A Deep Learning-Driven Lung Disease Classification Using CCRBM Algorithm for Future Prognosis and Forecasting

The necessity for precise and timely prediction models is highlighted by the substantial burden that the global spread of lung illnesses places on healthcare systems. Conventional classification techniques frequently have trouble handling heterogeneous and incomplete data. In order to categorize and forecast the course of lung disease, this study presents a deep learning-driven method that combines a Collaborative Clinical Restricted Boltzmann Machine (CCRBM) with the SEIRF framework—Susceptible, Exposed, Isolation, Recovery, Forecasting. The innovation is in matching epidemiological stages to RBM-based classification to improve clinical relevance and interpretability. The model analyzes clinical, diagnostic, and demographic information from patient records. The CCRBM method outperforms more conventional classifiers like SVM, logistic regression, and decision trees through thorough preprocessing, feature extraction, and model training. The model's suitability for future disease forecasting and clinical decision-making is validated by the findings, which demonstrate good accuracy, precision, recall, and F1 score.