2 Sources
[1]
$3.1 million NIH grant funds AI-generated pediatric head imaging database
University of ChicagoNov 17 2026Reviewed The University of Chicago has received a $3.1 million grant from the National Institutes of Health to create innovative AI tools to generate a database of pediatric head images that will advance research on cranial development while protecting patient
[2]
UChicago receives $3.1M NIH grant to create AI-generated database to support pediatric imaging research | Newswise
The University of Chicago has received a $3.1 million grant from the National Institutes of Health to create innovative AI tools to generate a database of pediatric head images that will advance research on cranial development while protecting patient privacy. The project will be led by Antonio
Share
Copy Link
The National Institutes of Health has awarded the University of Chicago $3.1 million to create a database of synthetic pediatric head images, generated by AI, for research on how children's skulls develop.
2
1

Source: News-Medical
The reason hospitals need a workaround is patient privacy. Real craniofacial images of children often cannot be shared between institutions for research, so each hospital's data stays locked inside its own walls. Antonio Porras, an associate professor of neurological surgery at UChicago who leads the project, said new AI methods built specifically for pediatric imaging are starting to change that. "We have not been able to share this data in the past, but our advances in new pediatric-centric AI methodologies are starting to allow us to generate synthetic datasets that do not compromise privacy," he said.
2
The clinical problem the database addresses is craniosynostosis, a condition in which the fibrous joints of an infant's skull close too early, before the brain has finished forming. Surgeons treat it by reopening the fused bones, and planning the operation means predicting how a rapidly growing skull can develop. Even in the few weeks between the initial scan and surgery, the skull can change.
2
1
Porras will lead the work with four partner institutions. They are Children's National Hospital in Washington, Children's Hospital Los Angeles with the University of Southern California, and Children's Hospital Colorado with the University of Colorado Anschutz Medical Campus. The team will work from one of the largest curated collections of pediatric clinical images. Generative AI will turn that material into realistic synthetic images covering all ages and stages of cranial development.
2
1
The grant funds one of the earliest projects of the university's newly launched Section of Computational Neurosurgery, described in the announcement as the first such section in the country. Porras leads it as scientific director, with Rohaid Ali as clinical director.
2
The project will produce a publicly available platform holding synthetic images of both healthy children and children with craniosynostosis. Porras said the plan runs in sequence: a complete dataset of healthy children's images, then a repository of craniosynostosis patients' images from before and after surgery. The public platform comes last, built with a graphical interface that surgeons can use without specialized technical skills. He said the goal is to let other researchers and surgeons build their own projects and evaluate their own patients using scientifically accurate synthetic datasets.
2
1
The projected benefit for patients is fewer repeat scans. Surgeons at more hospitals, large and small, could plan procedures by comparison against a reference dataset of expected development, rather than taking extra images that expose children to additional radiation.
2
As Ali noted, not every hospital has surgeons with thousands of these procedures behind them.2
Mohamad Bydon, who holds the Ginni & Mark Rometty Chair of Neurological Surgery at UChicago, framed the project as a way to study and predict patient outcomes. "Images are a window into a patient's condition," he said. "This project creates a paradigm that enables us to study and predict outcomes for our patients and ultimately have more effective surgeries and treatments."
2
Ali said the synthetic-imaging method could translate to surgical and disease areas well beyond neurosurgery.
2
"Not everyone works at a large, high-volume academic medical center or has access to experienced surgeons who have done thousands of these procedures," he said. "So, these datasets are going to help guide surgeons everywhere to perform procedures in the most effective way possible."2
Porras said the approach would, for the first time, let surgeons compare the outcomes of different operations to plan the best treatment for each child.2
Related Stories
Summarized by
Navi
08 Oct 2025•Health
25 Apr 2025•Science and Research

02 Aug 2025•Science and Research
