Thoracic diseases present substantial health risks to a significant portion of the population, necessitating prompt diagnosis and treatment for effective management. Chest X-rays are vital diagnostic instruments that shed light on a variety of thoracic disorders. Using technology to automate analysis can improve the accuracy of thoracic illness identification and speed up abnormality discovery. By helping radiologists spot trends and abnormalities, machine learning algorithms—trained on large datasets such as the NIH Chest X-ray dataset improve diagnostic accuracy. We use an extensive data pretreatment pipeline using publicly available datasets, such as a set of 5606 photos and an associated CSV file with metadata. Leveraging robust CNN and transfer learning models, we achieve strong predictive performance through rigorous testing and model optimization, with the MobileNet model attaining an accuracy of 0.8849. Automated analysis in clinical practice improves patient outcomes in thoracic disorder management through machine learning, contingent upon ethical compliance and patient privacy safeguards.