A Hybrid Efficientnet Densenet Deep Learning Framework for Automated Detection of Osteopenia and Osteoporosis

DOI

10.2991/978-94-6239-616-6_65How to use a DOI?

Keywords

Osteopenia; Osteoporosis; Deep Learning; EfficientNet-B0; DenseNet-121; Hybrid Architecture; Computer-Aided Diagnosis; Medical Imaging

Abstract

Osteopenia and osteoporosis are progressive conditions that weaken bones, making them more vulnerable to fractures, especially in older adults. Early detection is vital for preventing serious complications and ensuring prompt medical care. However, traditional diagnostic methods often depend on manual interpretation of bone scans, which can be slow and inconsistent. To address these issues, this study presents an automated bone health assessment system based on a hybrid deep learning framework that combines EfficientNet-B0 and DenseNet-121 architectures. EfficientNet-B0 captures global structural patterns effectively through compound scaling, while DenseNet-121 targets detailed local features using dense connectivity. By merging these strengths, the system improves diagnostic accuracy, speed, and reliability [1]. The combined features are classified using a Softmax layer to determine bone conditions as normal, osteopenia, or osteoporosis. Experimental results indicate that the proposed model achieves higher accuracy, faster convergence, and strong generalization across various datasets. This hybrid AI-based approach provides a reliable and scalable solution to assist clinicians in early diagnosis, reduce manual workload, and enhance personalized bone health management.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - R. Suresh
AU  - A. Supha Lakshmi
AU  - K. Keerthika
AU  - R. Raja Ragavi
AU  - B. Divya
PY  - 2026
DA  - 2026/03/31
TI  - A Hybrid Efficientnet Densenet Deep Learning Framework for Automated Detection of Osteopenia and Osteoporosis
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 872
EP  - 883
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_65
DO  - 10.2991/978-94-6239-616-6_65
ID  - Suresh2026
ER  -