabilities, Williams-Beuren syndrome, a rare multisystem disease caused by a submicroscopic deletion of a region of chromosome 7 and Noonan syndrome, a genetic disease caused by a defective gene that prevents the normal development of various parts of the body.

Most children with genetic syndromes live in areas where resources and access to genetic services are limited. Genetic testing can come at a high price. Specialists are also insufficient to help identify genetic syndromes early in life when preventive care can save lives, especially in low-income areas, with limited resources and isolated communities.

“The technology presented can help pediatricians, neonatologists and family physicians in the routine or remote assessment of pediatric patients, especially in areas where access to specialist care is limited,” said Porras et al. “Our technology can be a step forward in the democratization of health resources for genetic testing.”

Researchers trained the technology using 2,800 retrospective facial photographs of children, with or without genetic syndrome, from 28 countries, such as Argentina, Australia, Brazil, China, France, Morocco, Nigeria, Paraguay, Thailand and the United States. was designed to take into account normal variations in facial appearance among populations of various demographic groups.

“The appearance of the face is influenced by the race and ethnicity of the patients. The wide variety of conditions and the diversity of populations have an impact on the early identification of these conditions due to the lack of data that can be used as a guide. benchmark, ”Linguraru said. . “Racial and ethnic disparities still exist in the survival of the genetic syndrome, even under some of the most common and best studied conditions. “

Like all machine learning tools, they are trained with the available dataset. The researchers expect that as more data from under-represented groups will become available, they will adapt the model to locate phenotypic variations within more specific demographic groups.

In addition to being an accessible tool that could be used in telehealth services to assess genetic risk, there are other potentials for this technology.

“I am also excited about the potential of the technology in newborn screening,” Linguraru said. “There are approximately 140 million newborns worldwide each year, eight million of whom are born with a serious birth defect of genetic or partially genetic origin, many of which are discovered late.”

Children’s National also recently announced that it has entered into a licensing agreement with MGeneRx Inc. for its patented pediatric medical device technology. MGeneRx is a spin-off of BreakThrough BioAssets LLC, a life science technology company focused on accelerating and commercializing new innovations, such as this technology, with a focus on positive social impact.

“The social impact of this technology cannot be underestimated,” said Nasser Hassan, Acting Managing Director of MGeneRx Inc. “We are delighted with this licensing agreement with the Children’s National Hospital and the opportunity to ” improve this technology and extend its application to populations where precision medicine and the earliest possible interventions are absolutely necessary to save and improve the lives of children. “


National Children’s Hospital

Journal reference:

Porras, AR, et al. (2021) Development and evaluation of a machine learning-based point-of-care screening tool for genetic syndromes in children: a multinational retrospective study. The Lancet Digital Health.


Leave A Reply