Researchers at Kyung Hee University Medical Center have introduced a new diagnostic concept for neonatal intestinal injury, the hospital announced on Monday.
The research team, led by Professor Choi Yong-sung of the hospital's Department of Pediatrics, presented the findings at the “2026 Pediatric Academic Societies Meeting” held in Boston, Mass., on April 27.
Kyung Hee University Medical Center said Monday that a research team led by Professor Choi Yong-sung of the Department of Pediatrics proposed a new diagnostic concept for neonatal intestinal injury. (Courtesy of Kyung Hee University Medical Center)
Bell’s staging system has long been used as the standard diagnostic criterion for necrotizing enterocolitis (NEC) in newborns. However, the system has limitations because it relies heavily on clinicians’ subjective judgment, and symptoms are often ambiguous, making it difficult to distinguish between early-stage disease and rapidly deteriorating conditions.
To address these limitations, the researchers applied an unsupervised machine learning method -- a form of artificial intelligence (AI) that independently identifies relationships within datasets -- to analyze data from 20,352 very low birth weight infants registered in the Korean Neonatal Network.
The collaborative study included Professor Josef Neu of the University of Florida, Professor David K. Stevenson of Stanford University, Professor Oh Seong-hee of the Department of Pediatrics at Gangneung Asan Hospital, Professor Jung Sung-hoon of the Department of Pediatrics at Kyung Hee University Hospital at Gangdong, and researcher Ahn Jae-hyuk of Kyung Hee University.
To maximize objectivity, the team excluded major variables that could directly affect outcomes, including NEC diagnoses, gestational age, and birth weight.
The analysis identified five distinct patterns of intestinal injury risk that may improve diagnostic accuracy beyond the existing Bell’s staging system. These patterns included prenatal antibiotic exposure, infertility treatment, multiple pregnancies, gestational hypertension, chorioamnionitis, and severe complications associated with prematurity.
The researchers said the findings demonstrate new possibilities for diagnosing neonatal intestinal diseases.
“This study shows the potential to shift the diagnostic paradigm toward a more objective and comprehensive concept of intestinal injury compared with the existing Bell’s staging system,” Professor Choi said. “We hope that research utilizing big data and AI will bring meaningful changes to accurate diagnosis, personalized treatment, and the development of prognostic prediction systems for neonatal intestinal diseases in the future.”
Kim Jung-hyun
댓글
0