Pulse.

a daily field guide to health research that matters

◆ Console

‹ Fri · 22 May 2026
Promising but preliminary

Machine Learning-Based Prediction of Central Line-Associated Bloodstream Infection in Children with Acute Leukemia

A machine learning model using basic blood counts predicts dangerous line infections in children with leukemia, enabling earlier prevention.

A TabPFN ML model trained on 407 pediatric leukemia patients achieved 91.2% accuracy in predicting central line-associated bloodstream infections, with neutrophil count and WBC (CBC-derived features) among the top predictors. This validates a CBC-based ML approach for infection risk stratification in a high-risk pediatric oncology setting.

What the study was

Study design
Retrospective ML model development
Population
Pediatric acute leukemia patients (n=407)
Sample size
407
Category
Diagnostics
Maturity
Exploratory
Journal
J Hosp Infect

Why it surfaced

CBC features for CLABSI risk in leukemia; high accuracy; needs external validation.

A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.