A Diagnostic Nomogram for Early Prediction of Post-Infectious Bronchiolitis Obliterans in Severe Pneumonia

Risk factors and prediction models for bronchiolitis obliterans after severe adenoviral pneumonia

Severe adenoviral pneumonia (SAP) can cause post-infectious bronchiolitis obliterans (PIBO) in children. We aimed to investigate the relevant risk factors for PIBO and develop a predictive nomogram for PIBO in children with SAP. This prospective study analysed the clinical data of hospitalised children with SAP and categorised them into the PIBO and non-PIBO groups. Least absolute shrinkage and selection operator (LASSO) regressions were applied to variables that exhibited significant intergroup differences. Logistic regression was adopted to analyse the risk factors for PIBO. Additionally, a nomogram was constructed, and its effectiveness was assessed using calibration curves, C-index, and decision curve analysis. A total of 148 hospitalised children with SAP were collected in this study. Among them, 112 achieved favourable recovery, whereas 36 developed PIBO. Multivariable regression after variable selection via LASSO revealed that aged < 1 year (OR, 2.38, 95% CI, 0.82-6.77), admission to PICU (OR, 24.40, 95% CI, 7.16-105.00), long duration of fever (OR, 1.16, 95% CI, 1.04-1.31), and bilateral lung infection (OR, 8.78, 95% CI, 1.32-195.00) were major risk factors for PIBO. The nomogram model included the four risk factors: The C-index of the model was 0.85 (95% CI, 0.71-0.99), and the area under the curve was 0.85 (95% CI, 0.78-0.92). The model showed good calibration with the Hosmer-Lemeshow test (χ2 = 8.52, P = 0.38) and was useful in clinical settings with decision curve analysis.

CONCLUSION

Age < 1 year, PICU admission, long fever duration, and bilateral lung infection are independent risk factors for PIBO in children with SAP. The nomogram model may aid clinicians in the early diagnosis and intervention of PIBO.

WHAT IS KNOWN

• Adenoviruses are the most common pathogens associated with PIBO. • Wheezing, tachypnoea, hypoxemia, and mechanical ventilation are the risk factors for PIBO.

WHAT IS NEW

• Age < 1 year, admission to PICU, long duration of fever days, and bilateral lung infection are independent risk factors for PIBO in children with SAP. • A prediction model presented as a nomogram may help clinicians in the early diagnosis and intervention of PIBO.

Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study

Objective

The study aimed to use supervised machine learning models to predict the length and risk of prolonged hospitalization in PLWHs to help physicians timely clinical intervention and avoid waste of health resources.

Methods

Regression models were established based on RF, KNN, SVM, and XGB to predict the length of hospital stay using RMSE, MAE, MAPE, and R2, while classification models were established based on RF, KNN, SVM, NN, and XGB to predict risk of prolonged hospital stay using accuracy, PPV, NPV, specificity, sensitivity, and kappa, and visualization evaluation based on AUROC, AUPRC, calibration curves and decision curves of all models were used for internally validation.

Results

In regression models, XGB model performed best in the internal validation (RMSE = 16.81, MAE = 10.39, MAPE = 0.98, R2 = 0.47) to predict the length of hospital stay, while in classification models, NN model presented good fitting and stable features and performed best in testing sets, with excellent accuracy (0.7623), PPV (0.7853), NPV (0.7092), sensitivity (0.8754), specificity (0.5882), and kappa (0.4672), and further visualization evaluation indicated that the largest AUROC (0.9779), AUPRC (0.773) and well-performed calibration curve and decision curve in the internal validation.

Conclusion

This study showed that XGB model was effective in predicting the length of hospital stay, while NN model was effective in predicting the risk of prolonged hospitalization in PLWH. Based on predictive models, an intelligent medical prediction system may be developed to effectively predict the length of stay and risk of HIV patients according to their medical records, which helped reduce the waste of healthcare resources.