Bilek ve Ön Kol X-Ray Görüntüleri Kullanılarak DenseNet Tabanlı Yeni Bir Anomali Tespiti Yöntemi

Year 2024, Volume: 5 Issue: 2, 18 - 29, 31.12.2024

Selahattin Güçlü , Durmuş Özdemir , Hamdi Melih Saraoğlu

https://doi.org/10.53608/estudambilisim.1550680

Cited By: 1

Abstract

X-ray görüntülerdeki anomalilerin, bilgisayar destekli tespiti büyük öneme sahiptir ve görüntü tanımanın önemli dallarından biridir. Bu çalışma, MURA (Musculoskeletal Radiographs) veri kümesinden bilek ve ön kol X-ray görüntüleri kullanılarak, paralel yapılar kullanılarak bir derin öğrenme modeli olan DenseNet'i geliştirmeyi amaçlamaktadır. Anomali tespiti için; AlexNet, DenseNet, Paralel DenseNet ve Önerilen Paralel DenseNet (ÖPDN) derin öğrenme modelleri kullanılarak, bilek ve ön kol kısmı için analiz sonuçları kıyaslanmıştır. Bilek kısmı için 1530 sağlıklı, 1523 anomali X-ray görüntüsü olmak üzere; ön kol kısmı için 1163 sağlıklı, 810 anomali X-ray görüntüsü derin öğrenme modellerinin eğitiminde kullanılmıştır. Bilek kısmı için istatiksel analiz sonucunda, test doğruluk değeri en başarılı modelin %73,86 ile ÖPDN modeli, sonraki başarılı modelin %72,11 ile AlexNet olduğu görülmüştür. Ön kol kısmı için istatiksel analiz sonucunda, test doğruluk değeri en başarılı modelin %74,07 ile ÖPDN modeli, %73,06 ile AlexNet ve Paralel DenseNet modeli olduğu görülmüştür. Kas iskelet bilek ve ön kol X-ray görüntüleri kullanılarak yapılan anomali tespitinde; DenseNet tabanlı geliştirilen ÖPDN modelinin, klasik DenseNet modele göre test doğruluk değeri açısından daha başarılı olduğu görülmüştür.

Keywords

Derin Öğrenme , DenseNet , MURA Veri Seti

A New DenseNet Based Anomaly Detection Method Using Wrist and Forearm X-Ray Images

Abstract

Computer-aided detection of anomalies in X-ray images is of great importance and is one of the important branches of image recognition. This study aims to develop DenseNet, a deep learning model using parallel structures, using wrist and forearm X-ray images from the MURA (Musculoskeletal Radiographs) dataset. For anomaly detection; Using AlexNet, DenseNet, Parallel DenseNet and Proposed Parallel DenseNet (ÖPDN) deep learning models, the analysis results for the wrist and forearm were compared. For the wrist part, 1530 healthy and 1523 anomaly X-ray images; 1163 healthy and 810 anomaly X-ray images for the forearm part were used to train deep learning models. As a result of the statistical analysis for the wrist part, it was seen that the most successful model with the test accuracy value was the ÖPDN model with 73.86%, and the next successful model was AlexNet with 72.11%. As a result of the statistical analysis for the forearm part, the most successful test accuracy value was found to be the ÖPDN model with 74.07%, AlexNet and Parallel DenseNet models with 73.06%. In anomaly detection using musculoskeletal wrist and forearm X-ray images, it was observed that the DenseNet-based ÖPDN model was more successful in terms of test accuracy than the classical DenseNet model.

