Tag Archive for: Marius Linguraru

Marius Linguraru, D.Phil., M.A., M.Sc., a co-principal investigator for the project, presents

Children’s National joins team to use AI to expand health knowledge in Kenya

Marius Linguraru, D.Phil., M.A., M.Sc., a co-principal investigator for the project, presentsChildren’s National Hospital is joining a team of global health researchers to use large language models (LLMs) like ChatGPT to help Kenyan youth learn about their health and adopt lifestyles that may prevent cancer, diabetes and other non-communicable diseases.

The work, which is one of nearly 50 Grand Challenges Catalyzing Equitable Artificial Intelligence (AI) Use grants announced by the Bill & Melinda Gates Foundation, will harness the emerging power of AI to empower young people with information that they can carry through adulthood to reduce rates of unhealthy behaviors including physical inactivity, unhealthy diet and use of tobacco and alcohol.

“We are thrilled to be part of this effort to bring our AI expertise closer to young patients who would benefit dramatically from technology and health information,” said Marius George Linguraru, D.Phil., M.A., M.Sc., a co-principal investigator for the project, the Connor Family Professor in Research and Innovation at Children’s National and principal investigator in the Sheikh Zayed Institute for Pediatric Surgical Innovation. “Using generative AI, we will build an application to enhance the knowledge, attitudes and healthy habits of Kenyan youth and use this as a foundation to improve health inequities around the globe.”

Why it matters

A lower middle-income country located on the east coast of Sub-Saharan Africa, Kenya is home to 50 million people and one of the continent’s fastest-growing economies. English is one of Kenya’s official languages, and the country has been recognized as a technology leader in Africa, with 82% of Kenyans having phone connectivity. Taken together, these factors make the country an ideal location to deploy an LLM-based platform designed to improve health information and attitudes.

The Gates Foundation selected this project from more than 1,300 grant applications. The nearly 50 funded projects are aimed at supporting low- and middle-income countries to harness the power of AI for good and help countries participate in the AI development process. The project’s findings will contribute to building an evidence base for testing LLMs that can fill wide gaps in access and equitable use of these tools. Each of the grants provides an opportunity to mitigate challenges experienced by communities, researchers and governments.

What’s next

The project development will be led by the National Cancer Institute of Kenya, with Linguraru and other global experts advising the effort from Kenyan institutions and Stanford University. Researchers plan to enroll youth from universities, shopping malls, markets, sporting events and other high-traffic locations. The study will look at participants’ risk factors and how their attitudes toward healthier lifestyles changed after engaging with the new LLM platform.

“The team is thrilled to be selected as one of the nearly 50 most promising AI proposals in the Gates Foundation Grand Challenge competition, and we look forward to seeing how our work can benefit the health of Kenyan youth,” said Dr. Martin Mwangi, principal investigator for the project and head of the Cancer Prevention and Control Directorate at the National Cancer Institute of Kenya. “If successful, we hope to share this model and the expertise we gain to expand health equity and knowledge to other regions.”

Hyperfine Swoop System

$1.6m grant to boost MRI access globally for maternal, child health

Researchers at Children’s National Hospital are investigating ways to bring more portable and accessible low-field magnetic resonance imaging (MRI) to parts of the world that lack access to this critical diagnostic tool, thanks to a grant from the Bill & Melinda Gates Foundation.

The nearly $1.6 million in funding will enable clinicians to better treat pediatric neurological conditions including ischemic brain injury, hydrocephalus, micro- and macrocephaly and more, using analysis tools that are designed to handle the loss in image quality and related challenges inherent to low-field MRI. The research brings together teams at Children’s National and Children’s Hospital Los Angeles — two organizations with extensive experience in designing processing software tools for pediatric brain MRI analysis and data enhancement.

The patient benefit

“For 30 years, MRI has primarily helped patients in high-income countries. Our team is thrilled by the prospect of expanding this powerful tool to patients coming from a wide range of nations, geographies and socioeconomic backgrounds,” said Marius George Linguraru, D.Phil., M.A., M.Sc., principal investigator at the Sheikh Zayed Institute for Pediatric Surgical Innovation (SZI). “Low-field MRI comes with great advantages including portability at the point of care of patients, lower clinical costs and the elimination of sedation for young children.”

