International Journal of Science, Engineering and TechnologyAuthors: Associate Professor ,Dr.D.Uma1, Yadavalli Sai Naga Sri Ramya2,, Latchipatula Lohitha3,, Bongu Sai Tulasi Pravallika4, Sabhavathula Pujyavenkata Krishnachaitanya
Abstract: Spinal disorders such as scoliosis, osteoporosis, spondylolisthesis, osteochondrosis, and vertebral compression fractures are major contributors to chronic back pain and physical disability worldwide. Early and precise diagnosis using spine X-ray imaging plays a crucial role in effective treatment planning and long-term patient care. However, manual analysis of radiographic images is time-consuming and highly dependent on the expertise of radiologists, which may lead to variability in diagnosis.To address this challenge, this study presents a deep learning-based multi-class classification framework for the automatic detection of various spinal conditions from X-ray images. The proposed system utilizes Convolutional Neural Networks (CNNs) for automated feature extraction and classification. Multiple pre-trained architectures, including VGGNet and ResNet, are evaluated and compared with a customized CNN model to identify the most effective approach. The dataset undergoes preprocessing steps such as resizing, normalization, and augmentation to improve model generalization and robustness.Experimental results demonstrate that the proposed CNN model achieves superior performance, with high accuracy, precision, recall, and F1-score across multiple spine condition categories. The system provides reliable and consistent predictions, highlighting its potential as a computer-aided diagnostic tool. By assisting medical professionals with faster and more standardized analysis, the proposed framework can contribute to improved clinical decision-making and better patient outcomes
DOI: http://doi.org/10.61463/ijset.vol.14.issue2.159
