CHICAGO — The University of Chicago Medicine will establish a new Section of Computational Neurosurgery on Sept. 1, bringing together clinical and scientific expertise in neurosurgery, artificial intelligence and data science to develop new approaches to the diagnosis and treatment of neurological disease.
The section, the first of its kind in the nation, will be led by Antonio Porras, PhD, scientific director, and Rohaid Ali, MD, clinical director. They will collaborate with neurosurgeons, computer scientists, engineers, data scientists and other researchers across the University of Chicago to translate computational methods into clinical care and scientific discovery.
“This section brings together two things that are essential to the future of neurosurgery: deep clinical expertise and sophisticated computational science,” said Mohamad Bydon, MD, Ginni & Mark Rometty Chair of the Department of Neurological Surgery and Stahl Professor of Neuroscience. “Our goal is to develop technologies that augment the expertise of our surgeons, help us make better-informed decisions, and improve surgical outcomes for our patients.”
The section will focus on applications of artificial intelligence, machine learning, medical image computing and predictive modeling across the continuum of neurosurgical care — from diagnosis and treatment planning to surgery and long-term outcomes.
“Neurosurgery generates an extraordinary amount of complex data, from medical images to clinical histories and long-term outcomes,” said Porras. “The opportunity is to bring together that information with computational methods to identify patterns, make more precise predictions and answer questions that have been difficult to address through traditional approaches. At UChicago, we have the great advantage of being able to do that by bringing computational scientists and clinicians together around real problems in patient care.”
Porras directs UChicago's Neural Image Computing Lab and brings expertise in medical image computing, machine learning and quantitative approaches to understanding disease and predicting patient outcomes.
“The most promising applications of AI in neurosurgery are those that can help physicians make better decisions and give patients and their loved ones a clearer understanding of their care,” said Ali. “We are not looking to replace clinical judgment; we are looking to give surgeons better information and better tools. The goal is to translate these technologies into meaningful improvements in how we diagnose, plan and deliver care.”
The section also will leverage UChicago's broader strength in interdisciplinary research. Its faculty will collaborate with experts across the university, including researchers in the Data Science Institute, computer science, statistics, neuroscience and other fields.
The University of Chicago's computational research ecosystem includes advanced data-science and biomedical informatics capabilities, large-scale clinical and biomedical data resources, and access to Argonne National Laboratory, one of the nation’s leading supercomputing facilities. For example, the university's Center for Translational Data Science operates a data ecosystem comprising 17 data commons and meshes, making more than 27 petabytes of data from nearly 2.4 million patients available to researchers through secure, compliant workspaces. UChicago researchers also have collaborated with Argonne and other institutions on the Sociome Data Commons, which integrates clinical data with social, environmental and other information to support research into the factors that shape health outcomes.
“We are fortunate to have an unparalleled computational infrastructure at University of Chicago,” said Bydon. “Our new section will leverage the university’s nation-leading resources to solve vexing problems in neurosurgery and neuroscience.”
“What makes this particularly exciting at UChicago is the depth of expertise we can bring to the same problem,” Porras said. “We can have a neurosurgeon, a computer scientist, a statistician and a biomedical researcher looking at a clinical challenge together. That kind of collaboration allows us to move from an interesting algorithm to something that can be tested, validated and ultimately used to help patients.”
The new section builds on the Department of Neurological Surgery’s expanding clinical and research programs while creating a dedicated home for computational approaches to neurosurgery.
“If we do this well, success won't simply be measured by the number of AI models we develop or papers we publish,” Ali said. “It will be measured by whether we can help physicians make better decisions, identify disease earlier, personalize treatment and improve outcomes for patients.”
