Enhancement of Computed Tomography (CT) Images of the Human Spine and Spine Segmentation

Yıl 2023, Sayı: 52, 95 - 103, 15.12.2023

Murat Türkmen , Zeynep Orman

Öz

Today, there are many health problems associated with the human spine. Biomedical images are frequently used in the detection of spinal disorders. Chief among these is IT technology. Accurate and rapid detection of spinal disorders from CT images plays an important role in patient treatment. For this, quality images are needed in the evaluation phase of CT images. In this study, the discrete wavelet transform method was used for image enhancement using VerSe dataset. The vertebrae in the image are labeled with the connected component method. The vertebrae in the enhanced CT images were segmented by adapting the convolutional neural network U-Net method. With respect to the enhancement and segmentation method, the proposed model had 99.4% accuracy rate, 99.8% specificity rate, and 99.2% sensitivity rate. The dice segmentation score was increased from 65.1% to 74.5% as a result of improving the raw images. The results of the study were compared with other segmentation results made with the VerSe data set in the literature; differences were stated and more successful results were shown.

Anahtar Kelimeler

Image processing, CNN U-Net model, computed tomography, spine segmentation

İnsan Omurgasına Ait Bilgisayarlı Tomografi Görüntülerinin İyileştirilmesi ve Omur Segmentasyonu

Öz

Günümüzde insan omurgası ile ilişkili birçok sağlık sorunu mevcuttur. Omurga rahatsızlıklarının tespitinde biyomedikal görüntüler sıklıkla kullanılmaktadır. Bunların önde geleni ise Bilgisayarlı Tomografi (BT) teknolojisidir. BT görüntülerinden omurga rahatsızlıklarının doğru ve hızlı tespit edilmesi hasta tedavisinde önemli rol oynar. Bunun için BT görüntülerinin değerlendirme aşamasında kaliteli görüntülere ihtiyaç duyulmaktadır. Bu çalışmada VerSe veri kümesi kullanılarak görüntü iyileştirilmesi amacıyla ayrık dalgacık dönüşüm (discrete wavelet transform) yöntemi kullanılmıştır. Bağlı bileşen yöntemi ile görüntüdeki omurlar etiketlenmiştir. İyileştirilmiş BT görüntülerindeki omurlar evrişimli sinir ağı olan U-Net yöntemi uyarlanarak segmente edilmiştir. İyileştirme ve segmentasyon yöntemi uygulandıktan sonra doğruluk oranı %99.4, özgüllük oranı %99.8 ve hassasiyet oranı %99.2 olarak elde edilmiştir. Dice segmentasyon skoru ham görüntülerin iyileştirilmesi sonucunda %65.1’den %74.5’e yükseltilmiştir. Çalışmanın sonuçları literatürde VerSe veri kümesi ile yapılan diğer segmentasyon sonuçları ile kıyaslanmış; farklılıkları belirtilmiş ve daha başarılı sonuçlar elde edildiği gösterilmiştir.

