An Efficient Method for Heart and Lung Disease Classification Using Siasme-BiLSTM Model

Systemic inflammation is a key factor in the development of chronic obstructive lung disease (COPD), heart failure (CHF), and cardiovascular disease (CAD). These three illnesses frequently occur together and exacerbate one another. This interdependent link, which is referred to as the “cardiopulmonary continuum,” highlights the significance of combined diagnostic methods in the categorization of lung and heart diseases. Indicators of systemic inflammatory reactions include biomarkers like interleukin-5 (IL-5) and C-reactive protein (CRP). The purpose of this research was to create a reliable pipeline for classifying lung and heart diseases from chest X-ray images. Methodology steps include improving image quality through preprocessing, precisely isolating regions of interest by segmentation, and extracting features using spatial fuzzy clustering. A siamese-BiLSTM model was trained using the retrieved characteristics. The suggested method outperformed more conventional BiLSTM and Siamese models. Classification accuracy improved substantially, reaching a remarkable 97.69 %, according to experimental evaluation. These findings demonstrate that the suggested model can correctly identify cardiopulmonary disorders and emphasize its promise for improving clinical diagnostic tools for the categorization of heart and lung diseases.