Nandkishor Narkhede; D. Girish Kumar; Dhanya Job; K. Varada Rajkumar; Tirumalasetti Lakshmi Narayana; B. Loveswara Rao

XrayBoneNet: Multi-Region Fracture Localization and OA Grading via Attention-Guided Transformers

PDF (1389KB), PP.165-183

Views: 0 Downloads: 0

Author(s)

Nandkishor Narkhede ^1,* D. Girish Kumar ² Dhanya Job ³ K. Varada Rajkumar ⁴ Tirumalasetti Lakshmi Narayana ⁵ B. Loveswara Rao ⁶

1. Department of Electronics and Computer Science, Shah and Anchor Kutchhi Engineering College, Chembur, Mumbai – 400088, India

2. Department of Electronics and Communication Engineering, Shri Vishnu Engineering College for Women, Bhimavaram – 534202, India

3. Department of Computer Applications, Baselios Poulose Second Catholicose College, Piravom, Ernakulam, Kerala, India

4. Department of Computer Science and Engineering (AIML), M L R Institute of Technology, Hyderabad, Telangana, India

5. Department of Electrical and Electronics Engineering, Aditya University, Surampalem, Andhra Pradesh – 533437, India

6. Department of Electrical and Electronics Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh – 522302, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijigsp.2026.04.09

Received: 23 Jan. 2026 / Revised: 20 Feb. 2026 / Accepted: 24 Mar. 2026 / Published: 8 Aug. 2026

Index Terms

Bone Fracture, Multi-Region X-ray Dataset, Bighorn Sheep Optimization, Logarithmic Mean Optimization, Attention mechanisms, Intersection-over-Union

Abstract

It is still hard to accurately find bone fractures and figure out how bad osteoarthritis (OA) is from X-ray pictures because of the complicated anatomical differences and the lack of contextual modelling in standard deep learning methods. This research presents XrayBoneNet, a hybrid deep learning system that combines Convolutional Neural Networks (CNNs), Transformer-based global feature modelling, and attention mechanisms for concurrent fracture identification, OA grading, and localization. The model has two heads: one for binary fracture classification and one for multi-class OA staging. It also has a bounding box regression head for accurate localization. To improve training efficiency and performance, a hybrid optimization technique that uses Bighorn Sheep Optimization (BSO) for global exploration and Logarithmic Mean Optimization (LMO) for fine-tuning is developed. The Bone Fracture Multi-Region X-ray dataset shows that XrayBoneNet works better than state-of-the-art models like ResNet50, DenseNet121, and Vision Transformer. It has 96.8% accuracy in fracture detection, 94.5% accuracy in OA classification, and an Intersection-over-Union (IoU) of 0.87 for localization. The suggested system offers an effective and comprehensible alternative for automated radiological diagnosis.

Cite This Paper

Nandkishor Narkhede, D. Girish Kumar, Dhanya Job, K. Varada Rajkumar, Tirumalasetti Lakshmi Narayana, B. Loveswara Rao, "XrayBoneNet: Multi-Region Fracture Localization and OA Grading via Attention-Guided Transformers", International Journal of Image, Graphics and Signal Processing(IJIGSP), Vol.18, No.4, pp. 165-183, 2026. DOI:10.5815/ijigsp.2026.04.09

