Posts

Coronavirus and lungs with world map in the background

Top AI models unveiled in COVID-19 challenge to improve lung diagnostics

/in , , , /by

Coronavirus and lungs with world map in the background

The top 10 results have been unveiled in the first-of-its-kind COVID-19 Lung CT Lesion Segmentation Grand Challenge, a groundbreaking research competition focused on developing artificial intelligence (AI) models to help in the visualization and measurement of COVID specific lesions in the lungs of infected patients, potentially facilitating more timely and patient-specific medical interventions.

Attracting more than 1,000 global participants, the competition was presented by the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National Hospital in collaboration with leading AI technology company NVIDIA and the National Institutes of Health (NIH). The competition’s AI models utilized a multi-institutional, multi-national data set provided by public datasets from The Cancer Imaging Archive (National Cancer Institute), NIH and the University of Arkansas, that originated from patients of different ages, genders and with variable disease severity. NVIDIA provided GPUs to the top five winners as prizes, as well as supported the selection and judging process.

“Improving COVID-19 treatment starts with a clearer understanding of the patient’s disease state. However, a prior lack of global data collaboration limited clinicians in their ability to quickly and effectively understand disease severity across both adult and pediatric patients,” says Marius George Linguraru, D.Phil., M.A., M.Sc., principal investigator at the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National, who led the Grand Challenge initiative. “By harnessing the power of AI through quantitative imaging and machine learning, these discoveries are helping clinicians better understand COVID-19 disease severity and potentially stratify and triage into appropriate treatment protocols at different stages of the disease.”

The top 10 AI algorithms were identified from a highly competitive field of participants who tested the data in November and December 2020. The results were unveiled on Jan. 11, 2021, in a virtual symposium, hosted by Children’s National, that featured presentations from top teams, event organizers and clinicians.

Developers of the 10 top AI models from the COVID-19 Lung CT Lesion Segmentation Grand Challenge are:

  1. Shishuai Hu, et al. Northwestern Polytechnical University, China. “Semi-supervised Method for COVID-19 Lung CT Lesion Segmentation”
  2. Fabian Isensee, et al. German Cancer Research Center, Germany. “nnU-Net for Covid Segmentation”
  3. Claire Tang, Lynbrook High School, USA. “Automated Ensemble Modeling for COVID-19 CT Lesion Segmentation”
  4. Qinji Yu, et al. Shanghai JiaoTong University, China. “COVID-19-20 Lesion Segmentation Based on nnUNet”
  5. Andreas Husch, et al. University of Luxembourg, Luxembourg. “Leveraging State-of-the-Art Architectures by Enriching Training Information – a case study”
  6. Tong Zheng, et al. Nagoya University, Japan. “Fully-automated COVID-19-20 Segmentation”
  7. Vitali Liauchuk. United Institute of Informatics Problems (UIIP), Belarus. “Semi-3D CNN with ImageNet Pretrain for Segmentation of COVID Lesions on CT”
  8. Ziqi Zhou, et al. Shenzhen University, China. “Automated Chest CT Image Segmentation of COVID-19 with 3D Unet-based Framework”
  9. Jan Hendrik Moltz, et al. Fraunhofer Institute for Digital Medicine MEVIS, Germany. “Segmentation of COVID-19 Lung Lesions in CT Using nnU-Net”
  10. Bruno Oliveira, et al. 2Ai – Polytechnic Institute of Cávado and Ave, Portugal. “Automatic COVID-19 Detection and Segmentation from Lung Computed Tomography (CT) Images Using 3D Cascade U-net”

Linguraru added that, in addition to an award for the top five AI models, these winning algorithms are now available to partner with clinical institutions across the globe to further evaluate how these quantitative imaging and machine learning methods may potentially impact global public health.

“Quality annotations are a limiting factor in the development of useful AI models,” said Mona Flores, M.D., global head of Medical AI, NVIDIA. “Using the NVIDIA COVID lesion segmentation model available on our NGC software hub, we were able to quickly label the NIH dataset, allowing radiologists to do precise annotations in record time.”

