Automatic Skull Stripping and Brain Segmentation with U-Net in MRI Database

Year 2022, Issue: 40, 75 - 81, 30.09.2022

Alperen Derin , Ahmet Furkan Bayram , Caglar Gurkan , Abdulkadir Budak , Hakan Karataş

https://doi.org/10.31590/ejosat.1173065

Cited By: 2

Abstract

Skull stripping has an important in neuroimaging workflow. Skull stripping is a time-consuming process in the Magnetic resonance imaging (MRI). For this reason, skull stripping and brain segmentation are aimed in this study. For the this purpose, the U-NET architecture design, which is one of the frequently used models in the field of medical image segmentation, was used. Also, different loss functions such as Cross Entropy (CE), Dice, IoU, Tversky, Focal Tversky and their compound forms were tested on U-Net architecture design. The compound loss function of CE and Dice loss functions achieved the best performace with the average dice score of 0.976, average IoU score of 0.964, sensitivity of 0.972, specificity of 0.985, precision of 0.960 and accuracy of 0.981. As a result, skull stripping was performed to facilitate the detection of brain diseases.

Keywords

Brain, Skull stripping, MRI, Segmentation, U-Net

Thanks

This paper has been prepared by AKGUN Computer Incorporated Company. We would like to thank AKGUN Computer Inc. for providing all kinds of opportunities and funds for the execution of this project.

MRG Veri Tabanında U-Net ile Otomatik Kafatası Çıkartma ve Beyin Segmentasyonu

Abstract

Kafatasının çıkartılması beyin görüntüleme iş akışında önemli bir yere sahiptir. Kafatasının çıkartılması, Manyetik Rezonans Görüntülemede (MRG) zaman alan bir işlemdir. Bu nedenle bu çalışmada kafatası çıkartma ve beyin segmentasyonu amaçlanmaktadır. Bu amaçla tıbbi görüntü segmentasyonu alanında sıklıkla kullanılan modellerden biri olan U-Net mimari tasarımı kullanılmıştır. Ayrıca Cross Entropy (CE), Dice, IoU, Tversky, Focal Tversky gibi farklı kayıp fonksiyonları ve bunların bileşik formları U-Net mimari tasarımı üzerinde test edilmiştir. CE ve Dice kayıp fonksiyonlarının bileşik kayıp fonksiyonu, 0.976 ortalama dice skoru, 0.964 ortalama IoU skoru, 0.972 sensivity, 0.985 specificity, 0.960 presicion ve 0.981 accuracy ile en iyi performansı elde etmiştir. Sonuç olarak, beyin hastalıklarının tespitini kolaylaştırmak için kafatasının çıkartılması işlemi yapılmıştır.

Keywords

Beyin, Kafatasının çıkartılması, MRG, Segmentasyon, U-Net

References

• X-rays, CT Scans and MRIs - OrthoInfo - AAOS (pp. 1–4). (2017). https://orthoinfo.aaos.org/en/treatment/x-rays-ct-scans-and-mris/
• Kalavathi, P., & Prasath, V. B. S. (2016). Methods on Skull Stripping of MRI Head Scan Images—a Review. In Journal of Digital Imaging (Vol. 29, Issue 3, pp. 365–379). Springer. https://doi.org/10.1007/s10278-015-9847-8
• Hwang, H., Ur Rehman, H. Z., & Lee, S. (2019). 3D U-Net for skull stripping in brain MRI. Applied Sciences (Switzerland), 9(3), 569. https://doi.org/10.3390/app9030569
• Qamar, S., Jin, H., Zheng, R., Ahmad, P., & Usama, M. (2020). A variant form of 3D-UNet for infant brain segmentation. Future Generation Computer Systems, 108, 613–623. https://doi.org/10.1016/j.future.2019.11.021
• Wang, X., Li, X. H., Cho, J. W., Russ, B. E., Rajamani, N., Omelchenko, A., Ai, L., Korchmaros, A., Sawiak, S., Benn, R. A., Garcia-Saldivar, P., Wang, Z., Kalin, N. H., Schroeder, C. E., Craddock, R. C., Fox, A. S., Evans, A. C., Messinger, A., Milham, M. P., & Xu, T. (2021). U-net model for brain extraction: Trained on humans for transfer to non-human primates. NeuroImage, 235, 118001. https://doi.org/10.1016/j.neuroimage.2021.118001
• Kleesiek, J., Urban, G., Hubert, A., Schwarz, D., Maier-Hein, K., Bendszus, M., & Biller, A. (2016). Deep MRI brain extraction: A 3D convolutional neural network for skull stripping. NeuroImage, 129, 460–469. https://doi.org/10.1016/j.neuroimage.2016.01.024
• Li, J., Luo, Y., Shi, L., Zhang, X., Li, M., Zhang, B., & Wang, D. (2020). Automatic fetal brain extraction from 2D in utero fetal MRI slices using deep neural network. Neurocomputing, 378, 335–349. https://doi.org/10.1016/j.neucom.2019.10.032
• Weng, W., & Zhu, X. (2021). INet: Convolutional Networks for Biomedical Image Segmentation. IEEE Access, 9, 16591–16603. https://doi.org/10.1109/ACCESS.2021.3053408
• HarisIqbal88/PlotNeuralNet: Latex code for making neural networks diagrams. (n.d.). Retrieved September 5, 2022, from https://github.com/HarisIqbal88/PlotNeuralNet