The COVID-19 pandemic has underscored the critical need for rapid and accurate diagnostic tools to manage the spread of the virus and optimize patient care. While RT-PCR testing remains the gold standard, chest radiographs (X-rays) have emerged as a valuable supplementary diagnostic resource due to their accessibility and ability to reveal characteristic pulmonary patterns associated with COVID-19. This study explores the use of Convolutional Neural Networks (CNNs) for the automated detection of COVID-19 from chest radiographs. We collected a diverse dataset of chest X-rays from Kaggle, applied various image enhancement techniques, and trained a CNN model to differentiate between COVID-19 pneumonia, other types of pneumonia, and healthy lungs. The model’s performance was rigorously evaluated using metrics such as accuracy, sensitivity and prevalence. Our results demonstrate that the application of image enhancement techniques significantly improves the diagnostic performance of the CNN model, highlighting the potential of AI-driven tools in augmenting traditional diagnostic methods. This study provides a foundation for further development and integration of AI-based diagnostic systems in clinical settings, particularly in resource-limited environments.