“I applaud the computer science, data science and image processing global academic community for rapidly teaming up to combine multi-disciplinary expertise towards development of potential automated and multi-parametric tools to better study and address the myriad of unmet clinical needs created by the pandemic,” said Bradford Wood, M.D., director, NIH Center for Interventional Oncology and chief, Interventional Radiology Section, NIH Clinical Center. “Thank you to each team for locking arms towards a common cause that unites the scientific community in these challenging times.”

communication network concept image

Children’s National joins international AI COVID-19 initiative

/in , , , , , , , /by
communication network concept image

Children’s National Hospital is the first pediatric partner to join an international initiative led by leading technology firm NVIDIA and Massachusetts General Brigham Hospital, focused on creating solutions through machine and deep learning to benefit COVID-19 healthcare outcomes.

Children’s National Hospital is the first pediatric partner to join an international initiative led by leading technology firm NVIDIA and Massachusetts General Brigham Hospital, focused on creating solutions through machine and deep learning to benefit COVID-19 healthcare outcomes. The initiative, known as EXAM (EMR CXR AI Model) is the largest and most diverse federated learning enterprise, comprised of 20 leading hospitals from around the globe.

Marius George Linguraru, D.Phil., M.A., M.Sc., principal investigator at the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National Hospital, noted that one of the core goals of the initiative is to create a platform which brings resources together, from a variety of leading institutions, to advance the care of COVID-19 patients across the board, including children.

“Children’s National Hospital is proud to be the first pediatric partner joining the world’s leading healthcare institutions in this collaboration to advance global health,” says Linguraru. “We are currently living in a time where rapid access to this kind of global data has never been more important — we need solutions that work fast and are effective. That is not possible without this degree of collaboration and we look forward to continuing this important work with our partners to address one of the most significant healthcare challenges in our lifetime.”

A recent systematic review and meta-analysis from Children’s National Hospital became another core contribution to understanding how children are impacted by COVID-19. Led by Linguraru and accepted to be published in Pediatric Pulmonology, it offers the first comprehensive summary of the findings of various studies published thus far that describe COVID-19 lung imaging data across the pediatric population.

The review examined articles based on chest CT imaging in 1,026 pediatric patients diagnosed with COVID-19, and concluded that chest CT manifestations in those patients could potentially be used to prompt intervention across the pediatric population.

Marius George Linguraru

“Children’s National Hospital is proud to be the first pediatric partner joining the world’s leading healthcare institutions in this collaboration to advance global health,” says Marius George Linguraru, D.Phil., M.A., M.Sc.

“Until this point, pediatric COVID-19 studies have largely been restricted to case reports and small case series, which have prevented the identification of any specific pediatric lung disease patterns in COVID-19 patients,” says Linguraru. “Not only did this review help identify the common patterns in the lungs of pediatric patients presenting COVID-19 symptoms, which are distinct from the signs of other viral respiratory infections in children, it also provided insight into the differences between children and adults with COVID-19.”

Earlier this month, NVIDIA announced the EXAM initiative had – in just 20 days – developed an artificial intelligence (AI) model to determine whether a patient demonstrating COVID-19 symptoms in an emergency room would require supplemental oxygen hours – even days – after the initial exam. This data ultimately aids physicians in determining the proper level of care for patients, including potential ICU placement.

The EXAM initiative achieved a machine learning model offering precise prediction for the level of oxygen incoming patients would require.

In addition to Children’s National Hospital, other participants included Mass Gen Brigham and its affiliated hospitals in Boston; NIHR Cambridge Biomedical Research Centre; The Self-Defense Forces Central Hospital in Tokyo; National Taiwan University MeDA Lab and MAHC and Taiwan National Health Insurance Administration; Tri-Service General Hospital in Taiwan; Kyungpook National University Hospital in South Korea; Faculty of Medicine, Chulalongkorn University in Thailand; Diagnosticos da America SA in Brazil; University of California, San Francisco; VA San Diego; University of Toronto; National Institutes of Health in Bethesda, Maryland; University of Wisconsin-Madison School of Medicine and Public Health; Memorial Sloan Kettering Cancer Center in New York; and Mount Sinai Health System in New York.

illustration of lungs surrounded by virus

COVID-19: First comprehensive review of pediatric lung imaging features

/in , , , , /by
illustration of lungs surrounded by virus

A systematic review and meta-analysis by Children’s National Hospital researchers, published in Pediatric Pulmonology, provides the first comprehensive review of the findings of published studies describing COVID-19 lung imaging data in children.

The number COVID-19 studies focused on children have been small and with limited data. This has prevented the identification of specific pediatric lung disease patterns in COVID-19. Although children make up around 9.5% of COVID-19 infections, less than 2% of the literature on the virus, its symptoms and effects, have focused on kids.

