Developing a Multi-Omics Inflammatory Profile to Predict Recurrence Risk in Allergic Bronchopulmonary Aspergillosis ABPA Patients

What this Trial Is About

This research aims to develop and validate a machine learning model to predict the one-year risk of relapse in patients with stable allergic bronchopulmonary aspergillosis ABPA. By identifying inflammatory phenotypes through advanced data analysis, the study seeks to improve how patients are managed based on their risk levels. The project combines multiple types of data including imaging, fungal profiles, inflammatory markers, lung function tests, and clinical records to better understand ABPA recurrence. The study collects detailed medical information from patients with stable ABPA who visited multiple hospitals between January 2021 and January 2025. These patients are grouped into relapse and non-relapse categories based on their one-year follow-up. Machine learning techniques are used to select key features from the collected data to build a prediction model. Patients identified as low risk will receive routine monitoring, while those at high risk will receive more intensive care and management. Participants will undergo evaluations including pulmonary function tests, inflammatory marker assessments, fungal omics, and radiological imaging. Researchers will monitor for disease recurrence during the remission period over one year. The study measures how well the model can predict relapse to help guide personalized treatment decisions. Data collection and analysis aim to enhance clinical outcomes by tailoring follow-up and interventions based on individual risk levels.

Multi-Omics Inflammatory Phenotype for ABPA Recurrence Risk Prediction