BT Görüntülerinde Non-Travmatik Vertebral Kompresyon Kırıklarının DenseNet ve GAN’ları Birleştiren Hibrit Derin Öğrenme Modeli ile Tanımlanması

Abstract

Vertebral kompresyon kırıkları, özellikle yaşlı nüfus arasında yaygın bir durumdur ve genellikle osteoporoz ile diğer dejeneratif hastalıklarla ilişkilidir. Travma belirtisi göstermeyen non travmatik vertebral kompresyon kırıkları (VK’lar) tıbbi görüntülerden tanımlanması zor olabilir. Bu durum, daha etkili ve otomatik tespit yöntemlerine olan talebi artırmıştır. Bu çalışma, bilgisayarlı tomografi (BT) görüntülerinden non-travmatik VK’ları tespit etmek için DenseNet ve Üretici Karşıt Ağlar (GAN’lar) kullanan hibrit bir derin öğrenme yaklaşımını önermektedir. Kesin kırıkları olan 101 görüntü ve kırık olmayan 99 görüntü içeren bir hasta BT tarama veri seti kullanılmıştır. Hibrit modelimiz, geleneksel yöntemlere kıyasla üstün bir doğruluk göstermiştir ve kırık ve kırık olmayan vertebra ayırt etme konusunda umut verici sonuçlar sunmuştur. Bu otomatik yöntem, radyologların erken tanı ve tedavi planlamasında yardımcı olabilir, manuel görüntü analizine gereken süreyi azaltarak tanısal doğruluğu artırır. DenseNet ve GAN’ların kombinasyonu, tıbbi görüntü sınıflandırması için ileri düzey derin öğrenme tekniklerinin etkinliğini ortaya koymakta ve otomatik tıbbi tanıda gelecekteki uygulamalara kapı açmaktadır.

Keywords

• Vertebral Kompresyon Kırıkları
• DenseNet
• Üretici Karşıt Ağlar (GAN’lar)
• Bilgisayarlı Tomografi (BT) Görüntüleme
• Otomatik Tanı

IDENTIFICATION OF NON-TRAUMATIC VERTEBRAL COMPRESSION FRACTURES IN CT IMAGES USING A HYBRID DEEP LEARNING MODEL COMBINING DENSENET AND GAN

Abstract

Vertebral compression fractures are common conditions, particularly in the aging population, often linked to osteoporosis and other degenerative diseases. Non-traumatic vertebral compression fractures (VCFs) can be difficult to identify from medical images, especially those that do not show signs of trauma. This has led to a demand for more effective and automated detection methods. This study proposes a hybrid deep learning approach that uses DenseNet and Generative Adversarial Networks (GANs) to detect nontraumatic VCFs from computed tomography (CT) images. A dataset consisting of patient CT scans was used, including 101 images with confirmed fractures and 99 images without fractures. Our hybrid model demonstrated superior accuracy to conventional methods, showing promising results in distinguishing between fractured and non-fractured vertebrae. This automated method could aid radiologists in early diagnosis and treatment planning by decreasing the time needed for manual image analysis and improving diagnostic accuracy. The combination of DenseNet and GANs demonstrates the effectiveness of using advanced deep-learning techniques for medical image classification, opening the door for future applications in automated medical diagnosis.

Keywords

• Vertebral Compression Fractures
• DenseNet
• Generative Adversarial Networks (GANs)
• Computed Tomography (CT) Imaging
• Automated Diagnosis

Ethical Statement

This research was conducted in accordance with the ethical guidelines outlined in the Declaration of Helsinki and was approved by the Ethics Committee of Istanbul University-Cerrahpasa (Approval No: 409356, Date: 20.06.2022).

References

1. Alsaidi, M., Jan, M. T., Altaher, A., Zhuang, H., & Zhu, X. (2024). Tackling the class imbalanced dermoscopic image classification using data augmentation and GAN. Multimedia Tools and Applications, 83(16), 49121-49147. doi.org/10.1007/s11042-023-17067-1
2. Atasever, S., Azginoglu, N., Terzi, D. S., & Terzi, R. (2023). A comprehensive survey of deep learning research on medical image analysis with focus on transfer learning. Clinical imaging, 94, 18-41. doi.org/10.1016/j.clinimag.2022.11.003
3. Bahrami, A., Karimian, A., & Arabi, H. (2021). Comparison of different deep learning architectures for synthetic CT generation from MR images. Physica Medica, 90, 99-107. doi.org/10.1016/j.ejmp.2021.09.006
4. Bastidas-Rodriguez, M. X., Polania, L., Gruson, A., & Prieto-Ortiz, F. (2020). Deep Learning for fractographic classification in metallic materials. Engineering Failure Analysis, 113, 104532. oi.org/10.1016/j.engfailanal.2020.104532
5. Ding, Z., Li, H., Guo, Y., Zhou, D., Liu, Y., & Xie, S. (2023). M4fnet: Multimodal medical image fusion network via multi-receptive-field and multi-scale feature integration. Computers in Biology and Medicine, 159, 106923. doi.org/10.1016/j.compbiomed.2023.106923
6. Faiella, E., Pacella, G., Altomare, C., Bernetti, C., Sarli, M., Cea, L., ... & Grasso, R. F. (2022). Percutaneous vertebroplasty: A minimally invasive procedure for the management of vertebral compression fractures. Osteology, 2(4), 139-151. doi.org/10.3390/osteology2040017
7. Ferdousi, R., Yang, C., Hossain, M. A., Laamarti, F., Hossain, M. S., & Saddik, A. E. (2024). Generative Model-Driven Synthetic Training Image Generation: An Approach to Cognition in Railway Defect Detection. Cognitive Computation, 1-16. doi.org/10.1007/s12559-024-10283-3
8. Fei, R., Yao, Q., Zhu, Y., Xu, Q., Li, A., Wu, H., & Hu, B. (2020). Deep Learning Structure for Cross‐Domain Sentiment Classification Based on Improved Cross Entropy and Weight. Scientific Programming, 2020(1), 3810261. doi.org/10.1155/2020/3810261

