Radiography, renowned for its diagnostic prowess and affordability, plays a key role in detecting diseases, including critical conditions. Chest radiography, focusing on a vital body area, poses interpretational challenges, necessitating experienced radiologists for accurate diagnosis. Scarcity of expertise and susceptibility to human error have driven researchers to explore Artificial Intelligence (AI), predominantly employing Convolutional Neural Networks (CNNs). This paper introduces a novel method of parallelizing multiple CNN architectures for chest X-ray classification. It conducts a thorough evaluation of existing architectures using this approach across four extensive datasets, including one non-medical dataset. Results exhibit enhanced accuracy for most labels on the primary evaluation datasets. The conclusion outlines system limitations and potential avenues for future enhancements. By leveraging AI and parallelization techniques, the proposed method showcases promising strides toward improving diagnostic accuracy and efficiency in chest radiography while also highlighting areas for further refinement and exploration at the intersection of AI and medical imaging.