A new model for anomaly detection in elbow and finger X-ray images: Proposed parallel DenseNet

• Autorzy: Güçlü Selahattin , Özdemir Durmuş , Saraoğlu Hamdi Melih

Identyfikatory

DOI

10.24425/bpasts.2025.153233

• Języki publikacji: EN

Abstrakty

EN

Image recognition is one of the essential branches of computer vision and has significant theoretical and practical importance. This study aims to enhance a deep learning model, DenseNet, by incorporating parallel structures using X-ray images from the MURA (musculoskeletal radiographs) dataset. X-ray images of the elbow and finger are analyzed using AlexNet, DenseNet, parallel DenseNet, and proposed parallel DenseNet (PPDN) deep learning models for anomaly detection, and the results are compared. For the elbow, 1534 healthy and 1630 anomaly X-ray images; for the finger, 1965 healthy and 1938 anomaly X-ray images were used to train the deep learning models. As a result of the statistical analysis, the most successful model with the test accuracy value for the elbow part was the suggested PPDN model (78.74%). The next successful model for the elbow part was AlexNet (77.05%). The most successful model for the finger part was again the PPDN model (69.97%), and the next successful model was the parallel DenseNet model for the finger part (68.94%). In anomaly detection of musculoskeletal elbow and finger X-ray images, the PPDN model is more successful than the classical DenseNet and AlexNet models in terms of test accuracy.

Słowa kluczowe

EN

DenseNet; convolution; classification; musculoskeletal radiograph

PL

DenseNet; splot; klasyfikacja; układ mięśniowo-szkieletowy; radiogram; zdjęcie rentgenowskie

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2025
• Tom: Vol. 73, nr 2
• Strony: art. no. e153233
• Opis fizyczny: Bibliogr. 46 poz., rys., tab., wykr.

Twórcy

autor

Güçlü Selahattin

• Department of Electrical and Electronics Engineering, Kütahya Dumlupınar University, Kütahya, Türkiye

autor

Özdemir Durmuş

• Department of Computer Engineering, Kütahya Dumlupınar University, Kütahya, Türkiye

• https://orcid.org/0000-0002-9543-4076

autor

Saraoğlu Hamdi Melih

• Department of Electrical and Electronics Engineering, Kütahya Dumlupınar University, Kütahya, Türkiye

