AI projects move forward in collaboration between Virginia Tech, Children’s National Hospital
Organizers seek to build momentum and expand the scope of artificial intelligence to improve pediatric health.
Physician scientists and researchers from Children’s National Hospital and Virginia Tech will increase efforts to harness artificial intelligence (AI) to help children who are struggling with medical conditions.
The innovators will meet in September at the Children’s National Research & Innovation Campus in Washington, D.C.
“It’s clear that harnessing the power of artificial intelligence is the way forward in advancing children's health,” said Lance Collins, vice president and executive director of the Virginia Tech Innovation Campus in Alexandria. “Virginia Tech researchers are building momentum and solidifying our collaborative goals in this important area.”
The effort involves Virginia Tech’s Sanghani Center for Artificial Intelligence and Data Analytics, Children’s National Hospital, and the Fralin Biomedical Research Institute at VTC, which has labs at the Children’s National Research & Innovation Campus.
This meeting builds on the momentum from last year’s workshop, which featured sessions on smart surgery, rare diseases, and emergency medicine with talks by both Virginia Tech and Children’s National faculty and researchers.
“Now we are expanding the scope of this collaboration to more units at Children’s National Hospital and Virginia Tech,” said co-organizer Naren Ramakrishnan, director of the Sanghani Center and the Thomas L. Phillips Professor in the College of Engineering. “We will hear from new groups from Children’s National Hospital, and we will have more Virginia Tech people joining from areas such as security, conversational AI, and federated learning.”
The organizers seek to remove barriers between clinicians and AI scientists.
“The rapid evolution of AI technology is unlocking unprecedented possibilities to transform pediatric health care,” said co-organizer Marius George Linguraru, a global leader in harnessing the power of imaging and machine learning to advance children’s health and the Connor Family Professor of Research and Innovation at Children’s National. “AI’s potential to offer life-changing solutions for children with rare medical conditions is immense, and it’s essential that we collaborate with clinicians, AI scientists and partner organizations to tap into this potential. Together, we must craft AI tools specifically designed for the unique needs of children—beyond simply adapting models built for adults—to shape the future of pediatric medicine."
Collaboration between clinicians and AI scientists resulted after Michael Friedlander, vice president for health sciences and technology at Virginia Tech, introduced the leadership of the Sanghani Center to teams at Children's National Hospital.
It set the stage for further exploration of how this technology can be used to help children and adults.
“We are taking the next steps to explore how new technology can be incorporated into clinical practice to augment our intelligence and decision-making on the diagnostics, therapeutics and implementation frontiers,” Friedlander said. “AI-based tools have dramatically enhanced our ability to understand complex health data to benefit patients, and it will increasingly be used to understand someone’s personal health data to predict and eventually prevent a problem long before it arises.”
This year’s session will also provide updates on the progress of five projects that were jointly supported by Virginia Tech and Children’s National, including:
- Predicting single-cell responses to genetic perturbations in pediatric developmental disorders: AI models predicting how single cells respond to genetic changes could help overcome challenges in studying pediatric developmental disorders, especially those involving rare cell types. These models would pinpoint potential treatments, filling gaps in current approaches to pediatric genetic diseases. Principal investigators are Wei Li, assistant professor, Center for Genetic Medicine, Children’s National Hospital, and Jia-Ray Yu, assistant professor at Virginia Tech’s Fralin Biomedical Research Institute Cancer Research Center — D.C.
- Forecasting emergency department surges: Emergency department crowding leads to surges in patient numbers, system breakdowns, lower satisfaction, and higher rates of patients who leave without being seen. The proposed solution is to develop forecasting models to predict emergency room surges to use backup resources more effectively. Principal investigators are Kenneth McKinley, assistant professor at Children’s National Hospital, and Patrick Butler, a senior research associate at Virginia Tech’s Sanghani Center.
- Improving accuracy in identifying rare genetic syndromes in children through generative models: Identifying rare genetic syndromes in children is challenging. Researchers propose using facial analysis and diffusion models, a type of technology that’s good at making realistic images with little data, to simulate disease traits and better detect and classify genetic syndromes. Co-principal investigators are Yanardag Delul, assistant professor in the Department of Computer Science at Virginia Tech, and Xinyang Liu, staff scientist at Precision Medical Imaging lab of Children's National Hospital.
- Rethinking privacy in federated learning: Sharing data is crucial for training large-scale deep learning models in health care, but privacy concerns hinder the practice, especially in pediatric health care involving rare diseases, where datasets are limited. This project proposes a federated-learning approach, where individual patients can collaboratively train a large deep-learning model without sharing their individual data. Principal investigators are Wenjing Lou, the W. C. English Endowed Professor of Computer Science, Virginia Tech, and Syed Muhammad Anwar at the Sheikh Zayed Institute for Pediatric Surgical Innovation, Children’s National Hospital.
- Weakly-supervised clinical variables extraction for sepsis research with large language models: Pediatric sepsis is a major cause of child mortality worldwide and requires advanced strategies for to predict and prevent. This project aims to develop a method to automatically extract clinical variables from documents, radiology reports, and pediatric emergency provider notes for better prediction of sepsis risks. Principal investigators are Xuan Wang, assistant professor of computer science at Virginia Tech, and Ioannis Koutroulis, research director of emergency medicine at Children’s National Hospital.