Anahtar Kelimeler

Görüntü işleme, CNN U-Net modeli, bilgisayarlı tomografi, omur segmentasyonu

Kaynakça

• Mallat, S. (1999). A wavelet tour of signal processing. Elsevier.
• DeVore, R. A., Jawerth, B., & Lucier, B. J. (1992). Image compression through wavelet transform coding. IEEE Transactions on information theory, 38(2), 719-746.
• Mihcak, M. K., Kozintsev, I., Ramchandran, K., & Moulin, P. (1999). Low-complexity image denoising based on statistical modeling of wavelet coefficients. IEEE Signal Processing Letters, 6(12), 300-303.
• Sekuboyina, A., Husseini, M. E., Bayat, A., Löffler, M., Liebl, H., Li, H., ... & Kirschke, J. S. (2021). VerSe: A vertebrae labelling and segmentation benchmark for multi-detector CT images. Medical image analysis, 73, 102166.
• Löffler, M. T., Sekuboyina, A., Jacob, A., Grau, A. L., Scharr, A., El Husseini, M., ... & Kirschke, J. S. (2020). A vertebral segmentation dataset with fracture grading. Radiology: Artificial Intelligence, 2(4), e190138.
• Liebl, H., Schinz, D., Sekuboyina, A., Malagutti, L., Löffler, M. T., Bayat, A., ... & Kirschke, J. S. (2021). A computed tomography vertebral segmentation dataset with anatomical variations and multi-vendor scanner data. Scientific Data, 8(1), 284.
• Pang, C., Su, Z., Lin, L., Lin, G., He, J., Lu, H., ... & Pang, S. (2023). Automated measurement of spine indices on axial MR images for lumbar spinal stenosis diagnosis using segmentation‐guided regression network. Medical Physics, 50(1), 104-116.
• Qadri, S. F., Lin, H., Shen, L., Ahmad, M., Qadri, S., Khan, S., ... & Qamar, S. (2023). CT-Based Automatic Spine Segmentation Using Patch-Based Deep Learning. International Journal of Intelligent Systems, 2023.
• Yang, Z., Wang, Q., Zeng, J., Qin, P., Chai, R., & Sun, D. (2023). RAU-net: U-net network based on residual multi-scale fusion and attention skip layer for overall spine segmentation. Machine Vision and Applications, 34(1), 10.
• Cao, Y., Tan, C., Qian, W., Chai, W., Cui, L., Yang, W., ... & Shen, X. (2022, October). Automatic Spinal Ultrasound Image Segmentation and Deployment for Real-time Spine Volumetric Reconstruction. In 2022 IEEE International Conference on Unmanned Systems (ICUS) (pp. 922-927). IEEE.
• Fatima, J., Mohsan, M., Jameel, A., Akram, M. U., & Muzaffar Syed, A. (2022). Vertebrae localization and spine segmentation on radiographic images for feature‐based curvature classification for scoliosis. Concurrency and Computation: Practice and Experience, 34(26), e7300.
• Zhao, J., Sun, L., Zhou, X., Huang, S., Si, H., & Zhang, D. (2022). Residual-atrous attention network for lumbosacral plexus segmentation with MR image. Computerized Medical Imaging and Graphics, 100, 102109.
• Aydogdu, S., Stoyanov, D., Kalaskar, D., & Mazomenos, E. (2022, July). Vertebral Column Segmentation Using Single-staged CNNs. Cambridge University Press.
• Wang, D., Yang, Z., Huang, Z., & Gu, L. (2022, July). Spine Segmentation with Multi-view GCN and Boundary Constraint. In 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) (pp. 2136-2139). IEEE.
• Yamakawa, S., Ono, K., Makihara, E., Tawara, D., Yakushijin, S., & Ikushima, N. (2022, July). Textonmap optimization for spine segmentation using adaptive differential evolution. In Proceedings of the Genetic and Evolutionary Computation Conference Companion (pp. 75-76).
• Sekuboyina, A., Rempfler, M., Valentinitsch, A., Menze, B. H., & Kirschke, J. S. (2020). Labeling vertebrae with two-dimensional reformations of multidetector CT images: an adversarial approach for incorporating prior knowledge of spine anatomy. Radiology: Artificial Intelligence, 2(2), e190074.
• Agboola, H. A., & Zaccheus, J. E. (2023). Wavelet image scattering based glaucoma detection. BMC Biomedical Engineering, 5(1), 1.
• Unser, M., & Aldroubi, A. (1996). A review of wavelets in biomedical applications. Proceedings of the IEEE, 84(4), 626-638.
• Murala, S., Maheshwari, R. P., & Balasubramanian, R. (2012). Directional binary wavelet patterns for biomedical image indexing and retrieval. Journal of Medical Systems, 36, 2865-2879.
• He, L., Chao, Y., Suzuki, K., & Wu, K. (2009). Fast connected-component labeling. Pattern recognition, 42(9), 1977-1987.
• Javanmardi, M., & Tasdizen, T. (2018, April). Domain adaptation for biomedical image segmentation using adversarial training. In 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018) (pp. 554-558). IEEE.
• Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18 (pp. 234-241). Springer International Publishing.
• Falk, T., Mai, D., Bensch, R., Çiçek, Ö., Abdulkadir, A., Marrakchi, Y., ... & Ronneberger, O. (2019). Author Correction: U-Net: deep learning for cell counting, detection, and morphometry. Nature Methods, 16(4), 351-351.
• Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., ... & Darrell, T. (2014, November). Caffe: Convolutional architecture for fast feature embedding. In Proceedings of the 22nd ACM international conference on Multimedia (pp. 675-678).
• Zakharov, A., Pisov, M., Bukharaev, A., Petraikin, A., Morozov, S., Gombolevskiy, V., & Belyaev, M. (2023). Interpretable vertebral fracture quantification via anchor-free landmarks localization. Medical Image Analysis, 83, 102646.
• Altini, N., De Giosa, G., Fragasso, N., Coscia, C., Sibilano, E., Prencipe, B., ... & Bevilacqua, V. (2021, June). Segmentation and identification of vertebrae in CT scans using CNN, k-means clustering and k-NN. In Informatics (Vol. 8, No. 2, p. 40).
• Rahman, M. A., & Wang, Y. (2016). Optimizing intersection-over-union in deep neural networks for image segmentation. In International symposium on visual computing (pp. 234-244). Springer, Cham.
• Warfield, S. K., Zou, K. H., & Wells, W. M. (2004). Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation. IEEE transactions on medical imaging, 23(7), 903-921.
• Milletari, F., Navab, N., & Ahmadi, S. A. (2016, October). V-net: Fully convolutional neural networks for volumetric medical image segmentation. In 2016 fourth international conference on 3D vision (3DV) (pp. 565-571). Ieee.
• Naik, S., Doyle, S., Agner, S., Madabhushi, A., Feldman, M., & Tomaszewski, J. (2008, May). Automated gland and nuclei segmentation for grading of prostate and breast cancer histopathology. In 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro (pp. 284-287). IEEE.