Keywords

Deep Learning , DenseNet , MURA Data Set

References

• Barhoom, A. M., Al-Hiealy, M. R. J., & Abu-Naser, S. S. 2022. Bone Abnormalities Detection and Classification Using Deep Learning-VGG16 Algorithm. Journal of Theoretical and Applied Information Technology, 100(20), 6173-6184.
• Lu, S., Wang, S., & Wang, G. 2022. Automated universal fractures detection in X-ray images based on deep learning approach. Multimedia Tools and Applications, 81(30), 44487-44503.
• Kandel, I., & Castelli, M. 2021. Improving convolutional neural networks performance for image classification using test time augmentation: a case study using MURA dataset. Health Information Science and Systems, 9(1). doi:10.1007/s13755-021-00163-7
• He, Minliang & Wang, Xuming & Zhao, Yijun. 2021. A calibrated deep learning ensemble for abnormality detection in musculoskeletal radiographs. Scientific Reports. 11. 10.1038/s41598-021-88578-w.
• Suganyadevi, S., Seethalakshmi, V., & Balasamy, K. 2022. A review on deep learning in medical image analysis. International Journal of Multimedia Information Retrieval, 11(1), 19-38.
• Guo, X., Gichoya, J. W., Trivedi, H., Purkayastha, S., & Banerjee, I. 2021. MedShift: identifying shift data for medical dataset curation. arXiv preprint arXiv:2112.13885.
• Urakawa, T., Tanaka, Y., Goto, S., Matsuzawa, H., Watanabe, K., & Endo, N. 2019. Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network. Skeletal radiology, 48, 239-244.
• Harini, N., Ramji, B., Sriram, S., Sowmya, V., & Soman, K. P. 2020. Musculoskeletal radiographs classification using deep learning. In Deep learning for data analytics (pp. 79-98). Academic Press.
• Barhoom, A. M., Al-Hiealy, M. R. J., & Abu-Naser, S. S. 2022. Bone Abnormalities Detection and Classification Using Deep Learning-VGG16 Algorithm. Journal of Theoretical and Applied Information Technology, 100(20), 6173-6184.
• Nguyen, H. P., Hoang, T. P., & Nguyen, H. H. 2021. A deep learning based fracture detection in arm bone X-ray images. In 2021 international conference on multimedia analysis and pattern recognition (MAPR) (pp. 1-6). IEEE.
• Alzubaidi, L., Salhi, A., A. Fadhel, M., Bai, J., Hollman, F., Italia, K., ... & Gu, Y. 2024. Trustworthy deep learning framework for the detection of abnormalities in X-ray shoulder images. Plos one, 19(3), e0299545.
• Manoila, C. P., Ciurea, A., & Albu, F. 2022. SmartMRI Framework for Segmentation of MR Images Using Multiple Deep Learning Methods. In 2022 E-Health and Bioengineering Conference (EHB) (pp. 01-04). IEEE.
• Akgül, İ., Kaya, V., Karavaş, E., Aydın, S., Baran, A. 2024. A Novel Artificial Intelligence-Based Hybrid System to Improve Breast Cancer DetectionUsing DCE-MRI. BULLETIN OF THE POLISH ACADEMY OF SCIENCES. TECHNICAL SCIENCES , vol.72, no. 3, 1-11.
• Akgül, İ., Kaya, V., Ünver, E., Karavaş, E., Baran, A., & Tuncer, S., 2023. Covid-19 detection on x-ray images using a deep learning architecture. JOURNAL OF ENGINEERING RESEARCH , vol.11, no.2B, 15-26.
• Polamuri, Dr & Kumbhkar, Makhan & Daniel, Dr. 2022. Introduction to Deep Learning.
• Ibrahem Hamdy Abdelhamid Kandel.2021. Deep Learning Techniques for Medical Image Classification. NOVA Information Management School Universidade Nova de Lisboa Lisbon, Portugal.
• McCulloch, W. S., Pitts, W. “A logical calculus of the ideas immanent in nervous activity,” The bulletin of mathematical biophysics 1943 5:4, vol. 5, no. 4, pp. 115–133, Dec. 1943, doi: 10.1007/BF02478259.
• Nassa, V. K., Satpathy, S. K., Pathak, M. K., Takale, D. G., Rawat, S., & Rana, S. 2023. A Comparative Analysis in Using Deep Learning Models Which Results in Efficient Image Data Augmentation. In 2023 4th International Conference on Computation, Automation and Knowledge Management (ICCAKM) (pp. 1-6). IEEE.
• Lecun, Y., Bengio, Y., and G. Hinton, G. 2015. “Deep learning,” Nature 2015 521:7553, vol. 521, no. 7553, pp. 436–444, doi: 10.1038/nature14539.
• Alammar, Z., Alzubaidi, L., Zhang, J., Santamaría, J., Li Y. and Gu, Y. 2022. "A Concise Review on Deep Learning for Musculoskeletal X-ray Images," 2022 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Sydney, Australia, pp. 1-8, doi: 10.1109/DICTA56598.2022.10034618.
• Piccialli, F., Di Somma, V., Giampaolo, F., Cuomo, S., & Fortino, G. 2021. A survey on deep learning in medicine: Why, how and when?. Information Fusion, 66, 111-137.
• Liu, T., Zheng, H., Zheng, P., Bao, J., Wang, J., Liu, X., & Yang, C. 2023. An expert knowledge-empowered CNN approach for welding radiographic image recognition. Advanced Engineering Informatics, 56, 101963.
• Li, C., Li, X., Chen, M., & Sun, X. 2023. Deep learning and image recognition. In 2023 IEEE 6th International Conference on Electronic Information and Communication Technology (ICEICT) (pp. 557-562). IEEE.
• Ariff, N. A. M., & Ismail, A. R. 2023. Study of adam and adamax optimizers on alexnet architecture for voice biometric authentication system. In 2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM) (pp. 1-4). IEEE.
• Sarkar, A., Maniruzzaman, M., Alahe, M. A., & Ahmad, M. 2023. An effective and novel approach for brain tumor classification using AlexNet CNN feature extractor and multiple eminent machine learning classifiers in MRIs. Journal of Sensors, 2023.
