Year: 2025
Type: article
Abstract: —Pneumonia is a serious lung infection and one of the leading causes of illness and death globally, particularly affecting children, the elderly, and individuals with weakened immune systems. The diagnosis of pneumonia typically relies on chest X-rays that are interpreted by radiologists, a process that can be time-consuming and susceptible to errors. To address these challenges, this study introduces an AI-driven framework for pneumonia detection that utilizes a Region-based Convolutional Neural Network (RCNN) based on the VGG16 architecture. The model was trained on a dataset comprising 5,856 pediatric chest X-ray images, which included 4,273 cases of pneumonia and 1,583 normal cases. To enhance the model's performance and address class imbalance, preprocessing techniques such as pixel normalization and data augmentation (including horizontal flipping, zooming, and shearing) were employed. The model demonstrated high accuracy in classifying X-rays as either "normal" or "pneumonia," highlighting the potential of AI-assisted diagnostics to reduce delays and aid radiologists in their work. Future efforts will focus on exploring advanced architectures, such as Transformer-based models, and expanding the dataset to improve the model's generalizability.
Authors: P.K Sangameswar, Rohit Manoj, Sayanth V, Sibin Babu
Institution: Ahalia School of Engineering and Technology
Language: English
Topic: COVID-19 diagnosis using AI
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W4409197451