A Novel Deep Learning Approach with Attention Mechanism for Early Osteoporosis Detection from Knee X-Ray Imaging

International Journal of Electronics and Communication Engineering

© 2025 by SSRG - IJECE Journal

Volume 12 Issue 6

Year of Publication : 2025

Authors : Athira O M, R Gunasudari

10.14445/23488549/IJECE-V12I6P117

How to Cite?

Athira O M, R Gunasudari, "A Novel Deep Learning Approach with Attention Mechanism for Early Osteoporosis Detection from Knee X-Ray Imaging," SSRG International Journal of Electronics and Communication Engineering, vol. 12,  no. 6, pp. 215-226, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I6P117

Abstract:

A bone disorder with decreased Bone Mineral Density (BMD), rendering bones weaker and susceptible to fractures, even from minor falls or daily activities, is osteoporosis. An accurate diagnosis is crucial for effective therapy to reduce the fracture risk, yet existing diagnostic procedures are time-consuming. The traditional models relied on manual radiological evaluation and hand-crafted Machine Learning (ML) features to diagnose knee osteoporosis. However, these approaches had limitations in accuracy and efficacy due to radiologists’ subjective interpretations and manual feature extraction. Early detection with X-ray imaging allows timely intervention and facilitates proper treatment. This study proposes a novel osteoporosis detection model using a Deep Learning (DL)-based approach enhanced by an attention mechanism to improve classification performance using knee X-ray dataset images. The DenseNet-121 model is the backbone, improving the vanishing gradient problem and ensuring efficient data flow across layers. Channel-wise and spatial attention techniques are utilized with a Convolutional Block Attention Module (CBAM) to refine feature representation. The model demonstrated superior performance in classifying osteoporotic and healthy knees from X-ray images, attaining a remarkable accuracy of 97.43%. This study enhances osteoporosis detection by employing knee X-ray images with excellent prediction and efficient classification, thereby reducing the socioeconomic burden of osteoporosis-related fracture risks and improving patient outcomes.

Keywords:

Osteoporosis, Deep Learning, DenseNet 121, Convolutional block attention module, Attention mechanism.

References:

[1] M.S. LeBoff et al., “The Clinician’s Guide to Prevention and Treatment of Osteoporosis,” Osteoporosis International, vol. 33, pp. 2049-2102, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Md Shakhawat Hossen et al., “Revolutionizing Osteoporosis and Bone Fracture Diagnostics: The Emergence of Microwave Antenna Technology,” IEEE Access, vol. 12, pp. 160418-160448, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Boyi Li et al., “Effect of Spectral Estimation on Ultrasonic Backscatter Parameters in Measurements of Cancellous Bones,” IEEE Access, vol. 7, pp. 83034-83045, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Wilson Ong et al., “Artificial Intelligence Applications for Osteoporosis Classification Using Computed Tomography,” Bioengineering, vol. 10, no, 12, pp. 1-26, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Chaewoo Kim et al., “A Deep Learning Harmonization of Multi-Vendor MRI for Robust Intervertebral Disc Segmentation,” IEEE Access, vol. 12, pp. 19482-19499, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[6] M.J. Aashik Rasool et al., “KONet: Toward a Weighted Ensemble Learning Model for Knee Osteoporosis Classification,” IEEE Access, vol. 12, pp. 5731-5742, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Amany M. Sarhan et al., “Knee Osteoporosis Diagnosis Based on Deep Learning,” International Journal of Computational Intelligence Systems, vol. 17, pp. 1-33, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Ryoungwoo Jang et al., “Prediction of Osteoporosis from Simple Hip Radiography using Deep Learning Algorithm,” Scientific Reports, vol. 11, pp. 1-9, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Róża Dzierżak, and Zbigniew Omiotek, “Application of Deep Convolutional Neural Networks in the Diagnosis of Osteoporosis,” Sensors, vol. 22, no. 21, pp. 1-18, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Yijie Fang et al., “Opportunistic Osteoporosis Screening in Multi-Detector CT Images Using Deep Convolutional Neural Networks,” European Radiology, vol. 31, pp. 1831-1842, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Takashi Nakamoto, Akira Taguchi, and Naoya Kakimoto, “Osteoporosis Screening Support System from Panoramic Radiographs Using Deep Learning by Convolutional Neural Network,” Dentomaxillofacial Radiology, vol. 51, no. 6, pp. 1-8, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Shintaro Sukegawa et al., “Identification of Osteoporosis Using Ensemble Deep Learning Model with Panoramic Radiographs and Clinical Covariates,” Scientific Reports, vol. 12, pp. 1-10, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Sangseok Oh et al., “Evaluation of Deep Learning-Based Quantitative Computed Tomography for Opportunistic Osteoporosis Screening,” Scientific Reports, vol. 14, pp. 1-9, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Soaad M. Naguib et al., “A New Superfluity Deep Learning Model for Detecting Knee Osteoporosis and Osteopenia in X-Ray Images,” Scientific Reports, vol. 14, pp. 1-18, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Liyu Liu et al., “A Hierarchical Opportunistic Screening Model for Osteoporosis using Machine Learning Applied to Clinical Data and CT Images,” BMC Bioinformatics, vol. 23, pp. 1-15, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Wen-Yu Ou Yang et al., “Development of Machine Learning Models for Prediction of Osteoporosis from Clinical Health Examination Data,” International Journal of Environmental Research and Public Health, vol. 18, no. 14, pp. 1-12, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Insha Majeed Wani, and Sakshi Arora, “Osteoporosis Diagnosis in Knee X-Rays by Transfer Learning Based on Convolution Neural Network,” Multimedia Tools and Applications, vol. 82, pp. 14193-14217, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Varada Vivek Khanna et al., “A Decision Support System for Osteoporosis Risk Prediction using Machine Learning and Explainable Artificial Intelligence,” Heliyon, vol. 9, no. 12, pp. 1-19, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Ronnie Sebro, and Cynthia De la Garza-Ramos, “Machine Learning for Opportunistic Screening for Osteoporosis from CT Scans of the Wrist and Forearm,” Diagnostics, vol. 12, no. 11, pp. 1-17, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Khasnur Hidjah et al., “Periapical Radiograph Texture Features for Osteoporosis Detection Using Deep Convolutional Neural Network,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 1, pp. 223-232, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[21] German Cuaya-Simbro et al., “Comparing Machine Learning Methods to Improve Fall Risk Detection in Elderly with Osteoporosis from Balance Data,” Journal of Healthcare Engineering, vol. 2021, no. 1, pp. 1-11, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Osteoporosis Knee X-ray Dataset, Kaggle. [Online]. Available: https://www.kaggle.com/datasets/stevepython/osteoporosis-knee-xray-dataset/data
[23] Saleh A. Albelwi, “Deep Architecture Based on DenseNet-121 Model for Weather Image Recognition,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 10, pp. 1-7, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Omar Elharrouss et al., “Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches,” Arxiv, pp. 1-23, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Sanghyun Woo et al., “CBAM: Convolutional Block Attention Module,” Proceedings of the 15th European Conference on Computer Vision, Munich, Germany, pp. 3-19, 2018.
[CrossRef] [Google Scholar] [Publisher Link]