CTSA-affiliated researchers’ application of AI leads to successful neonatal feeding interventions

Nature Medicine has published the results of a Stanford study where artificial intelligence was used to develop precise total perinatal nutrition (TPN) formulas for premature newborns. The laboratory of Nima Aghaeepour, PhD led the way in utilizing AI technology to the develop the formulas, and the study was conceived and conducted at Stanford Children’s Health’s NICUs.

“This study is an exciting step toward showing how AI could help us improve nutrition for premature infants,” stated Dr. Aghaeepour.

The research team developed a data-driven approach to optimize and standardize nutrition (through GI feeding tube, intravenous, or by mouth) for preterm babies using information collected routinely in electronic health records. This TPN 2.0 approach used machine learning (ML) to reduce subjectivity by standardizing feeding formulas, which could potentially be mass-manufactured. The TPN2.0 approach also increases efficiency by making recommendations based on a patient’s clinical characteristics.

The team performed external validations with a second hospital cohort at UCSF where they tested the TPN2.0 approach with a multidisciplinary healthcare team. The TPN2.0 approach identified 15 formulas with the potential to increase safety and reduce costs.

“I am proud of the Stanford team’s dedication to building and testing these tools, and grateful to our colleagues at UCSF who enabled complete independent validation of the model,” said Dr. Aghaeepour. “While this work is still in the research phase, it highlights the potential of data-driven approaches to make care for our most vulnerable patients safer and more proactive.”

Dr. Aghaeepour is an alumnus of the Spectrum CTSA’s SPARK program, which funded the initial research, and a former Stanford Biodesign Faculty Fellow. He also participated in Spectrum’s KL2 program as an Education Affiliate.

The study received a special mention from the national NCATS/CTSA leadership.

“This AI application that packs a decade’s worth of expert experience regarding nutrition for infants in the NICU and makes it available for every hospital to use is a great example of CTSAs delivering 21st century medicine to those who need it most,” stated Michael Kurilla, M.D., Ph.D., director of NCATS’ Division of Clinical Innovation, which oversees the national CTSA program.

Read the article in Nature Medicine. Study results were also publicized on the NIH/NCATS news website.