Cardiomegaly Detection using Deep Convolutional Neural Network with U-Net

Cardiomegaly is indeed a medical disease in which the heart is enlarged. Cardiomegaly is better to handle if caught early, so early detection is critical. The chest X-ray, being one of the most often used radiography examinations, has been used to detect and visualize abnormalities of human organs for decades. X-ray is also a significant medical diagnosis tool for cardiomegaly. Even for domain experts, distinguishing the many types of diseases from the X-ray is a difficult and time-consuming task. Deep learning models are also most effective when used on huge data sets, yet due to privacy concerns, large datasets are rarely available inside the medical industry. A Deep learning-based customized retrained U-Net model for detecting Cardiomegaly disease is presented in this research. In the training phase, chest X-ray images from the "ChestX-ray8" open source real dataset are used. To reduce computing time, this model performs data preprocessing, picture improvement, image compression, and classification before moving on to the training step. The work used a chest x-ray image dataset to simulate and produced a diagnostic accuracy of 94%, a sensitivity of 96.2 percent, and a specificity of 92.5 percent, which beats prior pre-trained model findings for identifying Cardiomegaly disease.