9. Fooladgar, F., & Kasaei, S. (2020). Lightweight residual densely connected convolutional neural network. Multimedia Tools and Applications, 79, 25571-25588. doi.org/10.1007/s11042-020-09223
10. Frid-Adar, M., Diamant, I., Klang, E., Amitai, M., Goldberger, J., & Greenspan, H. (2018). GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification. Neurocomputing, 321, 321-331. doi.org/10.1016/j.neucom.2018.09.013
11. Ghazouani, H., & Barhoumi, W. (2021). Towards non-data-hungry and fully-automated diagnosis of breast cancer from mammographic images. Computers in Biology and Medicine, 139, 105011. doi.org/10.1016/j.compbiomed.2021.105011
12. Gutiérrez-González, R., Royuela, A., & Zamarron, A. (2023). Survival following vertebral compression fractures in population over 65 years old. Aging Clinical and Experimental Research, 35(8), 1609-1617. doi.org/10.1007/s40520-023-02445-4
13. Hemalatha, J., Roseline, S. A., Geetha, S., Kadry, S., & Damaševičius, R. (2021). An efficient densenet-based deep learning model for malware detection. Entropy, 23(3), 344. doi.org/10.3390/e23030344
14. Huang, W., Feng, J., Wang, H., & Sun, L. (2020). A new architecture of densely connected convolutional networks for pan-sharpening. ISPRS International Journal of Geo-Information, 9(4), 242. doi.org/10.3390/ijgi9040242
15. İncir, R., & Bozkurt, F. (2024). A study on effective data preprocessing and augmentation method in diabetic retinopathy classification using pre-trained deep learning approaches. Multimedia Tools and Applications, 83(4), 12185-12208. doi.org/10.1007/s11042-023-15754-7
16. Jiang, X., Hu, Z., Wang, S., & Zhang, Y. (2023). Deep learning for medical image-based cancer diagnosis. Cancers, 15(14), 3608. doi.org/10.3390/cancers15143608
17. Jin, C. B., Kim, H., Liu, M., Han, I. H., Lee, J. I., Lee, J. H., ... & Cui, X. (2019). DC2Anet: generating lumbar spine MR images from CT scan data based on semi-supervised learning. Applied Sciences, 9(12), 2521. doi.org/10.3390/app9122521
18. Kazeminia, S., Baur, C., Kuijper, A., van Ginneken, B., Navab, N., Albarqouni, S., & Mukhopadhyay, A. (2020). GANs for medical image analysis. Artificial intelligence in medicine, 109, 101938. doi.org/10.1016/j.artmed.2020.101938
19. Kolanu, N., Silverstone, E. J., Ho, B. H., Pham, H., Hansen, A., Pauley, E., ... & Pocock, N. A. (2020). Clinical utility of computer‐aided diagnosis of vertebral fractures from computed tomography images. Journal of Bone and Mineral Research, 35(12), 2307-2312. doi.org/10.1002/jbmr.4146
20. Kumar, M. L., Sampath, P., Srinivas, P. V. V. S., Dommeti, D., & Nallapati, S. R. (2023, September). Leveraging Deep Learning for Accurate Detection and Precise Localization of Vertebral Fractures in Medical Imaging. In 2023 4th International Conference on Smart Electronics and Communication (ICOSEC) (pp. 826-833). IEEE. doi: 10.1109/ICOSEC58147.2023.10275972.
21. Karras, T., Aittala, M., Hellsten, J., Laine, S., Lehtinen, J., & Aila, T. (2020). Training generative adversarial networks with limited data. Advances in neural information processing systems, 33, 12104-12114.
22. Lindsey, R., Daluiski, A., Chopra, S., Lachapelle, A., Mozer, M., Sicular, S., ... & Potter, H. (2018). Deep neural network improves fracture detection by clinicians. Proceedings of the National Academy of Sciences, 115(45), 11591-11596. doi:10.1073/pnas.1806905115
23. Meena, T., & Roy, S. (2022). Bone fracture detection using deep supervised learning from radiological images: A paradigm shift. Diagnostics, 12(10), 2420. doi.org/10.3390/diagnostics12102420
24. Nguyen, T., Le, T., Vu, H., & Phung, D. (2017). Dual discriminator generative adversarial nets. Advances in neural information processing systems, 30. doi.org/10.48550/arXiv.1709.03831