A systematic review and meta-analysis by Children’s National Hospital researchers, published in Pediatric Pulmonology, provides the first comprehensive review of the findings of published studies describing COVID-19 lung imaging data in children. The analysis concludes that chest CT manifestations in children with COVID‐19 could potentially prompt intervention in the pediatric population.

Marius George Linguraru, D.Phil., M.A., M.Sc., principal investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National, discusses the importance of this work.

Q: What findings stand out to you?

A: We found that more than a third of children with COVID-19 had normal imaging. The lung imaging findings in these children were overall less frequent and less severe than in adult patients, but they were also more heterogeneous than in adults. Importantly, children with COVID-19 were three times more likely to have a normal exam than adults.

Several common lung imaging findings reported in adults were extremely rare or not found in the pediatric studies. These discoveries, and other recent reports in this space, support the fact that children’s symptoms may be less obvious than adults or even absent, but they still carry the virus and may be at risk for serious and life-threatening illness.

Marius George Linguraru

Marius George Linguraru, D.Phil., M.A., M.Sc., principal investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National.

Q: How will the findings of this study benefit pediatric care?

A: In our study, we showed how the health of the lungs of these children is impacted. Our results from data from 1,026 children (from newborns to 18 year old) with COVID-19 present chest manifestations that could potentially prompt informed intervention and better recovery.

Another conclusion of our study is that the abnormalities reported on the chest scans of children infected with COVID-19 are distinct from the typical lung images seen during other viral respiratory infections in the pediatric population. This is important for preparing for the cold and flu season.

Q: Why was this review important to our understanding of how COVID-19 impacts children?

A: This is the first systematic review and meta-analysis focused on the manifestation of the COVID-19 infection in the lungs of children. Our study, and others from colleagues at Children’s National, helps lead the efforts on elucidating how the pandemic affects the health of children.

Though children were initially thought to be less susceptible to infection, the data has made it clear that many children are at high risk for hospitalization and severe health complications. Although there are similarities between how children and adults are affected by the pandemic, there are also critical differences.

Given the limited knowledge in the manifestation of COVID-19 in children, with children susceptible to infection and hospitalization, and with children returning to school, continued efforts to understand the impact of COVID-19 on young patients is critically important. Understanding how children fare through the pandemic is the foundation of discovering better ways to take care of young patients and their health.

You can find the full study published in Pediatric Pulmonology. Learn more about the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National.

$1M grant funds research on quantitative imaging for tumors

/in , , , /by

“For children who are at risk of losing their vision, this project will bring a window of opportunity for physicians to start treatment earlier and save their vision,” says Marius George Linguraru, DPhil, MA, MSc.

A team from Children’s National Hospital is part of a project receiving a two-year grant of nearly $1,000,000 from the National Institutes of Health (NIH) for the first pediatric project in the Quantitative Imaging Network (QIN) of the National Cancer Institute (NCI). Marius George Linguraru, DPhil, MA, MSc, principal investigator from the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National Hospital in Washington, D.C., is one of two principal investigators on the project, which focuses on developing quantitative imaging (QI) tools to improve pediatric tumor measurement, risk predictions and treatment response. Roger Packer, M.D., Senior Vice President of the Center for Neuroscience & Behavioral Health, Director of the Gilbert Neurofibromatosis Institute and Director of the Brain Tumor Institute, is co-investigator.

The project, in collaboration with Children’s Hospital of Philadelphia and Children’s Hospital Colorado, centers on the most common type of brain tumor in children, called a low-grade glioma. This project focuses on a clinically challenging group of children with neurofibromatosis type 1 (NF1), the most common inherited tumor predisposition syndrome. Nearly 20% of children with NF1 will develop a low-grade glioma called optic pathway glioma (OPG). In children with this type of brain tumor, the growth occurs around the optic nerve, chiasm and tracts, also called the optic pathway, which connects the eye to the brain. OPGs can cause vision loss and even blindness. Permanent vision loss usually occurs between one and eight years of age with doctors closely monitoring the tumor with magnetic resonance imaging (MRI) to assess the disease progression.

“Our traditional two-dimensional measures of tumor size are not appropriate to assess the changes in these amorphous tumors over time or how the tumor responds to treatment,” says Linguraru. “This means physicians have difficulty determining the size of the tumor as well as when treatment is working. Research such as this can lead to innovative medical technologies that can improve and possibly change the fate of children’s lives.”

