Predicting Lung Disease Progression Using Deep Learning on Pulmonary Function Test Data

Indeed, asthma, chronic obstructive pulmonary disease (COPD), and pulmonary fibrosis are major health issues among people worldwide. Early diagnosis and accurate predictions of illness progression are essential for their management and treatment. Pulmonary Function Tests (PFTs) are valuable physiological tests reflecting lung function and disease severity changes over time. In this article, we show a deep learning approach to predict developments in lung diseases by PFT data. Our model is based on temporal patterns and characteristics from PFT readings that enable early diagnosis and prognosis for lung diseases. Unfortunately, due to a specific PFT dataset, collecting representative data for model training is always a challenge. Meanwhile, we address various sources of access to PFT datasets from public repositories through partnerships with healthcare institutions or data-sharing platforms. Once sufficient data is collected, we plan to perform preprocessing, develop deep learning models, and evaluate their effectiveness regarding lung disease development prediction.