Analysis of A Deep Learning Algorithm for Fracture Detection In X-Ray Images
DOI:
https://doi.org/10.59395/ijadis.v6i3.1451Keywords:
Densenet, Efficientnet, Fracture, Resnet, X-rayAbstract
Identifying bone fractures in X-ray images is a complex task that requires special expertise from radiologists and can be time-consuming in clinical workflows. Deep learning offers a significant automated diagnostic solution to improve accuracy and efficiency. This study aims to analyze the performance of three Convolutional Neural Network (CNN) architectures namely, ResNet50, DenseNet169, and EfficientNet-B3 and specifically compare the performance of models trained using augmented data with that of models trained without augmentation. The research method utilizes a local dataset, which is divided equally between the fractured and non-fractured classes. Preprocessing techniques such as Contrast Limited Adaptive Histogram Equalization (CLAHE) were applied, and the models were evaluated on a separate test set (hold-out test set). Model evaluation was conducted using accuracy, precision, recall, F1-score, ROC-AUC metrics, as well as analysis through confusion matrix, classification report, sensitivity, specificity, and calibration curve to assess overall performance. The experimental results show that the application of data augmentation consistently improves the accuracy and robustness of all three models. In the augmentation scenario, EfficientNet-B3 showed the best performance, achieving an accuracy of 93.33%. This study concludes that the combination of the EfficientNet-B3 architecture with the data augmentation strategy is the most optimal and recommended approach for developing a reliable automatic detection system on local X-ray image datasets.
Downloads
References
[1] A. Alam et al., “Novel transfer learning based bone fracture detection using radiographic images,” BMC Med. Imaging, vol. 25, no. 1, pp. 1–16, Dec. 2025, doi: 10.1186/s12880-024-01546-4. DOI: https://doi.org/10.1186/s12880-024-01546-4
[2] T. Aldhyani et al., “Diagnosis and detection of bone fracture in radiographic images using deep learning approaches,” Front. Med., vol. 11, no. January, pp. 1–15, Jan. 2025, doi: 10.3389/fmed.2024.1506686. DOI: https://doi.org/10.3389/fmed.2024.1506686
[3] R. Najjar, “Redefining Radiology: A Review of Artificial Intelligence Integration in Medical Imaging,” Diagnostics 2023, Vol. 13, Page 2760, vol. 13, no. 17, p. 2760, Aug. 2023, doi: 10.3390/DIAGNOSTICS13172760. DOI: https://doi.org/10.3390/diagnostics13172760
[4] M. Li, Y. Jiang, Y. Zhang, and H. Zhu, “Medical image analysis using deep learning algorithms,” Front. Public Heal., vol. 11, no. November, pp. 1–28, Nov. 2023, doi: 10.3389/fpubh.2023.1273253. DOI: https://doi.org/10.3389/fpubh.2023.1273253
[5] A. Tahir, A. Saadia, K. Khan, A. Gul, A. Qahmash, and R. N. Akram, “Enhancing diagnosis: ensemble deep-learning model for fracture detection using X-ray images,” Clin. Radiol., vol. 79, no. 11, pp. e1394–e1402, Nov. 2024, doi: 10.1016/j.crad.2024.08.006. DOI: https://doi.org/10.1016/j.crad.2024.08.006
[6] D. Shen, G. Wu, and H. Il Suk, “Deep Learning in Medical Image Analysis,” Annu. Rev. Biomed. Eng., vol. 19, no. Volume 19, 2017, pp. 221–248, Jun. 2017, doi: 10.1146/ANNUREV-BIOENG-071516-044442/CITE/REFWORKS. DOI: https://doi.org/10.1146/annurev-bioeng-071516-044442
[7] H. Q. Nguyen et al., “VinDr-CXR: An open dataset of chest X-rays with radiologist’s annotations,” Sci. Data, vol. 9, no. 1, pp. 1–10, 2022, doi: 10.1038/s41597-022-01498-w. DOI: https://doi.org/10.1038/s41597-022-01498-w
[8] I. Banerjee et al., “‘Shortcuts’ Causing Bias in Radiology Artificial Intelligence: Causes, Evaluation, and Mitigation,” vol. 20, no. 9, pp. 842–851, 2024, doi: 10.1016/j.jacr.2023.06.025. DOI: https://doi.org/10.1016/j.jacr.2023.06.025
[9] S. Thota, P. Kandukuru, M. Sundaram, A. Ali, S. M. Basha, and N. Hima Bindu, “Deep Learning based Bone Fracture Detection,” Int. Conf. Smart Syst. Appl. Electr. Sci. ICSSES 2024, 2024, doi: 10.1109/ICSSES62373.2024.10561360. DOI: https://doi.org/10.1109/ICSSES62373.2024.10561360
