D2.132 - Prediction of Clinical Outcomes of Acute Bronchitis with Bronchospasm Based on Trigger Factors Using a Neural Network Model

A model based on a multilayer perceptron (MLP) was developed to predict clinical outcomes in children who had experienced acute bronchitis with obstructive syndrome.

A model based on a multilayer perceptron (MLP) was developed to predict clinical outcomes in children who had experienced acute bronchitis with obstructive syndrome.

A model based on a multilayer perceptron (MLP) was developed to predict clinical outcomes in children who had experienced acute bronchitis with obstructive syndrome.

The area under the ROC curve (AUC) was 0.611, indicating moderate discriminative ability of the model. The variables exerting the greatest influence on disease outcome were: household allergy — normalized importance 100.0%; ARVI — 61.2%; weather changes — 54.0%; dietary disturbances — 41.8%; and pneumonia — 11.5%.

Thus, the model highlights the significance of trigger factors—particularly household allergy and exposure to respiratory infections—in predicting a complicated course of acute bronchitis with bronchospasm. These findings may serve as a basis for the development of preventive recommendations.