• Zahan, N., Hasan, M. Z., Uddin, M. S., Hossain, S., & Islam, S. F. 2022. A deep learning-based approach for mushroom diseases classification. In Application of Machine Learning in Agriculture (pp. 191-212). Academic Press.
• Yan, F., Huang, X., Yao, Y., Lu, M., & Li, M. 2019. Combining lstm and densenet for automatic annotation and classification of chest x-ray images. IEEE Access, 7, 74181-74189.
• Dalvi, P. P., Edla, D. R., & Purushothama, B. R. 2023. Diagnosis of coronavirus disease from chest x-ray images using DenseNet-169 architecture. SN Computer Science, 4(3), 214.
• Shaik, S., & Kirthiga, S. 2021. Automatic modulation classification using DenseNet. In 2021 5th International Conference on Computer, Communication and Signal Processing (ICCCSP) (pp. 301-305). IEEE.
• Khoei, T. T., & Kaabouch, N. 2022. Densely connected neural networks for detecting denial of service attacks on smart grid network. In 2022 IEEE 13th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) (pp. 0207-0211). IEEE.
• Zhou, T., Ye, X., Lu, H., Zheng, X., Qiu, S., & Liu, Y. 2022. Dense convolutional network and its application in medical image analysis. BioMed Research International, 2022(1), 2384830.
• Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. 2017. Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4700-4708).
• Podder, P., Alam, F. B., Mondal, M. R. H., Hasan, M. J., Rohan, A., & Bharati, S. 2023. Rethinking Densely Connected Convolutional Networks for Diagnosing Infectious Diseases. Computers, 12(5), 95.
• Yang, J., Zhang, L., Tang, X. “CrodenseNet: An efficient parallel cross DenseNet for COVID-19 infection detection”, Biomedical Signal Processing and Control, 2022, doi:10.1016/j.bspc.2022.103775.
• Yin, L., Hong, P., Zheng, G., Chen, H., Deng, W. 2022. A Novel Image Recognition Method Based on DenseNet and DPRN. Appl. Sci. 12, 4232. https://doi.org/10.3390/ app12094232
• Mehr, Goodarz. 2020. Automating Abnormality Detection in Musculoskeletal Radiographs through Deep Learning.
• Mehta, R., Pareek, P., Jayaswal, R., Patil, S., & Vyas, K. 2023. A bone fracture detection using ai-based techniques. Scalable Computing: Practice and Experience, 24(2), 161-171.
• Arangarajan, P., Kumar, C. S., Shunmugakarpagam, N., Vijayabhasker, R., & Gayathri, C. 2023. A Improved Training Method for Deep Learning Based Anatomical Classification of X-Rays. In 2023 International Conference on System, Computation, Automation and Networking (ICSCAN) (pp. 1-6). IEEE.
• Jaros Vojtech. 2022. Detecting abnormalities in X-Ray images using Neural Networks. Bachelor’s thesis. Czech Technical University in Prague, Faculty of Information Technology.
• Kandel, I., Castelli, M., & Popovič, A. 2021. Comparing Stacking Ensemble Techniques to Improve Musculoskeletal Fracture Image Classification. Journal of Imaging, 7(6), 100. doi:10.3390/jimaging7060100
• Siddiqui, A. 2020. neXt-Ray: Deep Learning on Bone X-Rays.
• Liao, L., Liu, W., & Liu, S. 2023. Effect of Bit Depth on Cloud Segmentation of Remote-Sensing Images. Remote Sensing, 15(10), 2548.
• Karna, Ankit & Jha, Aadarsh & Dahal, Alish & Pandey, Anup & Jha, Tantra. 2023. Chest X-Ray Classification using DenseNet.
• Jain, G., Mittal, D., Thakur, D., & Mittal, M. K. 2020. A deep learning approach to detect Covid-19 coronavirus with X-Ray images. Biocybernetics and biomedical engineering, 40(4), 1391-1405.
• El Asnaoui, K., & Chawki, Y. 2021. Using X-ray images and deep learning for automated detection of coronavirus disease. Journal of Biomolecular Structure and Dynamics, 39(10), 3615-3626.
• Normawati, D.,D. Ismi, P. 2019. “K-fold cross validation for selection of cardiovascular disease diagnosis features by applying rule-based datamining”. Signal and Image Processing Letters, 1(2), 62-72.
• Karabağ, C., Ter-Sarkisov, A., Alonso, E., & Reyes-Aldasoro, C. C. 2020. Radiography classification: A comparison between eleven convolutional neural networks. In 2020 Fourth International Conference on Multimedia Computing, Networking and Applications (MCNA) (pp. 119-125). IEEE.
• Fushiki, T. 2011. “Estimation of prediction error by using K-fold cross-validation”. Stat Comput 21, 137–146. doi: 10.1007/s11222-009-9153-8
• Yang, F., Wei, G., Cao, H., Xing, M., Liu, S., & Liu, J. 2020. Computer-assisted bone fractures detection based on depth feature. In IOP Conference Series: Materials Science and Engineering (Vol. 782, No. 2, p. 022114). IOP Publishing.
• Karthik, K., & Sowmya Kamath, S. 2023. MSDNet: A deep neural ensemble model for abnormality detection and classification of plain radiographs. Journal of Ambient Intelligence and Humanized Computing, 14(12), 16099-16113.
• Harini, N., Ramji, B., Sriram, S., Sowmya, V., & Soman, K. P. 2020. Musculoskeletal radiographs classification using deep learning. In Deep learning for data analytics (pp. 79-98). Academic Press.
• Kandel, I., Castelli, M., & Popovič, A. 2020. Musculoskeletal images classification for detection of fractures using transfer learning. Journal of imaging, 6(11), 127.