[10] D. Sutradhar, N. M. Fahad, M. A. Khan Raiaan, M. Jonkman, and S. Azam, “Cervical spine fracture detection utilizing YOLOv8 and deep attention-based vertebrae classification ensuring XAI,” Biomed. Signal Process. Control, vol. 101, no. March 2024, p. 107228, 2025, doi: 10.1016/j.bspc.2024.107228. DOI: https://doi.org/10.1016/j.bspc.2024.107228
[11] Y. Ma and Y. Luo, “Bone fracture detection through the two-stage system of Crack-Sensitive Convolutional Neural Network,” Informatics Med. Unlocked, vol. 22, no. June 2020, p. 100452, Jan. 2021, doi: 10.1016/j.imu.2020.100452. DOI: https://doi.org/10.1016/j.imu.2020.100452
[12] S. Ünalan, O. Günay, I. Akkurt, K. Gunoglu, and H. O. Tekin, “A comparative study on breast cancer classification with stratified shuffle split and K-fold cross validation via ensembled machine learning,” J. Radiat. Res. Appl. Sci., vol. 17, no. 4, p. 101080, Dec. 2024, doi: 10.1016/j.jrras.2024.101080. DOI: https://doi.org/10.1016/j.jrras.2024.101080
[13] A. M. Rifai, S. Raharjo, E. Utami, and D. Ariatmanto, “Analysis for diagnosis of pneumonia symptoms using chest X-ray based on MobileNetV2 models with image enhancement using white balance and contrast limited adaptive histogram equalization (CLAHE),” Biomed. Signal Process. Control, vol. 90, p. 105857, Apr. 2024, doi: 10.1016/J.BSPC.2023.105857. DOI: https://doi.org/10.1016/j.bspc.2023.105857
[14] R. Rokhana et al., “Convolutional Neural Network untuk Pendeteksian Patah Tulang Femur pada Citra Ultrasonik B–Mode,” J. Nas. Tek. Elektro dan Teknol. Inf., vol. 8, no. 1, p. 59, 2019, doi: 10.22146/jnteti.v8i1.491. DOI: https://doi.org/10.22146/jnteti.v8i1.491
[15] S. Chauhan, R. Chauhan, R. Rawat, S. Aluvala, and R. Singh, “Fracture Detection in Radiology with Resnet-152 & CNN,” 2024 IEEE 9th Int. Conf. Converg. Technol. I2CT 2024, 2024, doi: 10.1109/I2CT61223.2024.10543297. DOI: https://doi.org/10.1109/I2CT61223.2024.10543297
[16] C. Kochgaven, P. Mishra, and S. Shitole, “Detecting Presence of COVID-19 with ResNet-18 using PyTorch,” Proc. - Int. Conf. Commun. Inf. Comput. Technol. ICCICT 2021, pp. 1–6, 2021, doi: 10.1109/ICCICT50803.2021.9510085. DOI: https://doi.org/10.1109/ICCICT50803.2021.9510085
[17] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-December, pp. 770–778, Dec. 2016, doi: 10.1109/CVPR.2016.90. DOI: https://doi.org/10.1109/CVPR.2016.90
[18] V. Kumar, C. Prabha, P. Sharma, N. Mittal, S. S. Askar, and M. Abouhawwash, “Unified deep learning models for enhanced lung cancer prediction with ResNet-50–101 and EfficientNet-B3 using DICOM images,” BMC Med. Imaging, vol. 24, no. 1, pp. 1–21, 2024, doi: 10.1186/s12880-024-01241-4. DOI: https://doi.org/10.1186/s12880-024-01241-4
[19] C. Raptis, E. Karavasilis, G. Anastasopoulos, and A. Adamopoulos, “Comparative Analysis of Conventional CNN v’s ImageNet Pretrained ResNet in Medical Image Classification,” Inf., vol. 15, no. 12, 2024, doi: 10.3390/info15120806. DOI: https://doi.org/10.3390/info15120806
[20] R. Singh et al., “Advanced dynamic ensemble framework with explainability driven insights for precision brain tumor classification across datasets,” Sci. Rep., vol. 15, no. 1, pp. 1–35, 2025, doi: 10.1038/s41598-025-14917-w. DOI: https://doi.org/10.1038/s41598-025-14917-w
[21] S. Madan, S. Kesharwani, K. V. S. Akhil, S. Balaji, K. P. Bharath, and M. R. Kumar, “Abnormality Detection in Humerus Bone Radiographs Using DenseNet,” 3rd IEEE Int. Virtual Conf. Innov. Power Adv. Comput. Technol. i-PACT 2021, 2021, doi: 10.1109/I-PACT52855.2021.9696904. DOI: https://doi.org/10.1109/i-PACT52855.2021.9696904
[22] G. Vempati, T. Paladugu, J. Sama, R. T. Kamineni, and S. Kamepalli, “Evaluating CNN and Deep Learning Models for Bone Fracture Detection: A Comparative Study of VGG19, ResNet-50, LeNet and DenseNet,” 2024 1st Int. Conf. Innov. Commun. Electr. Comput. Eng. ICICEC 2024, 2024, doi: 10.1109/ICICEC62498.2024.10808468. DOI: https://doi.org/10.1109/ICICEC62498.2024.10808468
