Investigation and classification of bone fracture using a deep learning model

Kamran Khan; Ayaz Ali; Abid Haider; Numan Ahmad  Khan; Jamil Khan

doi:10.47264/idea.ajset/4.1.4

Authors

Kamran Khan Department of Information Technology, University of Technology, Nowshera, Pakistan. https://orcid.org/0009-0000-2887-463X
Ayaz Ali Department of Mathematics, Government Post Graduate Jahan Zeb College, Swat, Pakistan. https://orcid.org/0009-0000-7552-5870
Abid Haider Department of Information Technology, University of Technology, Nowshera, Pakistan.
Numan Ahmad Khan Department of Computer System Engineering, COMSATS Institute of Information Technology, Attock Campus, Pakistan. https://orcid.org/0009-0004-3167-9762
Jamil Khan College of Nuclear Science and Technology, Harbin Engineering University, Harbin, China. https://orcid.org/0009-0005-4416-8299

DOI:

https://doi.org/10.47264/idea.ajset/4.1.4

Keywords:

Deep learning, Fracture detection, Convolutional neural networks, Medical imaging, Bone fractures, Healthcare AI, Radiology, Automated diagnosis, Image processing, X-ray analysis

Abstract

Bone fractures represent a large percentage of medical cases worldwide, requiring accurate and precise detection to improve patient results. The current work, therefore, suggests a new deep learning model for detecting and classifying bone fractures from medical images that addresses the limitations of traditional diagnostic methods. Leveraging the power of Convolutional Neural Networks (CNNs), the model learns imaging data, identifies subtle fracture features, and labels them with high accuracy into the pre-existing categories. They trained their model on a comprehensive dataset comprising numerous triaged and undistributable fractures, which was coupled with data augmentation to improve the model’s robustness to variation in clinical presentation. Systematic regularisation strategies applied throughout the training prevented overfitting and improved model generalizability. Preliminary results indicate strong levels of accuracy, suggesting that the model can potentially complement or replace traditional diagnostic pathways. Implementing advanced AI-based systems into clinical workflows may transform radiology by speeding up diagnostic workflows and improving uniformity for identifying fractures. This research represents progress in the science behind automated fracture diagnosis techniques and the importance of artificial intelligence in healthcare, currently in implementing solutions to complex diagnostic challenges and improvements in related patient care outcomes.

References

Bandyopadhyay, O., Biswas, A., & Bhattacharya, B. B. (2016a). Classification of long-bone fractures based on digital-geometric analysis of X-ray images. Pattern Recognition and Image Analysis, 26, 742–757. https://doi.org/10.1134/S1054661816040027

Bandyopadhyay, O., Biswas, A., & Bhattacharya, B. B. (2016b). Long-bone fracture detection in digital X-ray images based on digital-geometric techniques. Computer Methods and Programs in Biomedicine, 123, 2–14. https://doi.org/10.1016/j.cmpb.2015.09.013

Bhangare, Y., Rajeswari, K., & Game, P. S. (2024). Wrist bone fracture classification using least entropy combiner for ensemble learning. Journal of Engineering Science & Technology Review, 17(3), 45–51. https://doi.org/10.25103/jestr.173.06

Brown, J. P., Engelke, K., Keaveny, T. M., Chines, A., Chapurlat, R., Foldes, A. J., ... & Libanati, C. (2021). Romosozumab improves lumbar spine bone mass and bone strength parameters relative to alendronate in postmenopausal women: results from the Active?Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk (ARCH) trial. Journal of Bone and Mineral Research, 36(11), 2139-2152.

Dell’Osa, A. H., Felice, C. J., & Simini, F. (2019). Bioimpedance and bone fracture detection. Journal of Physics: Conference Series, 1272. https://doi.org/10.1088/1742-6596/1272/1/012010

Deokar, N. D., & Thakur, A. G. (2016). Design, development and analysis of femur bone by using rapid prototyping. International Journal of Engineering Development and Research, 4(3), 881–886. https://rjwave.org/IJEDR/papers/IJEDR1603142

Deshmukh, S., Zalte, S., Vaidya, S., & Tangade, P. (2015). Bone fracture detection using image processing in Matlab. International Journal of Advent Research in Computer and Electronics (IJARCE), 15–19.

Devi, M. S., Aruna, R., Almufti, S., Punitha, P., & Kumar, R. L. (2024). Bone feature quantisation and systematised attention gate UNet-based deep learning framework for bone fracture classification. Intelligent Data Analysis, (Preprint), 1–29. https://content.iospress.com/articles/intelligent-data-analysis/ida240431

Edward, C. P., & Hepzibah, H. (2015). A robust approach for detection of the type of fracture from x-ray images. International Journal of Advanced Research in Computer and Communication Engineering, 4(3), 479–482.

