Impact of a Novel Convolution Neural Network on Chest X-Ray Images for Pneumonia Detection

Pneumonia is still considered as a major worldwide healthcare hazard which needs immediate treatment and timely and precise identification. The research paper shows the method for the precise identification of the disease with the use of Convolutional Neural Network (CNN) on the Chest X-ray(CXR) images with the help of Batch normalisation, padding and data augmentation. The research uses a comprehensive dataset of 3450 CXR images. The experiments were conducted on over 200 epochs to ensure accuracy. Batch normalization and padding contributed to the stability and better interpretability of the CNN. Data Augmentation including Colour jitter made the training set more diverse for better generalization. The classification parameters like precision, recall ,F1 score and accuracy were used for the estimation of the model's efficacy. The model showed a 96.02% accuracy reflecting its reliability in classification. This study gives medical professionals a dependable and effective tool that improves the state-of-the-art in pneumonia diagnosis.