The role of machine learning methods in assessing the risk of neonatal Sepsis: A study of biochemical markers and genetic variants

Le résumé fourni par la source

Abstract Neonatal Sepsis is a life-threatening infection that affects neonates, and its morbidity and mortality rates remain high. There is currently no effective method for timely diagnosis and prevention of neonatal Sepsis. Using machine learning techniques, we aimed to analyze the risk factors of neonatal Sepsis, including biochemical indicators and genetic variants. We collected data from 107 neonates, 56 of whom were in the sepsis cohort and 51 were not. We classified the data using support vector machine (SVM) and random forest (RF) models and evaluated model performance and feature significance. PCT (calcitoninogen level), WBC (white blood cell count), and IL-6 (interleukin-6 level) were the characteristics most strongly associated with sepsis risk. We analyzed the association between genetic variants in CRP (C-reactive protein) and IL-10 and biochemical markers using the Kruskal-Wallis H test and linear regression. CRP gene variants were significantly associated with CRP levels, whereas IL-10 gene variants were substantially distinct from Hb (hemoglobin) levels. These findings shed light on potential biomarkers and genetic correlates of neonatal Sepsis, which will inform future clinical diagnosis and treatment.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.