Optimized 1D CNNs for Enhanced Early Detection and Accurate Prediction of COPD and Other Pulmonary Diseases

Artificial intelligence (AI) advancements have opened new avenues in medical diagnostics, particularly for the early detection of respiratory conditions such as Chronic Obstructive Pulmonary Disease (COPD). COPD, the most common respiratory disease, is marked by breathing difficulties, mucus (sputum) production, coughing, and wheezing. Early diagnosis is crucial for effective management and treatment. A one-dimensional convolutional neural network (1D CNN) model optimized with Adam and RMSprop algorithms is proposed for COPD detection. Using the ICBHI 2017 dataset, the model's performance with each optimizer was evaluated based on training and validation accuracy. The Adam optimizer achieved a training accuracy of 94% and a validation accuracy of 90%. In comparison, the RMSprop optimizer attained a training accuracy of 92% and a validation accuracy of 88%. These findings indicate that the Adam optimizer outperformed RMSprop in terms of accuracy. It showcased the potential of the proposed 1D CNN model as a reliable diagnostic tool, contributing to the advancement of respiratory health analytics.