Dr. Linguraru is leading the technical trials on this project, which take place in the first two years, or phase one, starting in June 2020. Phase one focuses on improving the often inaccurate human measurements of tumor size by developing QI tools to make precise and automated measures of tumor volume and shape using machine learning. In this phase, the project will use and homogenize MRI data from multiple centers to develop predictive models of the treatment response based on the tumor volume that are agnostic to the differences in imaging protocols. By doing this, it will allow physicians to make more informed decisions about the treatment’s success and whether the child will recover their vision.

When phase one is complete, Linguraru and the project’s other principal investigator Robert A. Avery, DO, MSCE, neuro-ophthalmologist in the Division of Ophthalmology at Children’s Hospital of Philadelphia, will initiate the second phase, which includes validating the QI application on data from the first ever phase III clinical trial comparing two treatments for NF1-OPGs. Phase two is scheduled to start in the Summer 2022 and continue through Summer 2025.

“For children who are at risk of losing their vision, this project will bring a window of opportunity for physicians to start treatment earlier and save their vision,” says Linguraru. “For those children who won’t benefit from chemotherapy because the tumor poses no threat to their sight, this project will save them from having to go through that difficult treatment unnecessarily. It will be life-changing for the children and their families, which is what excites me about this QI application.”

This project is a collaboration between Children’s Hospital of Philadelphia and Children’s National Hospital in Washington, D.C., in partnership with Children’s Hospital of Colorado and University of Pennsylvania. Upon project completion, the QI application will provide a precision-medicine approach for NF1-OPGs and improve clinical outcomes for pediatric tumors.

kidney ultrasound

Using computers to enhance hydronephrosis diagnosis

/in , , , /by
kidney ultrasound

Researchers at Children’s National Hospital are using quantitative imaging and machine intelligence to enhance care for children with a common kidney disease, and their initial results are very promising. Their technique provides an accurate way to predict earlier which children with hydronephrosis will need surgical intervention, simplifying and enhancing their care.

We live in a time of great uncertainty yet great promise, particularly when it comes to harnessing technology to improve lives. Researchers at Children’s National Hospital are using quantitative imaging and machine intelligence to enhance care for children with a common kidney disease, and their initial results are very promising. Their technique provides an accurate way to predict earlier which children with hydronephrosis will need surgical intervention, simplifying and enhancing their care.

Hydronephrosis means “water in the kidney” and is a condition in which a kidney doesn’t empty normally. One of the most frequently detected abnormalities on prenatal ultrasound, hydronephrosis affects up to 4.5% of all pregnancies and is often discovered prenatally or just after birth.

Although hydronephrosis in children sometimes resolves by itself, identifying which kidneys are obstructed and more likely to need intervention isn’t particularly easy. But it is critical. “Children with severe hydronephrosis over long periods of time can start losing kidney function to the point of losing a kidney,” says Marius George Linguraru, DPhil, MA, MSc, principal investigator of the project; director of Precision Medical Imaging Group at the Sheikh Zayed Institute for Pediatric Surgical Innovation; and professor of radiology, pediatrics and biomedical engineering at George Washington University.

Children with hydronephrosis face three levels of examination and intervention: ultrasound, nuclear imaging testing called diuresis renogram and surgery for the critical cases. “What we want to do with this project is stratify kids as early as possible,” Dr. Linguraru says. “The earlier we can predict, the better we can plan the clinical care for these kids.”

Ultrasound is used to see whether there is a blockage and try to determine hydronephrosis severity. “Ultrasound is non-invasive, non-radiating, and does not expose the child to any risk prenatally or postnatally,” Dr. Linguraru says. Ultrasound evaluations require a trained radiologist, but there’s a lot of variability. Radiologists have a grading system based on the ultrasound appearance of the kidney to determine whether the hydronephrosis is mild, moderate or severe, but studies show this isn’t predictive of longer term outcomes.

Children whose ultrasounds show concern will be referred to diuresis renogram. Costly, complex, invasive and irradiating, it tests how well the kidney empties. Although appropriate for good clinical indications, doctors try to minimize its use. “Management of hydronephrosis is complex,” Dr. Linguraru says. “We want to use ultrasound as much as possible and much less diuresis renogram.”