Ganesan, P., Sivakumar, S., & Sundar, S. (2015). A Comparative study on MMDBM classifier incorporating various sorting procedure. Indian Journal of Science and Technology, 8(9), 868. https://doi.org/10.17485/ijst/2015/v8i9/53064

Gonzalez, C. I., Melin, P., Castro, J. R., Mendoza, O., & Castillo, O. (2016). An improved sobel edge detection method based on generalised type-2 fuzzy logic. Soft Computing, 20, 773–784. https://doi.org/10.1007/s00500-014-1541-0

Hareendranathan, A. R., Tripathi, A., Panicker, M. R., Zhang, J., Boora, N., & Jaremko, J. (2023). Deep learning approach for automatic wrist fracture detection using ultrasound bone probability maps. SN Comprehensive Clinical Medicine, 5(1), 276.

Hasnain, M. A. (2023). Deep learning-based classification of dental disease using X-rays. Journal of Computing & Biomedical Informatics, 5(01), 82–95. https://www.jcbi.org/index.php/Main/article/view/141

Jabbar, J., Hussain, M., Malik, H., Gani, A., Khan, A. H., & Shiraz, M. (2022). Deep learning based classification of wrist cracks from X-ray imaging. CMC-Computers Materials & Continua, 73(1), 1827–1844.

Johnson, B., Alizai, H., & Dempsey, M. (2021). Fast field echo resembling a CT using restricted echo-spacing (FRACTURE): a novel MRI technique with superior bone contrast. Skeletal Radiology, 50, 1705-1713.

Kaur, H. & Jain, A. (2017). Detection of Fractures in Orthopedic X-Ray Images. . International Journal of Advanced Research in Computer Science, 8(3), 545–551. https://www.researchgate.net/publication/352438797

Khan, M., Sirdeshmukh, S. P. S. M. A., & Javed, K. (2016). Evaluation of bone fracture in animal model using bio-electrical impedance analysis. Perspectives in Science, 8, 567–569. https://doi.org/10.1016/j.pisc.2016.06.022

Khatik, I. (2017). A study of various bone fracture detection techniques. Int J Eng Comput Sci, 6(5), 21418–21423.

Kishor, K., Sengar, A., Sharma, A., & Gautam, D. (2025). Osteo fracture identification using deep learning techniques. Health Services and Outcomes Research Methodology, 1–40. https://doi.org/10.1007/s10742-025-00340-1

Lanka, S. R., & Yarramalle, S. (2018). Bone fracture identification-a case study based on statistical modelling. International Journal of Advanced Research in Computer Science, 9(1). https://doi.org/1010.26483/ijarcs.v9i1.5326

Muchtar, M. A., Simanjuntak, S. E., Rahmat, R. F., Mawengkang, H., Zarlis, M., Sitompul, O. S., ... & Nasution, T. H. (2018, March). Identification tibia and fibula bone fracture location using scanline algorithm. Journal of Physics: Conference Series, 978, 012043. https://doi.org/10.1088/1742-6596/978/1/012043

Naeem, A., Khan, A. Haider., Ayubi, S., & Malik, H. (2023). Predicting the metastasis ability of prostate cancer using machine learning classifiers. Journal of Computing & Biomedical Informatics, 4(02), 1–7. https://www.jcbi.org/index.php/Main/article/view/139

Nowroozi, A., Salehi, M. A., Shobeiri, P., Agahi, S., Momtazmanesh, S., Kaviani, P., & Kalra, M. K. (2024). Artificial intelligence diagnostic accuracy in fracture detection from plain radiographs and comparing it with clinicians: a systematic review and meta-analysis. Clinical Radiology, 79(8), 579–588. https://doi.org/10.1016/j.crad.2024.04.009

Pal, O. K., Danmusa, N. A., & Yusuff, H. D. (2024, July). bonenet: human bone fractures localising and diagnosing using hybrid neural network. In 2024 10th International Conference on Smart Computing and Communication (ICSCC) (pp. 423–427). https://doi.org/10.1109/ICSCC62041.2024.10690459

Saraswat, A., Tooley, T., & Shrivastav, S. (2024). Osteoarthritis by deep learning approach. In A. Kalam, K. R. Niazi, A. Soni, S. A. Siddiqui, & A. Mundra (eds.), Intelligent computing techniques for smart energy systems. Springer.

Smith, T. O., Daniell, H., Geere, J. A., Toms, A. P., & Hing, C. B. (2012). The diagnostic accuracy of MRI for the detection of partial-and full-thickness rotator cuff tears in adults. Magnetic resonance imaging, 30(3), 336-346.

Windarto, A. P., & Alkairi, P. (2024). Bone fracture classification using convolutional neural network architecture for high-accuracy image classification. International Journal of Electrical & Computer Engineering, 14(6), 2088–8708. https://doi.org/10.11591/ijece.v14i6.pp6466-6477

Xie, Z., Lu, Q., Guo, J., Lin, W., Ge, G., Tang, Y., ... & Wang, W. (2024). Semantic segmentation for tooth cracks using improved DeepLabv3+ model. Heliyon, 10(4). e25892. https://www.cell.com/heliyon/fulltext/S2405-8440(24)01923-6