• https://orcid.org/0000-0002-5075-9504

Bibliografia

• [1] M. He, X. Wang, and Y. Zhao, “A calibrated deep learning ensemble for abnormality detection in musculoskeletal radiographs,” Sci. Rep., vol. 11, no. 1, Apr. 2021, doi: 10.1038/s41598-021-88578-w.
• [2] I. Kandel and M. Castelli, “Improving convolutional neural networks performance for image classification using test time augmentation: a case study using MURA dataset,” Health Inf. Sci. Syst., vol. 9, no. 1, Jul. 2021, doi: 10.1007/s13755-021-00163-7.
• [3] R. Lindsey et al., “Deep neural network improves fracture detection by clinicians,” Proc. Nation. Acad. Sci., vol. 115, no. 45, pp. 11591–11596, Oct. 2018, doi: 10.1073/pnas.1806905115.
• [4] V. Narayan, P.K. Mall, A. Alkhayyat, K. Abhishek, S. Kumar, and P. Pandey, “Enhance-Net: An Approach to Boost the Performance of Deep Learning Model Based on Real-Time Medical Images,” J. Sensors, vol. 2023, p. e8276738, May 2023, doi: 10.1155/2023/8276738.
• [5] S. Panda and M. Jangid, “Improving the Model Performance of Deep Convolutional Neural Network in MURA Dataset,” Smart Systems and IoT: Innovations in Computing. Smart Innovation, Systems and Technologies, vol 141, pp. 531–541, Oct. 2019, doi: 10.1007/978-981-13-8406-6_51.
• [6] G. Mehr, “Automating Abnormality Detection in Musculoskeletal Radiographs through Deep Learning,” arXiv:2010.12030, Oct. 2020, doi: 10.48550/arxiv.2010.12030.
• [7] S. Liang and Y. Gu, “Towards robust and accurate detection of abnormalities in musculoskeletal radiographs with a multi-network model,” Sensors, vol. 20, no. 11, p. 3153, Jun. 2020, doi: 10.3390/s20113153.
• [8] N. Harini, B. Ramji, S. Sriram, V. Sowmya, and K.P. Soman, “Musculoskeletal radiographs classification using deep learning,” Elsevier eBooks, pp. 79–98, Jan. 2020, doi: 10.1016/b978-0-12-819764-6.00006-5.
• [9] K. Cheng, C. Iriondo, F. Calivá, J. Krogue, S. Majumdar, and V. Pedoia, “Adversarial policy gradient for deep learning image augmentation,” Lecture Notes in Computer Science, pp. 450–458, Jan. 2019, doi: 10.1007/978-3-030-32226-7_50.
• [10] S. Lysdahlgaard, “Utilizing heat maps as explainable artificial intelligence for detecting abnormalities on wrist and elbow radiographs,” Radiography, vol. 29, no. 6, pp. 1132–1138, Oct. 2023, doi: 10.1016/j.radi.2023.09.012.
• [11] A. Solovoya, I. Solovyov, and Traumai, “X-Ray bone abnormalities detection using MURA dataset,” 2020. [Online]. Available: https://arxiv.org/pdf/2008.03356.pdf (Accessed: Feb. 05, 2024).
• [12] I. Kandel, M. Castelli, and A. Popovič, “Comparing Stacking ensemble techniques to improve musculoskeletal fracture image classification,” J. Imag., vol. 7, no. 6, p. 100, Jun. 2021, doi: 10.3390/jimaging7060100.
• [13] T.C. Mondol, H. Iqbal, M.M.A. Hashem, “Deep CNN-based ensemble CADx model for musculoskeletal abnormality detection from radiographs,” in 2019 5th International Conference on Advances in Electrical Engineering (ICAEE), September 2019, pp. 392–397.
• [14] L. Morra, L. Piano, F. Lamberti, T. Tommasi, “Bridging the gap between natural and medical images through deep colorization,” in 2020 25th International Conference on Pattern Recognition (ICPR), January 2021, pp. 835–842.
• [15] M. Puttagunta and S. Ravi, “Medical image analysis based on deep learning approach,” Multimed. Tools Appl., vol. 80, pp. 24365–24398, Apr. 2021, doi: 10.1007/s11042-021-10707-4.
• [16] Z. Alammar, L. Alzubaidi, J. Zhang, J. Santamaría, Y. Li, and Y. Gu, “A concise review on deep learning for musculoskeletal X-ray Images,” 2022 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Australia, 2022, pp. 1–8, doi: 10.1109/DICTA56598.2022.10034618.
• [17] F. Piccialli, V. D. Somma, F. Giampaolo, S. Cuomo, and G. Fortino, “A survey on deep learning in medicine: Why, how and when?,” Inf. Fusion, vol. 66, pp. 111–137, Feb. 2021, doi: 10.1016/j.inffus.2020.09.006.
• [18] L. Yin, P. Hong, G. Zheng, H. Chen, and W. Deng, “A Novel Image Recognition Method Based on DenseNet and DPRN,” Appl. Sci., vol. 12, no. 9, pp. 4232–4232, Apr. 2022, doi: 10.3390/app12094232.
• [19] R. Ghalyan, A. Singh, K. Kadian, V. Kumar, and D. Yadav, “Bone X-ray classification for upper extremity radiographs,” J. Survey Fish. Sci., vol. 10, no. 1S, pp. 7082–7087, 2023, doi: 10.53555/sfs.v10i1S.2316.
• [20] I.H.A. Kandel, “Deep Learning Techniques for Medical Image Classification,” Doctoral dissertation, Universidade NOVA de Lisboa, Portugal, 2021.
• [21] S. Lu, Z. Lu, and Y.-D. Zhang, “Pathological brain detection based on AlexNet and transfer learning,” J. Comput. Sci., vol. 30, pp. 41–47, Jan. 2019, doi: 10.1016/j.jocs.2018.11.008.
• [22] S. Li, L. Wang, J. Li, and Y. Yao, “Image Classification Algorithm Based on Improved AlexNet,” J. Phys.-Conf. Series, vol. 1813, no. 1, p. 012051, Feb. 2021, doi: 10.1088/1742-6596/1813/1/012051.