For those patients whose kidney is obstructed and eventually need surgical intervention, the sooner that decision can be made the better. “The more you wait for a kidney that is severely obstructed, the more function may be lost. If intervention is required, it’s preferable to do it early,” Dr. Linguraru says. Of course for the child whose hydronephrosis will likely resolve itself, intervention is not the best option.

Marius George Linguraru

“With our technique we are measuring physiological and anatomical changes in the ultrasound image of the kidney,” says Marius George Linguraru, DPhil, MA, MSc. “The human eye may find it difficult to put all this together, but the machine can do it. We use quantitative imaging to do deep phenotyping of the kidney and machine learning to interpret the data.”

Dr. Linguraru and the multidisciplinary team at Children’s National Hospital, including radiology and urology clinicians, are putting the power of computers to work interpreting subtleties in the ultrasound data that humans just can’t see. In their pilot study they found that 60% of the nuclear imaging tests could have been safely avoided without missing any of the critical cases of hydronephrosis. “With our technique we are measuring physiological and anatomical changes in the ultrasound image of the kidney,” Dr. Linguraru says. “The human eye may find it difficult to put all this together, but the machine can do it. We use quantitative imaging to do deep phenotyping of the kidney and machine learning to interpret the data.”

Results of the initial study indicate that kids who have a mild condition can be safely discharged earlier and the model can predict all those kids with obstructions and accelerate their diagnosis by sending them earlier to get further investigation. Dr. Linguraru says. “There are only benefits: some kids will get earlier diagnosis, some earlier discharges.”

The team also has a way to improve the interpretation of diuresis renograms. “We analyze the dynamics of the kidney’s drainage curve in quantifiable way. Using machine learning to interpret those results, we showed we can potentially discharge some kids earlier and accelerate intervention for the most severe cases instead of waiting and repeating the invasive tests,” he says. The framework has 93% accuracy, including 91% sensitivity and 96% specificity, to predict surgical cases, a significant improvement over clinical metrics’ accuracy.

The next step is a study connecting all the protocols. “Right now we have a study on ultrasound, a study on nuclear imaging, but we need to connect them so a child with hydronephrosis immediately benefits,” says Dr. Linguraru. Future work will focus on streamlining and accelerating diagnosis and intervention for kids who need it, both in prospective studies and hopefully clinically as well.

Hydronephrosis is an area in which machine learning can be applied to pediatric health in meaningful ways because of the sheer volume of cases.

“Machine learning algorithms work best when they are trained well on a lot of data,” Dr. Linguraru says. “Often in pediatric conditions, data are sparse because conditions are rare. Hydronephrosis is one of those areas that can really benefit from this new technological development because there is a big volume of patients. We are collecting more data, and we’re becoming smarter with these kinds of algorithms.”

Learn more about the Precision Medical Imaging Laboratory and its work to enhance clinical information in medical images to improve children’s health.

Marius George Linguraru

Marius George Linguraru, D.Phil., M.A., M.Sc., awarded Department of Defense grant for Neurofibromatosis application development

/in , , , /by
Marius George Linguraru

Marius George Linguraru, D.Phil., M.A., M.Sc., is a principal investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National, where he founded and directs the Precision Medical Imaging Laboratory. He’s an expert in quantitative imaging and artificial intelligence.

Marius George Linguraru, D.Phil., M.A., M.Sc., a principal investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National has been awarded a Congressionally Directed Medical Research Program (CDMRP) grant through the Department of Defense. This grant allows Dr. Linguraru to develop a novel quantitative MRI application that can inform treatment decisions by accurately identifying which children with Neurofibromatosis type 1 (NF1) and optic pathway glioma (OPG) are at risk of losing their vision.

This grant is part of the Neurofibromatosis Research Program of the CDMRP, which fills research gaps by funding high impact, high risk and high gain projects. Dr. Linguraru, who directs the Precision Medical Imaging Laboratory in the Sheikh Zayed Institute, is collaborating with the Gilbert Family Neurofibromatosis Institute and the Children’s Hospital of Philadelphia on this project.

An expert in quantitative imaging and artificial intelligence, Dr. Linguraru has published several peer-reviewed studies on NF1 and OPG, a tumor that develops in 20 percent of children with NF1. The OPG tumor can cause irreversible vision loss, leading to permanent disability in about 50 percent of children with the tumor. This project, titled “MRI Volumetrics for Risk Stratification of Vision Loss in Optic Pathway Gliomas Secondary to NF1” will provide doctors certainty when identifying which children with NF1-OPG will lose vision and when the vision loss will occur.