Linguraru and his long-time collaborator, Natasha Lepore, Ph.D., principal investigator at The Saban Research Institute at Children’s Hospital Los Angeles, will analyze data from the brains of children from birth for the maternal and child health studies. The MRI data analyzed will form the basis for future studies of children’s brain anatomy in health and disease.

The big picture

Through the new grant, researchers will develop a suite of tools to help clinicians better analyze data and images from low-field MRI systems. These systems already have been integrated into interventional and observational studies to help characterize early neurodevelopmental patterns and identify drivers of abnormal development. They are also evaluating the efficacy of maternal and infant-focused interventions aimed at improving neurodevelopmental outcomes.

Why we’re excited

At Children’s National, SZI has installed a Hyperfine Swoop system, and Linguraru’s team is creating image enhancement tools tailored to the unique challenges of low-field MRI. Chief among them, conventional processing tools developed over the past several decades remain incompatible with the low-field data and require new software to take full advantage of the diagnostic power of imaging.

The work brings together a prestigious international consortium of scientists and clinicians from around the world to harness the power of computing and expand the reach of diagnostic imaging. Lepore said the team is eager to bring modern medical imaging to parts of the world that have missed its many benefits.

“Children’s brain development in underserved areas can be affected by so many factors, like malnutrition or anemia,” Lepore said. “The software we will design for the Hyperfine scanners will improve research into these factors, so the optimal interventions can be designed. We are excited to bring our expertise to this important and timely project.”

echocardiogram

AI may revolutionize rheumatic heart disease early diagnosis

echocardiogram

Researchers at Children’s National Hospital have created a new artificial intelligence (AI) algorithm that promises to be as successful at detecting early signs of rheumatic heart disease (RHD) in color Doppler echocardiography clips as expert clinicians.

Researchers at Children’s National Hospital have created a new artificial intelligence (AI) algorithm that promises to be as successful at detecting early signs of rheumatic heart disease (RHD) in color Doppler echocardiography clips as expert clinicians. Even better, this novel model diagnoses this deadly heart condition from echocardiography images of varying quality — including from low-resource settings — a huge challenge that has delayed efforts to automate RHD diagnosis for children in these areas.

Why it matters

Current estimates are that 40.5 million people worldwide live with rheumatic heart disease, and that it kills 306,000 people every year. Most of those affected are children, adolescents and young adults under age 25.

Though widely eradicated in nations such as the United States, rheumatic fever remains prevalent in developing countries, including those in sub-Saharan Africa. Recent studies have shown that, if detected soon enough, a regular dose of penicillin may slow the development and damage caused by RHD. But it has to be detected.

The hold-up in the field

Diagnosing RHD requires an ultrasound image of the heart, known as an echocardiogram. However, ultrasound in general is very variable as an imaging modality. It is full of texture and noise, making it one of the most challenging to interpret visually. Specialists undergo significant training to read them correctly. However, in areas where RHD is rampant, people who can successfully read these images are few and far between. Making matters worse, the devices used in these low resource settings have their own levels of varying quality, especially when compared to what is available in a well-resourced hospital elsewhere.

The research team hypothesized that a novel, automated deep learning-based method might detect successfully diagnose RHD, which would allow for more diagnoses in areas where specialists are limited. However, to date, machine learning has struggled the same way the human eye does with noisy ultrasound images.

Children’s National leads the way

Using approaches that led to successful objective digital biometric analysis software for non-invasive screening of genetic disease, researchers at the Sheikh Zayed Institute for Pediatric Surgical Innovation, including medical imaging scientist Pooneh Roshanitabrizi, Ph.D., and Marius Linguraru, D.Phil., M.A., M.Sc., principal investigator, partnered with clinicians from Children’s National Hospital, including Craig Sable, M.D., associate chief of Cardiology and director of Echocardiography, and cardiology fellow Kelsey Brown, M.D., who are heavily involved in efforts to research, improve treatments and ultimately eliminate the deadly impacts of RHD in children. The collaborators also included cardiac surgeons from the Uganda Heart Institute and cardiologists from Cincinnati Children’s Hospital Medical Center.

