Predicting early symptoms of lung disease through the CNN classifier in .html web page

Diseases have an impact on both the physical and psychological well-being of individuals. They generally arise from four main causes: (i) infections, (ii) deficiencies, (iii) hereditary factors, and (iv) dysfunction of body organs. In today’s society, it is the responsibility of medical experts to accurately identify and categorize diseases, providing appropriate medical treatments to either cure or manage the condition. While some diseases can be completely cured with treatment, others, known as chronic illnesses, persist over time but can be controlled to prevent further deterioration. Therefore, early detection and intervention are crucial in addressing diseases effectively. Advanced techniques such as Machine Learning, Deep Learning, and Artificial Intelligence play a vital role in disease identification. This study focuses on employing deep learning and image processing techniques to classify lung disorders, aiming to support healthcare professionals in their diagnostic processes. The initial step involves gathering lung images of affected individuals for the purpose of categorization. The CNN model is then trained using a designated dataset, with the primary goal of disease identification. Subsequently, the model is applied to classify new and unseen chest X-ray images from a test dataset. To make this model accessible to end-users, a webpage application is developed, enabling them to upload unseen chest X-ray images and receive predictions regarding the corresponding disease class. As a result, the proposed CNN model is capable of categorizing chest X-ray images into classes such as COVID-19, Pneumonia, Tuberculosis, and Normal by achieving an impressive accuracy of 97% in the classification of lung diseases.