Dr. Linguraru and his team will validate the quantitative MRI application that they’re developing by studying children at 25 NF1 clinics from around the world. Doctors using the application, which will perform comprehensive measurements of the OPG tumor’s volume, shape and texture, will upload their patient’s MRI into Dr. Linguraru’s application. Using recent advances in quantitative image analysis and machine learning, the application will then definitively determine whether the child’s NF1-OPG is going to cause vision loss and therefore requires treatment.

This diagnosis can occur before visual acuity starts to decline, which provides an opportunity for early treatment in children at risk for vision loss. Dr. Linguraru believes that early diagnosis and treatment can help to avoid lifelong visual impairment for these patients while preventing unnecessary MRIs and aggressive chemotherapy in pediatric patients who are not at risk of vision loss.

Occurring in one in 3,000 to 4,000 live births, NF1 is a genetic condition that manifests in early childhood and is characterized by changes in skin coloring and the growth of tumors along nerves in the skin, brain and other parts of the body. It is unknown why the OPG tumor caused by NF1 only results in vision loss for 50 percent of children. Some children will sustain lifelong disability from their vision loss, despite receiving treatment for their tumor, likely because treatment was started late. In other instances, doctors are unknowingly treating NF1-OPGs that would never cause vision loss.

Dr. Linguraru and his team have already proven that their computer-based, quantitative imaging measures are more objective and reliable than the current clinical measures, enabling doctors to make earlier and more accurate diagnoses and develop optimal treatment plans.

Antonio R. Porras

Antonio R. Porras, Ph.D., awarded prestigious NIH grant for craniosynostosis modeling, career advancement

/in , , /by
Antonio R. Porras

Antonio R. Porras, Ph.D., is a staff scientist in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National Health System.

Antonio R. Porras, Ph.D., a staff scientist in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National Health System, has received the prestigious Pathway to Independence Award from the National Institutes of Health (NIH). This award funds Dr. Porras’ research for the next five years, enabling him to develop two bone growth models that will better inform clinicians treating patients with craniosynostosis and help to optimize outcomes. Also referred to as the K99/R00 grant, this NIH award is for researchers who are either in the postdoctoral/residency period or who are early career investigators. It is designed to transition them from mentored positions to independent, tenure-track or equivalent faculty positions so that they can launch competitive research careers.

Marius George Linguraru, D.Phil., M.A., M.Sc., a principal investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation, is Dr. Porras’ primary mentor on this research project along with co-mentors Robert Keating, M.D., division chief of neurosurgery at Children’s National, and Maximilian Muenke, M.D., chief in the Medical Genetics Branch at the National Human Genome Research Institute.

Dr. Porras has taken a research interest in craniosynostosis, the early fusion of one or more cranial sutures that may lead to craniofacial malformations and brain growth constraints during childhood. With this NIH K99/R00 award, Dr. Porras will employ his expertise in computer science, biomedical engineering, quantitative imaging and statistical modeling to create a personalized computational predictive model of cranial bone growth for subjects without cranial pathology and for patients with craniosynostosis. Dr. Porras will also quantify the coupled growth patterns of the cranial bones and the brain using an existing brain growth model.

Affecting one in 2,100 to 2,500 live births, craniosynostosis complications can result in elevated intra-cranial pressure and subsequent impaired brain growth. While treatable, optimal outcomes are stymied by subjectivity in the evaluation of cranial malformations and prediction of cranial bone development. There are currently no personalized clinical tools available to predict healthy or pathological cranial growth and no objective techniques to optimize the long-term outcome of treatment for patients with craniosynostosis.

photos used for facial analysis technology

Facial analysis technology successful in identifying Williams-Beuren syndrome in diverse populations

/in , /by
photos used for facial analysis technology

Image Credit: Darryl Leja, NHGRI.

In an international study led by the National Human Genome Research Institute (NHGRI), researchers have successfully identified Williams-Beuren syndrome in diverse populations using clinical information and objective facial analysis technology developed by the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National.

The technology, which was featured by STAT as an ‘Editor’s Pick’ finalist in their recent competition to find the best innovation in science and medicine, enables users to compare the most relevant facial features characteristic of Williams-Beuren syndrome in diverse populations.