Dr. Linguraru’s team of AI and imaging scientists spent hours working with cardiologists, including Dr. Sable, to truly understand how they approach and assess RHD from echocardiograms. Building the tool based on that knowledge is why this tool stands apart from other efforts to use machine-learning for this purpose. Orienting the approach to the clinical steps of diagnosis is what led to the very first deep learning algorithm that diagnoses mild RHD with similar success to the specialists themselves. After the platform was built, 2,136 echocardiograms from 591 children treated at the Uganda Heart Institute fed the learning algorithm.

What’s next

The team will continue to collect data points based on clinical imaging data to refine and validate the tool. Ultimately, researchers will look for a way that the algorithm can work directly with ultrasound/echocardiogram machines. For example, the program might be run through an app that sits on top of an ultrasound device and works on the same platform to communicate directly with it, right in the clinic. By putting the two technologies together, care providers on the ground will be able to diagnose mild cases and prescribe prophylactic treatments like penicillin in one visit.

The first outcomes from the program were showcased in a presentation by Dr. Roshanitabrizi at one of the biggest and most prestigious medical imaging and AI computing meetings — the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI).

Coronavirus and lungs with world map in the background

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

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.”

DNA Molecule

Decoding cellular signals linked to hypospadias

DNA Molecule

“By advancing our understanding of the genetic causes and the anatomic differences among patients, the real goal of this research is to generate knowledge that will allow us to take better care of children with hypospadias,” Daniel Casella, M.D. says.

Daniel Casella, M.D., a urologist at Children’s National, was honored with an AUA Mid-Atlantic Section William D. Steers, M.D. Award, which provides two years of dedicated research funding that he will use to better understand the genetic causes for hypospadias.

With over 7,000 new cases a year in the U.S., hypospadias is a common birth defect that occurs when the urethra, the tube that transports urine out of the body, does not form completely in males.

Dr. Casella has identified a unique subset of cells in the developing urethra that have stopped dividing but remain metabolically active and are thought to represent a novel signaling center. He likens them to doing the work of a construction foreman. “If you’re constructing a building, you need to make sure that everyone follows the blueprints.  We believe that these developmentally senescent cells are sending important signals that define how the urethra is formed,” he says.

His project also will help to standardize the characterization of hypospadias. Hypospadias is classically associated with a downward bend to the penis, a urethra that does not extend to the head of the penis and incomplete formation of the foreskin. Still, there is significant variability among patients’ anatomy and to date, no standardized method for documenting hypospadias anatomy.

“Some surgeons take measurements in the operating room, but without a standardized classification system, there is no definitive way to compare measurements among providers or standardize diagnoses from measurements that every surgeon makes,” he adds. “What one surgeon may call ‘distal’ may be called ‘midshaft’ by another.” (With distal hypospadias, the urethra opening is near the penis head; with midshaft hypospadias, the urethra opening occurs along the penis shaft.)

“By advancing our understanding of the genetic causes and the anatomic differences among patients, the real goal of this research is to generate knowledge that will allow us to take better care of children with hypospadias,” he says.

Parents worry about lingering social stigma, since some boys with hypospadias are unable to urinate while standing, and in older children the condition can be associated with difficulties having sex. Surgical correction of hypospadias traditionally is performed when children are between 6 months to 1 year old.

When reviewing treatment options with family, “discussing the surgery and postoperative care is straight forward. The hard part of our discussion is not having good answers to questions about long-term outcomes,” he says.

Dr. Casella’s study hopes to build the framework to enable that basic research to be done.

“Say we wanted to do a study to see how patients are doing 15-20 years after their surgery.  If we go to their charts now, often we can’t accurately describe their anatomy prior to surgery.  By establishing uniform measurement baselines, we can accurately track long-term outcomes since we’ll know what condition that child started with and where they ended up,” he says.

Dr. Casella’s research project will be conducted at Children’s National under the mentorship of Eric Vilain, M.D., Ph.D., an international expert in sex and genitalia development; Dolores J. Lamb, Ph.D., HCLD, an established leader in urology based at Weill Cornell Medicine; and Marius George Linguraru, DPhil, MA, MSc, an expert in image processing and artificial intelligence.