Development and validation of a nomogram for predicting hemoptysis risk in patients with bronchiectasis: A study...

Area of Science:

• Pulmonary Medicine
• Medical Informatics
• Biostatistics

Background:

• Hemoptysis is a severe complication of bronchiectasis with limited individualized risk prediction tools.
• Accurate prediction of hemoptysis risk is crucial for timely intervention in bronchiectasis patients.

Purpose of the Study:

• To develop and validate a nomogram for predicting hemoptysis risk in bronchiectasis patients using machine learning.
• To identify key clinical predictors for hemoptysis in this patient population.

Main Methods:

• Retrospective study of 131 bronchiectasis patients, divided into training (70%) and validation (30%) sets.
• Random Forest algorithm used for predictor screening, followed by multivariate logistic regression for nomogram construction.
• Model performance evaluated using ROC curves, calibration curves, decision curve analysis, and clinical impact curves.

Main Results:

• The nomogram identified body mass index, C-reactive protein, creatinine, age, and urea as independent predictors of hemoptysis.
• The nomogram demonstrated excellent predictive performance with an AUC of 0.944 in the validation set.
• The model showed good calibration and significant clinical utility across various threshold probabilities.

Conclusions:

• A validated nomogram integrating five readily available clinical parameters accurately predicts hemoptysis risk in bronchiectasis patients.
• This tool offers good calibration and clinical utility, assisting clinicians in early risk assessment and management.
• The developed nomogram serves as a valuable aid for identifying and managing high-risk bronchiectasis patients prone to hemoptysis.