From clusters to clinical rules: unsupervised machine learning identifies four newborn hearing phenotypes with bedside risk stratification

Chen Liu, Jiali Zhang, Liu Yang · 2026

Read From clusters to clinical rules: unsupervised machine learning identifies four newborn hearing phenotypes with bedside risk stratification by Chen Liu, Jiali Zhang, Liu Yang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13560213.txt.

Description

Chen Liu, Jiali Zhang, and Liu Yang use unsupervised machine learning to sort high-risk newborns into hearing-risk phenotypes. Their 2026 retrospective study covers 447 babies in a neonatal unit, clustering demographic, perinatal, and diagnostic data by a medoid method suited to mixed data types. A decision tree then translated the clusters into bedside rules.

Four phenotypes appeared. Term babies with jaundice failed hearing screening least often, at 20.5 percent of their group in the reported rates. Preterm boys with respiratory illness failed most, at nearly half. Preterm girls sat between, and term babies with respiratory or infectious problems scored moderately. Simple rules emerged, such as boys under a stated birth weight failing at very high rates. Network analysis showed respiratory illness, infection, and prematurity forming a tangled core, while jaundice stood apart. The authors argue pattern-based sorting complements single-factor thinking.

Catalog record

CREATORChen Liu, Jiali Zhang, Liu Yang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39895
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13560213

Browse the full library · About the Tower of Babel