[23] B. A. S. Al-rimy, F. Saeed, M. Al-Sarem, A. M. Albarrak, and S. N. Qasem, “An Adaptive Early Stopping Technique for DenseNet169-Based Knee Osteoarthritis Detection Model,” Diagnostics 2023, Vol. 13, Page 1903, vol. 13, no. 11, p. 1903, May 2023, doi: 10.3390/DIAGNOSTICS13111903. DOI: https://doi.org/10.3390/diagnostics13111903
[24] N. Aziz, N. Minallah, J. Frnda, M. Sher, M. Zeeshan, and A. H. Durrani, “Precision meets generalization: Enhancing brain tumor classification via pretrained DenseNet with global average pooling and hyperparameter tuning,” PLoS One, vol. 19, no. 9, pp. 1–16, 2024, doi: 10.1371/journal.pone.0307825. DOI: https://doi.org/10.1371/journal.pone.0307825
[25] P. Podder, F. B. Alam, M. R. H. Mondal, M. J. Hasan, A. Rohan, and S. Bharati, “Rethinking Densely Connected Convolutional Networks for Diagnosing Infectious Diseases,” Computers, vol. 12, no. 5, 2023, doi: 10.3390/computers12050095. DOI: https://doi.org/10.3390/computers12050095
[26] A. R. Pratap, A. Mallola, and M. S. Preethi, “An Enhanced Efficientnet Algorithm For Prediction Of Cervical Spine Fracture,” 2023 Glob. Conf. Inf. Technol. Commun. GCITC 2023, 2023, doi: 10.1109/GCITC60406.2023.10426506. DOI: https://doi.org/10.1109/GCITC60406.2023.10426506
[27] M. Jesher Joshua, J. S. Spoorthi, K. Keerthi, S. Divya Sankari, and V. Pandiyaraji, “Optimizing Bone Fracture Diagnosis: A Hybrid Approach using EfficientNet-B3 and SVM for Accurate Classification,” 2nd Int. Conf. Emerg. Res. Comput. Sci. ICERCS 2024, 2024, doi: 10.1109/ICERCS63125.2024.10894779. DOI: https://doi.org/10.1109/ICERCS63125.2024.10894779
[28] A. A. A. El-aziz, M. A. Mahmood, and S. A. El-ghany, “E ffi cientNet-B3-Based Automated Deep Learning Framework for Multiclass Endoscopic Bladder Tissue Classi fi cation,” pp. 1–38, 2025. DOI: https://doi.org/10.3390/diagnostics15192515
[29] M. A. Al-antari, M. A. Al-masni, M. T. Choi, S. M. Han, and T. S. Kim, “A fully integrated computer-aided diagnosis system for digital X-ray mammograms via deep learning detection, segmentation, and classification,” Int. J. Med. Inform., vol. 117, pp. 44–54, Sep. 2018, doi: 10.1016/J.IJMEDINF.2018.06.003. DOI: https://doi.org/10.1016/j.ijmedinf.2018.06.003
[30] V. Mudeng, J. W. Jeong, and S. W. Choe, “Simply Fine-Tuned Deep Learning-Based Classification for Breast Cancer with Mammograms,” Comput. Mater. Contin., vol. 73, no. 3, pp. 4677–4693, Jul. 2022, doi: 10.32604/CMC.2022.031046. DOI: https://doi.org/10.32604/cmc.2022.031046
[31] A. Tharwat, “Classification assessment methods,” Appl. Comput. Informatics, vol. 17, no. 1, pp. 168–192, Jul. 2020, doi: 10.1016/j.aci.2018.08.003. DOI: https://doi.org/10.1016/j.aci.2018.08.003
[32] C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, vol. 6, no. 1, pp. 1–48, Dec. 2019, doi: 10.1186/s40537-019-0197-0. DOI: https://doi.org/10.1186/s40537-019-0197-0
[33] S. Torne et al., “VGG-16, VGG-16 With Random Forest, Resnet50 With SVM, and EfficientNetB0 With XGBoost-Enhancing Bone Fracture Classification in X-Ray Using Deep Learning Models,” IEEE Access, vol. 13, no. February, pp. 25568–25577, 2025, doi: 10.1109/ACCESS.2025.3534818. DOI: https://doi.org/10.1109/ACCESS.2025.3534818
[34] K. Warin, W. Limprasert, S. Suebnukarn, T. Paipongna, P. Jantana, and S. Vicharueang, “Maxillofacial fracture detection and classification in computed tomography images using convolutional neural network-based models,” Sci. Rep., vol. 13, no. 1, pp. 1–9, Mar. 2023, doi: 10.1038/s41598-023-30640-w. DOI: https://doi.org/10.1038/s41598-023-30640-w
[35] K. Yasaka and O. Abe, “Deep learning and artificial intelligence in radiology: Current applications and future directions,” PLoS Med., vol. 15, no. 11, pp. 2–5, Nov. 2018, doi: 10.1371/journal.pmed.1002707. DOI: https://doi.org/10.1371/journal.pmed.1002707
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Alana Mulya Zebada, Endang Wahyu Pamungkas
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
Share
Plum Analytics