A NEW AUGMENTATION TECHNIQUE FOR IMPROVED BONE FRACTURES IMAGE CLASSIFICATION USING YOLO

Muntadhar Jabbar  Easa; Israa H.  Ali

doi:10.30572/2018/KJE/160426

Authors

Muntadhar Jabbar Easa Software Department, College of Information Technology, University of Babylon, Hilla, Iraq
Prof.Dr Israa Hadi Ali Software Department, College of Information Technology, University of Babylon, Hilla, Iraq

DOI:

https://doi.org/10.30572/2018/KJE/160426

Keywords:

Augmentation, Bone Fracture, Deep Learning, Image Classification, Preprocessing Techniques, Yolov8

Abstract

The classification of bone fractures in medical imaging distinguishing between fractured and non-fractured bones is one of the most critical and difficult tasks. Traditional manual examination is prone to errors, especially in cases of tiny or minute fractures. Similarly, these tiny fractures are a big challenge for deep learning models, which struggle to detect and classify them effectively due to their size and variability. In this paper, we propose a new augmentation method called the Cropping Small Fracture Patches (CSFP) aimed at enhancing the representation of small fractures within training data. Our approach integrates Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance image contrast, followed by a targeted patch-based augmentation strategy that extracts small fracture regions, upscales them using Lanczos interpolation, and then combines them with original train images to enrich the dataset and used to train the YOLOv8m classification model. The proposed methodology was applied to the FracAtlas dataset and the result demonstrates a significant improvement in classification performance achieving 100% accuracy outperforming previous methods. These findings confirm the effectiveness of this approach in improving model performance; by enhancing small fracture visibility, reducing class imbalance, and improving the diversity of the dataset. However, certain limitations exist. The augmentation method relies on bounding box annotations, which may not always be available in classification datasets. Additionally, this study currently focuses on binary classification (fractured vs. non-fractured) and does not classify fracture types or anatomical locations

Downloads

Download data is not yet available.

Author Biography

Prof.Dr Israa Hadi Ali, Software Department, College of Information Technology, University of Babylon, Hilla, Iraq

Prof Dr in Software Department, College of Information Technology, University of Babylon, Hilla, Iraq

References

Abedeen, I. et al. (2023) 'FracAtlas: A Dataset for Fracture Classification, Localization and Segmentation of Musculoskeletal Radiographs,' Scientific Data, 10(1). https://doi.org/10.1038/s41597-023-02432-4.

Akhlaq, F. et al. (2024) 'Diving deep into bone anomalies on the FRACAtlas dataset using deep learning and explainable AI,' IEEE, pp. 1–6. https://doi.org/10.1109/icect61618.2024.10581288.

Akyon, F.C., Altinuc, S.O. and Temizel, A. (2022) 'Slicing Aided Hyper Inference and Fine-Tuning for Small Object Detection,' 2022 IEEE International Conference on Image Processing (ICIP). https://doi.org/10.1109/icip46576.2022.9897990.

Alaa, R., Hussein, E.A., Al-Libawy, H., 2024. Object Detection Algorithms Implementation on Embedded Devices: Challenges and Suggested Solutions. Kufa Journal of Engineering 15, 148–169. https://doi.org/10.30572/2018/KJE/150309

Alshahrani, A. and Alsairafi, A. (2024) 'Bone Fracture Classification using Convolutional Neural Networks from X-ray Images,' Engineering Technology & Applied Science Research, 14(5), pp. 16640–16645. https://doi.org/10.48084/etasr.8050.

Azam, N.Z.F.N., Yazid, H. and Rahim, S.A. (2022) 'Performance Analysis on Interpolation-based Methods for Fingerprint Images,' IEEE, pp. 135–140. https://doi.org/10.1109/icspc55597.2022.10001744.

Bhuiyan, S.S., Tarin, L.T., Niaz, Md.S., Dolon, M.N., Afroz, A., Rahman, R., 2024. Fracture Classification in Musculoskeletal Radiographs Using Custom CNN and Ensemble Learning, in: 2024 6th International Conference on Electrical Engineering and Information & Communication Technology (ICEEICT). IEEE, pp. 764–769. https://doi.org/10.1109/ICEEICT62016.2024.10534439

Chen, P. et al. (2024) 'WCAY object detection of fractures for X-ray images of multiple sites,' Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-77878-6.

Gupta, S. and Sharma, D. (2024) 'Bone Fracture Classification using Transfer Learning,' arXiv (Cornell University). https://doi.org/10.48550/arxiv.2406.15958.

