CTSA News

A team from the University of Rochester has developed a time-series AI machine learning model that better predicts the lung disease Bronchopulmonary Dysplasia (BPD) in newborns than current tools. They shared their work in “Time-Series Machine Learning for Prediction of Bronchopulmonary Dysplasia” published in The Journal of Pediatrics.

BPD in Newborns

Newborns with BPD suffer from long-term respiratory function issues resulting from premature birth and low birth weight. Their underdeveloped lungs are exposed to oxygen via mechanical ventilation too early in their development, contributing to their long-term health challenges.

“We take great effort in the neonatal intensive care unit to prevent lung damage,” said Associate Professor Andrew Dylag, M.D., Department of Pediatrics, Neonatology. “Despite this, premature infants still develop BPD. There are BPD ‘calculators’ on the internet that can predict the severity of lung disease while the baby is still in the hospital, but they use a very limited set of data.”

A recent update to these calculators resulted in a decline in accuracy compared to previous versions, inspiring the team to develop another approach.

“We thought that using more detailed data from the University’s electronic health record would improve disease predictions and allow us to pinpoint vulnerable times when we might be able to intervene to prevent lung disease in newborns,” Dylag said.

The team sought out a Digital Health Seedling award from the Clinical and Translational Science Institute in 2023 to support this project.

Interest in AI Leads to Collaboration

“We hoped that CTSI could help us test the hypothesis that machine learning could improve disease predictions in hospitalized premature newborns,” Dylag said. “The 2023 Digital Health Seedling award was exactly the type of funding we needed to develop new collaborations across the University community and kickstart our team's academic interactions.”

With the pilot award, the research team grew to include Albert Arendt Hopeman Professor of Engineering Jiebo Luo, Ph.D., of the Department of Computer Science, who connected several graduate students to the project; and Professor Xing Qiu, Ph.D., of the Department of Biostatistics and Computational Biology.

“The neonatology team brought content and clinical expertise to the work, the computer science team developed and tested the models, and the biostatisticians ensured the rigor and testing of the models and algorithms,” Dylag said.

This effort also required the expertise of UR CTSI’s Informatics and Analytics group, building upon an established collaboration on previous projects involving the electronic health record.

Leveraging the Secure Environment for Research Data Analytics

“We initially got involved by helping the study team pull clinical data from eRecord,” said Jack Chang, Ph.D., associate director of Research Informatics. “Recognizing the project involved a very large patient population and massive data including Protected Health Information, we identified a need for a more secure analytical workspace.”

To ensure both regulatory compliance and accessibility, the Informatics team facilitated the data transition into Secure Environment for Research Data Analytics (SERDA). High-risk clinical data use has a high burden of administrative and security oversight to protect patient privacy and ensure its correct application.

“SERDA removes that obstacle by providing a secure, scalable, and 'ready-to-go' environment tailored for advanced analytics and machine learning,” Chang said. “It allows researchers to focus on their science while knowing their data is protected and compliant with all privacy regulations.”

The Informatics team had the expertise needed to support the team’s goals.

“Our team—working with our ISD partners—handled the technical heavy lifting: setting up virtual machines, configuring project shares, ensuring secure access with the security team, and customizing the environment with specialized analytical software,” Chang said. “We also provided training and facilitated the numerous exports of analytical outcomes from SERDA for their publication.”

Improving Care for Newborns

The improved model that the research team developed is more likely to select infants that are most likely to benefit from the precise diagnostic accuracy and treatment required to address BPD.

“We want to build clinical decision support tools to identify how disease predictions change in real time,” Dylag said. “If we validate our algorithm and can present the disease prediction to the clinical team, we can test guideline implementation for how to manage or treat infants that may reduce BPD severity.”

Further collaboration across departments and disciplines—as well as infrastructure supported by CTSI and ISD—will help the team move the research forward into the next phase.