Pneumonia is a significant respiratory disease that leads to acute inflammation in lung tissues, primarily affecting the terminal fine bronchioles and alveoli, resulting in solidification. As per the World Health Organization (WHO), approximately 4 million people lose their lives to pneumonia annually, and it is particularly prevalent among children, the elderly, and those with compromised immune systems. Due to the high rates of morbidity and mortality associated with pneumonia, an early and accurate diagnosis is crucial for effective treatment. Chest X-rays are a commonly used tool for diagnosing pneumonia as they help visualize abnormalities in the lungs. However, traditional manual analysis is time-consuming and susceptible to errors, creating a need for more efficient and precise diagnostic methods. This research focuses on developing a deep learning-based computer-aided diagnosis system (DAOC) for pneumonia classification using 5000 chest X-ray images. It demonstrates advancements due to its superior ability to extract features. Several deep learning models, including the Convolutional Neural Network (CNN), ResNet50, and Vision Transformer (ViT), are evaluated and compared to enhance diagnostic accuracy and efficiency. Deep learning techniques are at the forefront of medical image analysis due to their strengths in managing large datasets and recognizing complex patterns. Results indicate that the ResNet50 model achieves the highest classification accuracy and sensitivity, significantly surpassing other models. By employing ResNet50, pneumonia cases can be identified and classified more effectively, thus addressing the rising demand for healthcare professionals and ensuring timely identification and treatment of patients. This research illustrates the substantial potential and application value of deep learning in the classification of medical images, particularly in addressing the global prevalence of chest diseases.
Deep Learning-Based Diagnostic System Research for Pneumonia Image Classification
2024-10-23
803158 byte
Conference paper
Electronic Resource
English
Deep Learning Approach to detect Pneumonia
IEEE | 2020
|Deep Migration Learning-Based Aerospace Food Image Classification
Springer Verlag | 2023
|Secure Image Classification Using Deep Learning
Springer Verlag | 2023
|