Smart Pelvic Bone Disease Detection from X-ray Images Using Convolutional Neural Networks

¹MS. S. LAKSHMI AKSHAYA, ²MS. M. SHAMSEENA SHABNAM, ³Mr. ARUN. M,

⁴Dr. SENTHIL VELAN. G, ⁵Mrs. S. DIVYA

^1,2Student, Department of CSE, Dr. M.G.R. Educational and Research Institute, Chennai, India. ^3,4Assistant professor, Department of CSE, Dr.M.G.R. Educational and Research Institute, Chennai, India. 5Assistant Professor, Department of CFIS, Dr.M.G.R. Educational and Research Institute, Chennai, India.

Abstract:

Pelvic bone disorders such as fractures, osteoporosis, and structural abnormalities can significantly affect a patient's mobility and overall quality of life, making early and accurate diagnosis essential. X-ray imaging is the most commonly used technique for pelvic bone examination due to its speed, accessibility, and cost-effectiveness. However, its interpretation relies heavily on experienced radiologists, and the manual diagnostic process is time-consuming, resource-intensive, and prone to human error. This research proposes an intelligent system for pelvic bone disease detection using X-ray images and a CNN-based deep learning model. By leveraging a simple yet effective Convolutional Neural Network architecture with four convolutional layers, the system can automatically extract relevant features from pelvic X-rays and accurately classify them into normal and abnormal categories. The proposed system integrates data preprocessing, augmentation techniques, and a Flask-based web interface for real- time analysis. Experimental results demonstrate that the model achieves an accuracy of 95.2% on synthetic datasets, with bone density estimation and risk assessment capabilities. The system reduces diagnostic time from approximately 30-45 minutes (manual analysis) to less than 2 seconds, representing a significant advancement in automated orthopedic diagnostics.

Keywords: Pelvic Bone, Disease Detection, Convolutional Neural Networks, Deep Learning, X- ray Analysis, Bone Density, Medical Imaging