Williams-Beuren syndrome affects an estimated 1 in 7,500 to 10,000 people, with the most significant medical problems being cardiovascular, including high blood pressure. Though the syndrome is a genetic condition, most cases are not inherited. Signs and symptoms include intellectual disability and distinctive facial features including puffiness around the eyes, a short nose with a broad tip, full cheeks and a wide mouth with full lips.

Using the facial analysis technology, the researchers compared 286 African, Asian, Caucasian and Latin American children and adults with Williams-Beuren syndrome with 286 people of the same age, sex and ethnicity without the disease. They were able to correctly identify patients with the disease from each ethnic group with 95 percent or higher accuracy.

“Our algorithm found that the angle at the nose root is the most significant facial feature of the Williams-Beuren syndrome in all ethnic groups and also highlighted facial features that are relevant to diagnosing the syndrome in each group,” said Marius George Linguraru, D.Phil., developer of the facial analysis technology and an investigator in the study from Children’s National.

Linguraru and his team are working to create a simple tool that will enable doctors in clinics without state-of-the-art genetic facilities to take photos of their patients on a smartphone and receive instant results.

The technology was also highly accurate in identifying Noonan syndrome according to a study published in Sept. 2017, DiGeorge syndrome (22q11.2 deletion syndrome) in April 2017 and Down syndrome in Dec. 2016. The next study in the series will focus on Cornelia de Lange syndrome.

STAT Madness

Voters select Children’s National innovation as runner-up in national competition

/in , , , , , , /by
STAT Madness

Facial recognition technology developed and tested by researchers with the Sheikh Zayed Institute for Pediatric Surgical Innovation and Rare Disease Institute at Children’s National was the runner-up in this year’s STAT Madness 2018 competition.

Facial recognition technology developed and tested by researchers with the Sheikh Zayed Institute for Pediatric Surgical Innovation and Rare Disease Institute at Children’s National was the runner up in this year’s STAT Madness 2018 competition. Garnering more than 33,000 overall votes in the bracket-style battle that highlights the best biomedical advances, the Children’s National entry survived five rounds and made it to the championship before falling short of East Carolina University’s overall vote count.

Children’s entry demonstrates the potential widespread utility of digital dysmorphology technology to diverse populations with genetic conditions. The tool enables doctors and clinicians to identify children with genetic conditions earlier by simply taking the child’s photo with a smartphone and having it entered into a global database for computer analyses.

The researchers partnered with the National Institutes of Health National Human Genome Research Institute and clinicians from 20 different countries to acquire pictures from local doctors for the study. Using the facial analysis technology, they compared groups of Caucasians, Africans, Asians and Latin Americans with Down syndrome, 22q11.2 deletion syndrome (also called DiGeorge syndrome) and Noonan syndrome to those without it. Based on more than 125 individual facial features, they were able to correctly identify patients with the condition from each ethnic group with more than a 93 percent accuracy rate. Missed diagnoses of genetic conditions can negatively impact quality of life and lead to premature death.

Children’s National also was among four “Editor’s Pick” finalists, entries that span a diverse range of scientific disciplines. Journalists at the digital publication STAT pored through published journal articles for 64 submissions in the single-elimination contest to honor a select group of entries that were the most creative, novel, and most likely to benefit the biomedical field and the general public.

Each year, 1 million children are born worldwide with a genetic condition that requires immediate attention. Because many of these children experience serious medical complications and go on to suffer from intellectual disability, it is critical that doctors accurately diagnose genetic syndromes as early as possible.

“For years, research groups have viewed facial recognition technology as a potent tool to aid genetic diagnosis. Our project is unique because it offers the expertise of a virtual geneticist to general health care providers located anywhere in the world,” says Marius George Linguraru, D.Phil., M.A., M.S., a Sheikh Zayed Institute for Pediatric Surgical Innovation principal investigator who invented the technology. “Right now, children born in under-resourced regions of the U.S. or the world can wait years to receive an accurate diagnosis due to the lack of specialized genetic expertise in that region.”

In addition to providing patient-specific benefits, Marshall Summar, M.D., director of Children’s Rare Disease Institute that partners in the facial recognition technology research, says the project offers a wider societal benefit.

“Right now, parents can endure a seemingly endless odyssey as they struggle to understand why their child is different from peers,” says Dr. Summar. “A timely genetic diagnosis can dispel that uncertainty and replace it with knowledge that can speed patient triage and deliver timely medical interventions.”

