Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{196382,
        author = {Dr.B.Arun Kumar and R.Bindu and Y.Vennela and R.Ch.Lakshmidevi and V.Amrutha Varshini},
        title = {PNEUMONIA DETECTION DEEP LEARNING- BASED MEDICAL IMAGE CLASSIFICATION SYSTEM},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {3424-3428},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196382},
        abstract = {This project developed a deep learning-based medical image classification system for the automatic detection of pneumonia from chest X-rays. Pneumonia is an acute and serious respiratory sickness that needs an accurate diagnosis as soon as possible to avoid severe problems that could lead to death; however, manual interpretations of the radiograph by a clinician can take a significant amount of time and may include errors from the clinician's inability to detect subtle differences in the degree of pacification of the lungs or overlapping anatomy. This research proposes an intelligent and automated diagnostic support system using advanced deep learning methods to address these challenges. The proposed system uses Convolutional Neural Networks (CNN) and transfer learning to achieve a high diagnostic accuracy, despite having a limited number of labeled medical images. Pre-trained architectures, such as VGG16, ResNet (Residual Networks), and MobileNet, were used to retrieve deep hierarchical features from chest X-ray images. A multi-model comparison framework was implemented to compare the different models based on their accuracy, precision, recall and computational efficiency. To help alleviate the vanishing gradient issue, the proposed system uses residual connections; while at the same time lightweight architectures help to ensure that the system can be deployed in healthcare environments where resources are limited. The system has been trained and validated on publicly available datasets, including the Kaggle Chest X-Ray (Pneumonia) Dataset, the RSNA Pneumonia Detection Challenge Dataset and the NIH Chest-X-ray Dataset.},
        keywords = {Convolutional Neural Networks (CNN), VGG16, ResNet (Residual Networks), Mobile Net.},
        month = {April},
        }