• [23] N. Zahan, M. Z. Hasan, M.S. Uddin, S. Hossain, and S.F. Islam, “A deep learning-based approach for mushroom diseases classification,” Elsevier eBooks, pp. 191–212, Jan. 2022, doi: 10.1016/b978-0-323-90550-3.00005-9.
• [24] M.K. Bohmrah and H. Kaur, “Classification of Covid-19 patients using efficient Fine-tuned Deep learning DenseNet Model,” Global Transit. Proc., vol. 2, no. 2, pp. 476–483, Aug. 2021, doi: 10.1016/j.gltp.2021.08.003.
• [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, p. 95, May 2023, doi: 10.3390/computers12050095.
• [26] B. Chen, T. Zhao, J. Liu, and L. Lin, “Multipath feature recalibration DenseNet for image classification,” Int. J. Mach. Learn. Cybern., vol. 12, no. 3, pp. 651–660, Sep. 2020, doi: 10.1007/s13042-020-01194-4.
• [27] G. Huang, Z. Liu, L. Van Der Maaten, and K.Q. Weinberger, “Densely Connected Convolutional Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2261–2269, Jul. 2017, doi: 10.1109/cvpr.2017.243.
• [28] T. Zhou, X. Ye, H. Lu, X. Zheng, S. Qiu, and Y. Liu, “Dense Convolutional Network and Its Application in Medical Image Analysis,” BioMed Res. Int., vol. 2022, pp. 1–22, Apr. 2022, doi: 10.1155/2022/2384830.
• [29] J. Yang, L. Zhang, and X. Tang, “CrodenseNet: An efficient parallel cross DenseNet for COVID-19 infection detection,” Biomed. Signal Proces. Control, vol. 77, p. 103775, 2022, doi: 10.1016/j.bspc.2022.103775.
• [30] N. Kumar and V. Cutsuridis, “Deep convolutional neural networks with transfer learning for bone fracture recognition using small exemplar image datasets,” 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW), Rhodes Island, Greece, 2023, pp. 15, doi: 10.1109/ICASSPW59220.2023.10193015.
• [31] L. Alzubaidi et al., “Trustworthy deep learning framework for the detection of abnormalities in X-ray shoulder images,” Plos one, vol. 19, no. 3, p. e0299545, 2024, doi: 10.1371/journal.pone.0299545.
• [32] J. Vojtech, “Detecting abnormalities in X-Ray images using Neural Networks,” Bachelor’s thesis, Czech Technical University, Prague, 2022.
• [33] A. Siddiqui, “neXt-Ray: Deep Learning on Bone X-Rays.” [Online]. Available: https://cs230.stanford.edu/projects_fall_2020/reports/55773729.pdf (Accessed: Feb. 05, 2024).
• [34] L. Liao, W. Liu, and S. Liu, “Effect of Bit Depth on Cloud Segmentation of Remote-Sensing Images,” Remote Sens., vol. 15, no. 10, pp. 2548–2548, May 2023, doi: 10.3390/rs15102548.
• [35] A. Karna, A. Jha, A. Dahal, A. Pandey, and T. Jha, “Chest X-Ray Classification using DenseNet,” Proc 13th IOE Graduate Conference, 2023.
• [36] G. Jain, D. Mittal, D. Thakur, and M.K. Mittal, “A deep learning approach to detect Covid-19 coronavirus with X-Ray images,” Biocybern. Biomed. Eng., vol. 40, no. 4, pp. 1391–1405, Oct. 2020, doi: 10.1016/j.bbe.2020.08.008.
• [37] Ananda, C. Karabağ, A. Ter-Sarkisov, E. Alonso, and C. Carlos Reyes-Aldasoro, “Radiography Classification: A Comparison between Eleven Convolutional Neural Networks,” 2020 Fourth International Conference on Multimedia Computing, Networking and Applications (MCNA), Spain, 2020, pp. 119–125, doi: 10.1109/MCNA50957.2020.9264285.
• [38] T. Fushiki, “Estimation of prediction error by using K-fold cross-validation,” Stat. Comput., vol. 21, pp. 137–146, 2011, doi: 10.1007/s11222-009-9153-8.
• [39] D. Normawati and D.P. Ismi, “K-fold cross validation for selection of cardiovascular disease diagnosis features by applying rule-based datamining,” Signal Image Process. Lett., vol. 1, no. 2, pp. 62–72, 2019.
• [40] A.J. Bowers and X. Zhou, “Receiver operating characteristic (ROC) area under the curve (AUC): A diagnostic measure for evaluating the accuracy of predictors of education outcomes,” J. Edu. Stud. Placed Risk, vol. 24, no. 1, pp. 20–46, 2019.
• [41] F. Yang and B. Ding, “Computer aided fracture diagnosis based on integrated learning,” 2020 IEEE 3rd International Conference on Information Systems and Computer Aided Education (ICISCAE), Sep. 2020, doi: 10.1109/iciscae51034.2020.9236917.
• [42] K. Karthik and S. Sowmya Kamath, “Correction to: MSDNet: a deep neural ensemble model for abnormality detection and classification of plain radiographs,” J. Ambient Intell. Humaniz. Comput., vol. 14, pp. 17009–17010, May 2022, doi: 10.1007/s12652-022-03906-w.
• [43] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press.2016
• [44] A. Geron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, O’Reilly Media, 2019.
• [45] S.W. Wolfe, W.C. Pederson, S.H. Kozin, and M.S. Cohen, Green’s Operative Hand Surgery, Elsevier, 2010.
• [46] W.P. Cooney, The Wrist: Diagnosis and Operative Treatment. Wolters Kluwer Health, 2011.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).