Reference

[1]M. Kutbi. “Artificial intelligence-based applications for bone fracture detection using medical images: a systematic review.” Diagnostics, vol.14, no.17, 1879, 2024.
[2]C. Sethi, S. Singh, P. S. Chauhan, & B. Raj. “Bone Fracture Detection Using Machine Learning.” In Distributed Intelligent Circuits and Systems, pp. 77-109, 2024.
[3]A. J. Alkhatib, M. Alharoun, & A. AlZoubi. “A deep learning framework for timely bone fracture detection and prevention.” Information Sciences with Applications, vol.1, pp.52-61, 2024.
[4]A. Bhatnagar, A. L. Kekatpure, V. R. Velagala, & A. Kekatpure. A review on the use of artificial intelligence in fracture detection. Cureus, vol.16, no.4, 2024.
[5]G. Deo, J. Totlani, & C. V. Mahamuni. “A survey on bone fracture detection methods using image processing and artificial intelligence (AI) approaches.” In AIP Conference Proceedings, Vol. 2802, No. 1, 2024, January. AIP Publishing.
[6]J. Zou, & M. R. Arshad. “Detection of whole-body bone fractures based on improved yolov7.” Biomedical Signal Processing and Control, vol. 91, 105995, 2024.
[7]S. Parvin, & A. Rahman. “A real-time human bone fracture detection and classification from multi-modal images using deep learning technique.” Applied Intelligence, vol.54, no.19, pp.9269-9285, 2024.
[8]S. Thota, P. Kandukuru, M. Sundaram, A. Ali, S. M. Basha, & N. H. Bindu. “Deep Learning based Bone Fracture Detection.” In 2024 International Conference on Smart Systems for applications in Electrical Sciences (ICSSES) pp. 1-7, 2024, May. IEEE.
[9]A. Tieu, E. Kroen, Y. Kadish, Z. Liu, N. Patel, A. Zhou, & T. Deyer. “The role of artificial intelligence in the identification and evaluation of bone fractures.” Bioengineering, vol.11, no.4, 338, 2024.
[10]A. Verma, V. Kumar, & R. K. Yadav. “Humerus bone fracture detection utilizing YOLOv4 algorithm: a deep learning approach.” In 2024 2nd International Conference on Disruptive Technologies (ICDT) pp. 1191-1196, 2024, March. IEEE.
[11]S. Baswaraju, A. Thirumalraj, & B. Manjunatha, “Unlocking the Potential of Deep Learning in Knee Bone Cancer Diagnosis Using MSCSA-Net Segmentation and MLGC-LT Net Classification.” In Sustainable Development Using Private AI, pp. 190-213. CRC Press.
[12]S. Ghosh, J. Mallick, M. Chakraborty, & S.Gupta. “Automated bone fracture detection in X-ray imaging to improve orthopaedic diagnostics in healthcare.” Procedia Computer Science, vol.233, pp.832-840, 2024.
[13]M. D. Baig, H. B. U. Haq, W. Akram, & A. Shahzad. “Transfer Learning Empowered Bone Fracture Detection.” Decision Making Advances, vol.2, no.1, pp.60-72, 2024.
[14]B. Pujitha, K. R. Sravya, N. Krishnasai, & C. Aparna. “Detection of bone fracture using deep learning.” In 2024 International Conference on Emerging Systems and Intelligent Computing (ESIC) pp. 703-708, 2024, February. IEEE.
[15]H. T. Nguyen, T. B. Tran, & T. T. Tran. “Fracture Detection in Bone: An Approach with Versions of YOLOv4.” SN Computer Science, vol.5, no.6, 765, 2024.
[16]B. N. K. Rao, B. Pranitha, B. K. Reddy, D. Varsha, & N. N. Reddy. “Bone Fracture Detection and Classification Using Deep Learning Techniques.” In Driving Smart Medical Diagnosis Through AI-Powered Technologies and Applications pp. 92-101, 2024. IGI Global.
[17]H. A. Alwzwazy, L. Alzubaidi, Z. Zhao, & Y. Gu. “FracNet: An end-to-end deep learning framework for bone fracture detection.” Pattern Recognition Letters, 2025.
[18]A. Alam, A. S. Al-Shamayleh, N. Thalji, A. Raza, E. A. Morales Barajas, E. B. Thompson, & I. Ashraf. “Novel transfer learning-based bone fracture detection using radiographic images.” BMC Medical Imaging, vol.25, no.1, 5, 2025.
[19]W. Wei, Y. Huang, J. Zheng, Y. Rao, Y. Wei, X. Tan, & H. OuYang. “YOLOv11-based multi-task learning for enhanced bone fracture detection and classification in X-ray images.” Journal of Radiation Research and Applied Sciences, vol.18 no.1, 101309, 2025.
[20]A. Abdusalomov, S. Mirzakhalilov, S. Umirzakova, O. Ismailov, D. Sultanov, R. Nasimov, & Y. I. Cho. “Lightweight Deep Learning Framework for Accurate Detection of Sports-Related Bone Fractures.” Diagnostics, vol.15 no.3, 271, 2025.
[21]T. Aldhyani, Z. A. Ahmed, B. M. Alsharbi, S. Ahmad, M. H. Al-Adhaileh, A. H. Kamal, & J. Nazeer. “Diagnosis and detection of bone fracture in radiographic images using deep learning approaches.” Frontiers in Medicine, vol.11, 1506686, 2025.
[22]M. I. Khan, J. Amin, M. A. Shehzad, & S.Iqbal. “AN INNOVATIVE HYBRID MODEL FOR ELBOW BONE FRACTURE DETECTION: INTEGRATING VIT AND CNN.” Contemporary Journal of Social Science Review, vol.3, no.1, pp.1-21, 2025.
[23]A. K. Pandey, S. Kedarnath, I. K. Argyros, & G. P. Pateel. “Improving Vertebral Fracture Detection in C-Spine CT Images Using Bayesian Probability-Based Ensemble Learning.” Algorithms, vol.18, no.4, 181, 2025.
[24]Z. Zheng, B. Y. Ryu, S. E. Kim, D. S. Song, S. H. Kim, J. W. Park, & D. H. Ro. “Deep learning for automated hip fracture detection and classification: achieving superior accuracy.” The bone & joint journal, vol.107, no.2, pp.213-220, 2025.
[25]D. Sutradhar, N. M. Fahad, M. A. K. Raiaan, M. Jonkman, & S. Azam. “Cervical spine fracture detection utilizing YOLOv8 and deep attention-based vertebrae classification ensuring XAI.” Biomedical Signal Processing and Control, vol.101, 107228, 2025.
[26]https://www.kaggle.com/datasets/bmadushanirodrigo/fracture-multi-region-x-ray-data
[27]LeCun Y, Bengio Y, Hinton G. Deep learning. nature. 2015 May 28;521(7553):436-44.
[28]He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. InProceedings of the IEEE conference on computer vision and pattern recognition 2016 (pp. 770-778).
[29]Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. InProceedings of the IEEE conference on computer vision and pattern recognition 2017 (pp. 4700-4708).
[30]Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł. and Polosukhin, I., 2017. Attention is all you need. Advances in neural information processing systems, 30.
[31]Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. InInternational Conference on Medical image computing and computer-assisted intervention 2015 Oct 5 (pp. 234-241). Cham: Springer international publishing.

International Journal of Image, Graphics and Signal Processing (IJIGSP)