Construction and Validation of a Nomogram Model to Predict the Severity of Mycoplasma pneumoniae Pneumonia in Children

Abstract

Background This study aimed to develop a nomogram model for early prediction of the severe Mycoplasma pneumoniae pneumonia (MPP) in children. Methods A retrospective analysis was conducted on children with MPP, classifying them into severe and general MPP groups. The risk factors for severe MPP were identified using Logistic Stepwise Regression Analysis, followed by Multivariate Regression Analysis to construct the nomogram model. The model’s discrimination was evaluated using a receiver operating characteristic curve, its calibration with a calibration curve, and the results were visualized using the Hosmer–Lemeshow goodness-of-fit test. Results Univariate analysis revealed that age, duration of fever, length of hospital-stay, decreased sounds of breathing, respiratory rate, hypokalemia, and incidence of co-infection were significantly different between severe and general MPP. Significant differences ( p < 0.05) were also observed in C-reactive protein, procalcitonin, peripheral blood lymphocyte count, neutrophil-to-lymphocyte ratio, ferritin, lactate dehydrogenase, alanine aminotransferase, interleukin-6, immunoglobulin A, and CD4 + T cells between the two groups. Logistic Stepwise Regression Analysis showed that age, decreased sounds of breathing, respiratory rate, duration of fever (OR = 1.131; 95% CI: 1.060–1.207), length of hospital-stay (OR = 1.415; 95% CI: 1.287–1.555), incidence of co-infection (OR = 1.480; 95% CI: 1.001–2.189), ferritin level (OR = 1.003; 95% CI: 1.001–1.006), and LDH level (OR = 1.003; 95% CI: 1.001–1.005) were identified as risk factors for the development of severe MPP ( p < 0.05 in all). The above factors were applied in constructing a nomogram model that was subsequently tested with 0.862 of the area under the ROC curve. Conclusion Age, decreased sound of breathing, respiratory rate, duration of fever, length of hospital-stay, co-infection with other pathogen(s), ferritin level, and LDH level were the significant contributors for the establishment of a nomogram model to predict the severity of MPP in children.

Development of machine learning-based differential diagnosis model and risk prediction model of organ damage for severe Mycoplasma pneumoniae pneumonia in children

“…Machine learning algorithms excel at exploring intricate relationships within multidimensional data, extracting hidden and valid knowledge from vast datasets, and making more accurate predictions and diagnoses of diseases 16 . Historically, numerous studies have developed MPP prediction models focusing on diagnosis, severity, risk factors, treatment, and prognosis [17][18][19][20] . These studies primarily relied on constructing nomogram models for the early identi cation and intervention of MPP.…”

“…However, complex ML methods can manage a broader array of variables, often yielding more accurate and precise results than traditional modeling methods 21 . Additionally, one of the main challenges in applying MPP prediction models to clinical practice is the lack of external validation [19][20][21][22] .…”