25. Rather, I. H., & Kumar, S. (2024). Generative adversarial network based synthetic data training model for lightweight convolutional neural networks. Multimedia Tools and Applications, 83(2), 6249-6271. doi.org/10.1007/s11042-023-15747-6
26. Saxena, D., & Cao, J. (2021). Generative adversarial networks (GANs) challenges, solutions, and future directions. ACM Computing Surveys (CSUR), 54(3), 1-42. doi.org/10.1145/3446374
27. Shazia, A., Xuan, T. Z., Chuah, J. H., Usman, J., Qian, P., & Lai, K. W. (2021). A comparative study of multiple neural network for detection of COVID-19 on chest X-ray. EURASIP journal on advances in signal processing, 2021, 1-16. doi.org/10.1186/s13634-021-00755-1
28. Shin, H. C., Tenenholtz, N. A., Rogers, J. K., Schwarz, C. G., Senjem, M. L., Gunter, J. L., ... & Michalski, M. (2018). Medical image synthesis for data augmentation and anonymization using generative adversarial networks. In Simulation and Synthesis in Medical Imaging: Third International Workshop, SASHIMI 2018, Springer International Publishing. doi.org/10.48550/arXiv.1807.10225
29. Story, M., & Congalton, R. G. (1986). Accuracy assessment: a user’s perspective. Photogrammetric Engineering and remote sensing, 52(3), 397-399.
30. Teoh, L., Ihalage, A. A., Harp, S., F. Al-Khateeb, Z., Michael-Titus, A. T., Tremoleda, J. L., & Hao, Y. (2022). Deep learning for behaviour classification in a preclinical brain injury model. PLoS one, 17(6), e0268962. doi.org/ 10.1371/journal.pone.0268962
31. Tian, Y., Wang, Q., Huang, Z., Li, W., Dai, D., Yang, M., ... & Fink, O. (2020). Off-policy reinforcement learning for efficient and effective gan architecture search. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VII 16 (pp. 175-192). Springer International Publishing. doi.org/10.48550/arXiv.2007.09180
32. Verma, R., Mehrotra, R., Rane, C., Tiwari, R., & Agariya, A. K. (2020). Synthetic image augmentation with generative adversarial network for enhanced performance in protein classification. Biomedical Engineering Letters, 10, 443-452. doi.org/10.1007/s13534-020-00162-9
33. Wang, J., Zhu, H., Wang, S. H., & Zhang, Y. D. (2021). A review of deep learning on medical image analysis. Mobile Networks and Applications, 26(1), 351-380. doi.org/10.1007/s11036-020-01672-7
34. Wardhani, N. W. S., Rochayani, M. Y., Iriany, A., Sulistyono, A. D., & Lestantyo, P. (2019, October). Cross-validation metrics for evaluating classification performance on imbalanced data. In 2019 international conference on computer, control, informatics and its applications (IC3INA) (pp. 14-18). IEEE. doi.org/10.1109/IC3INA48034.2019.8949568
35. Xue, Y., Ye, J., Zhou, Q., Long, L. R., Antani, S., Xue, Z., ... & Huang, X. (2021). Selective synthetic augmentation with HistoGAN for improved histopathology image classification. Medical image analysis, 67, 101816. doi.prg/10.1016/j.media.2020.101816
36. Yu, C., He, X., Ma, H., Qi, X., Lu, J., & Zhao, Y. (2019). S-DenseNet: a DenseNet compression model based on convolution grouping strategy using skyline method. IEEE Access, 7, 183604-183613. doi.org/10.1109/ACCESS.2019.2960315
37. Yu, H., Yang, L. T., Zhang, Q., Armstrong, D., & Deen, M. J. (2021). Convolutional neural networks for medical image analysis: state-of-the-art, comparisons, improvement and perspectives. Neurocomputing, 444, 92-110. doi.org/10.1016/j.neucom.2020.04.157
38. Zhang, J., Zheng, B., Gao, A., Feng, X., Liang, D., & Long, X. (2021). A 3D densely connected convolution neural network with connection-wise attention mechanism for Alzheimer's disease classification. Magnetic Resonance Imaging, 78, 119-126. doi.org/10.1016/j.mri.2021.02.001
39. Zhou, S. K., Greenspan, H., Davatzikos, C., Duncan, J. S., Van Ginneken, B., Madabhushi, A., ... & Summers, R. M. (2021). A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises. Proceedings of the IEEE, 109(5), 820-838. doi.org/10.1109/JPROC.2021.3054390