Children’s National leaders join with Governor Martin O'Malley

Facial analysis technology successfully used to identify Noonan syndrome in diverse populations

/in , /by

facial recognition of noonan syndrome

According to an international study led by the National Human Genome Research Institute (NHGRI), researchers have successfully used facial analysis software, developed by the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National, to identify Noonan syndrome in diverse populations.

Noonan syndrome is relatively common, affecting between 1 in 1,000 to 1 in 2,500 children, however few studies have been conducted in non-Europeans. For this study, the researchers evaluated children (average age of eight) with Noonan syndrome from 20 countries. Using the facial analysis software and clinical criteria, the researchers compared 161 white, African, Asian and Latin American children with Noonan syndrome with 161 people of the same age and gender without the disease. Using the software to analyze facial features, they were able to correctly diagnose patients with the disease from each ethnic group with 94 percent or higher accuracy.

“Our algorithm found widely spaced eyes as a significant facial feature in all ethnic groups and also highlighted facial features that are relevant to diagnosing the syndrome in each group,” said

Marius George Linguraru, D.Phil., developer of the facial analysis technology and an investigator in the study from Children’s National.

Linguraru and his team are working to create a simple tool that will enable doctors in clinics without state-of-the-art genetic facilities to take photos of their patients on a smartphone and receive instant results.

Facial analysis technology helps diagnose rare genetic disease

/in , /by
Facial Analysis Technology

A new study uses facial analysis technology developed at Children’s National to diagnose 22q1.2 deletion syndrome, also known as DiGeorge syndrome.

According to a new study led by the National Human Genome Research Institute (NHGRI), facial analysis technology can assist clinicians in making accurate diagnosis of 22q1.2 deletion syndrome, also known as DiGeorge syndrome. Using objective facial analysis software, developed by researchers from the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National, the study compared the most relevant facial features characteristic of DiGeorge syndrome in diverse populations. Based on a selection of 126 individual facial features, the researchers were able to correctly diagnose patients with the disease from different ethnic groups with 96.6 percent or higher accuracy.

“The results of the study demonstrated that the identification of rare diseases benefits from adapting to ethnic and geographic populations,” said Marius George Linguraru, D.Phil., developer of the facial analysis technology and an investigator in the study from Children’s National.

Linguraru and his team are also working on a simple tool that will enable doctors in clinics without state-of-the-art genetic facilities to take photos of their patients on a smartphone and receive instant results.

How technology can predict vision loss in neurofibromatosis patients

/in , /by

Roger Packer and patient

For the first time, scientists have been able to definitively connect tumor volume and vision loss for children with neurofibromatosis type 1 (NF1). The first study to use quantitative imaging technology to accurately assess the total volume of individual optic nerve glioma (OPG) in NF1 was published in the November 4, 2016 issue of Neurology.

NF1 is a genetic condition that occurs in one in 3,500 births. Children with NF1 develop tumors in multiple locations across the nervous system. About 20 percent of children with NF1 will develop optic pathway gliomas, or tumors that occur in the visual system. Half of those with OPG will have irreversible vision loss, which occurs at a very young age, usually before age 3.

“Neuroradiologists typically assess these tumors through a measurement of the tumor’s radii using magnetic resonance images (MRI) of the patient,” said Marius George Linguraru, D.Phil., M.A., M.S., Principal Investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation at Children’s National Health System, who is senior author on the study.

“These measurements aren’t detailed enough to serve as a good indicator of whether an OPG will cause vision loss for a child. Through automated computerized analysis, however, we’ve taken the MRI data and systematically analyzed the size and shape, as well as documented changes over time, all in 3-D, to pinpoint the volume of each tumor.”

A look inside the study

The study included children with NF1-related OPGs who are currently cared for at the Gilbert Family Neurofibromatosis Institute at Children’s National. Investigators compared the MRI analysis to the patients’ retinal nerve fiber layer (RNFL), a measure of the health of the visual system. The analysis showed a quantifiable negative relationship between increasing tumor volume within the structures of the anterior visual pathway (the optic nerve, chiasm, and tract) and decreasing thickness of the RNFL, indicating damage to the visual system and vision loss.

“Measuring the tumors in a precise, systematic manner, along with knowing how they grow, is the first step in recognizing which children are at highest risk for vision loss and to potentially identifying them before they suffer any visual symptoms,” added Dr. Linguraru. “If we know which children will probably lose vision, we can treat earlier, and perhaps improve how patients respond to treatment.”

A multicenter collaborative study to validate the findings will begin in 2017.