Hayati, M. et al. (2023) 'Impact of CLAHE-based image enhancement for diabetic retinopathy classification through deep learning,' Procedia Computer Science, 216, pp. 57–66. https://doi.org/10.1016/j.procs.2022.12.111.

Hussain, M. (2024) 'YOLOv5, YOLOv8 and YOLOv10: The Go-To Detectors for Real-time Vision,' arXiv (Cornell University). https://doi.org/10.48550/arxiv.2407.02988.

Joshi, D. and Singh, T.P. (2020) 'A survey of fracture detection techniques in bone X-ray images,' Artificial Intelligence Review, 53(6), pp. 4475–4517. https://doi.org/10.1007/s10462-019-09799-0.

Kisantal, M. et al. (2019) 'Augmentation for small object detection,' arXiv (Cornell University). https://doi.org/10.48550/arxiv.1902.07296.

Kraus, M., Anteby, R., Konen, E., Eshed, I., Klang, E., 2024. Artificial intelligence for X-ray scaphoid fracture detection: a systematic review and diagnostic test accuracy meta-analysis. Eur Radiol. https://doi.org/10.1007/s00330-023-10473-x

Kutbi, M. (2024) 'Artificial Intelligence-Based Applications for Bone Fracture Detection Using Medical Images: A Systematic Review,' Diagnostics, 14(17), p. 1879. https://doi.org/10.3390/diagnostics14171879.

Meena, T. and Roy, S. (2022) 'Bone Fracture Detection Using Deep Supervised Learning from Radiological Images: A Paradigm Shift,' Diagnostics, 12(10), p. 2420. https://doi.org/10.3390/diagnostics12102420.

Mohammed, H.A., Kareem, S.W., Mohammed, A.S., 2022. A Comparative Evaluation of Deep Learning Methods in Digital Image Classification. Kufa Journal of Engineering 13, 53–69. https://doi.org/10.30572/2018/KJE/130405

Moraes, T. et al. (2019b) 'Medical image interpolation based on 3D Lanczos filtering,' Computer Methods in Biomechanics and Biomedical Engineering Imaging & Visualization, 8(3), pp. 294–300. https://doi.org/10.1080/21681163.2019.1683469.

Mumuni, A. and Mumuni, F. (2022) 'Data augmentation: A comprehensive survey of modern approaches,' Array, 16, p. 100258. https://doi.org/10.1016/j.array.2022.100258.

Musa, P., Rafi, F.A. and Lamsani, M. (2018) 'A Review: Contrast-Limited Adaptive Histogram Equalization (CLAHE) methods to help the application of face recognition,' 2018 Third International Conference on Informatics and Computing (ICIC), pp. 1–6. https://doi.org/10.1109/iac.2018.8780492.

Padilla, R., Netto, S.L. and Da Silva, E. a. B. (2020) 'A Survey on Performance Metrics for Object-Detection Algorithms,' IEEE, pp. 237–242. https://doi.org/10.1109/iwssip48289.2020.9145130.

Ruhi, S.S.F., Nahar, F. and Ashrafi, A.F. (2024) 'A novel approach towards the classification of Bone Fracture from Musculoskeletal Radiography images using Attention Based Transfer Learning,' arXiv (Cornell University). https://doi.org/10.48550/arxiv.2410.14833.

Sales, V. et al. (2023) 'Evaluation of Resampling Techniques to Provide Better Synthesized Input Data to Super-Resolution Deep Learning Model Training,' IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium, pp. 7368–7371. https://doi.org/10.1109/igarss52108.2023.10281470.

Sivalingan, H., 2024. Cloud-Smart Surveillance: Enhancing Anomaly Detection in Video Streams with Df-Convlstm-Based Vae-Gan. Kufa Journal of Engineering 15, 125–140. https://doi.org/10.30572/2018/KJE/150409

Sohan, M., Ram, T.S. and Reddy, Ch.V.R. (2024) 'A Review on YOLOv8 and Its Advancements,' Algorithms for Intelligent Systems, pp. 529–545. https://doi.org/10.1007/978-981-99-7962-2_39.

Terven, J., Córdova-Esparza, D.-M. and Romero-González, J.-A. (2023) 'A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS,' Machine Learning and Knowledge Extraction, 5(4), pp. 1680–1716. https://doi.org/10.3390/make5040083.

Unel, F.O., Ozkalayci, B.O. and Cigla, C. (2019) 'The Power of Tiling for Small Object Detection,' 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 582–591. https://doi.org/10.1109/cvprw.2019.00084.