Résumé
Objective:To construct prediction models of necrotizing enterocolitis (NEC) using machine learning (ML) methods.Methods:From January 2015 to October 2021, neonates with suspected NEC symptoms receiving abdominal ultrasound examinations in our hospital were retrospectively analyzed. The neonates were assigned into NEC group (modified Bell's staging≥Ⅱ) and non-NEC group for diagnostic prediction analysis (dataset 1). The NEC group was subgrouped into surgical NEC group (staging≥Ⅲ) and conservative NEC group for severity analysis (dataset 2). Feature selection algorithms including extremely randomized trees, elastic net and recursive feature elimination were used to screen all variables. The diagnostic and severity prediction models for NEC were established using logistic regression, support vector machine (SVM), random forest, light gradient boosting machine and other ML methods. The performances of different models were evaluated using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, negative predictive value and positive predictive value.Results:A total of 536 neonates were enrolled, including 234 in the NEC group and 302 in the non-NEC group (dataset 1).70 were in the surgical NEC group and 164 in the conservative NEC group (dataset 2). The variables selected by extremely randomized trees showed the best predictive performance in two datasets. For diagnostic prediction models, the SVM model had the best predictive performance, with AUC of 0.932 (95% CI 0.891-0.973) and accuracy of 0.844 (95% CI 0.793-0.895). A total of 11 predictive variables were determined, including portal venous gas, intestinal dilation, neutrophil percentage and absolute monocyte count at the onset of illness. For NEC severity prediction models, the SVM model showed the best predictive performance, with AUC of 0.835 (95% CI 0.737-0.933) and accuracy of 0.787 (95% CI 0.703-0.871). A total of 25 predictive variables were identified, including age of onset, C-reactive protein and absolute neutrophil count at clincial onset. Conclusions:NEC prediction model established using feature selection algorithm and SVM classification model in ML is helpful for the diagnosis of NEC and grading of disease severity.
Résumé
Objective:To establish the reference range of red blood cell parameters within 24 hours after birth of premature infants with different gestational ages and genders.Methods:According to the inclusion criteria and exclusion criteria, a retrospective analysis was performed in premature infants who were admitted to the neonatal intensive care unit from January 1, 2018 to December 31, 2019. These newborns were delivered in the obstetrics department of our hospital or came from other parts of Jilin Province. All of their radial artery blood were collected within 24 hours after birth. According to the blood examination results, we analyzed reference range of red blood cell parameters of these premature infants.Results:With the increase of gestational age, the number of RBC, HGB, HCT, MCHC gradually increases and the number of MCV, MCH gradually decreases. There are differences in some red blood cell parameters of premature infants with 34 week≤gestational age<37 week between different genders. Compared with boys, the number of RBC, HGB, HCT and MCV in girls were higher. The number of RBC in premature infants with 23 week≤gestational age<28 week and 28 week ≤ gestion age<34 week are 2.58×10 12-5.45×10 12/L and 2.97×10 12-5.86×10 12/L respectively. In the group of premature infants with 34 week ≤gestion age<37 week, the number of RBC in boys is 3.38×10 12-5.83×10 12/L, while the number of RBC in girls is 3.18-5.89×10 12/L. There're no difference in RDW among preterm infants with different gestational ages and genders, which is 14.8%-20.6%. Conclusions:The study established the reference range of red blood cell parameters of 23 w≤gestational age<37 w premature infants within 24 hours after the birth and explored the differences in red blood cell parameters of premature infants with different gestational ages and genders.