Logo PTI Logo FedCSIS

Communication Papers of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS)

Annals of Computer Science and Information Systems, Volume 45

3D Brain Extraction from Magnetic Resonance Imaging Using Knowledge Distillation

, ,

DOI: http://dx.doi.org/10.15439/2025F2011

Citation: Communication Papers of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS), M. Bolanowski, M. Ganzha, L. Maciaszek, M. Paprzycki, D. Ślęzak (eds). ACSIS, Vol. 45, pages 71–76 ()

Full text

Abstract. Brain extraction, or skull stripping, is a crucial preprocessing step in magnetic resonance imaging (MRI), isolating brain tissue from surrounding structures like the skull and scalp. However, existing methods have limitations, such as parameter sensitivity in traditional approaches and computational complexity in advanced deep learning architectures. This study proposes a knowledge distillation framework utilizing two UNet++ models---a high-capacity teacher network and an efficient student network---for 3D brain extraction tasks. The teacher network generates detailed grayscale brain predictions, capturing subtle intensity transitions and anatomical boundaries. The student network learns to produce precise binary segmentation masks from the teacher's feature representations, guided by a hybrid loss function combining Dice, Structural Similarity Index Measure (SSIM), and Mean Squared Error (MSE). Evaluations conducted on T1-weighted, T2-weighted, and proton-density weighted MRI images from the IXI dataset demonstrated the student model's superior performance, achieving a Dice coefficient of 0.97857. These findings suggest that the proposed framework may offer a practical and accurate solution for brain extraction in diverse medical imaging scenarios.

References

  1. P. Kalavathi and V. S. Prasath, “Methods on skull stripping of MRI head scan images—a review,” J. Digit. Imaging, vol. 29, pp. 365–379, 2016.
  2. R. de Boer et al., “Accuracy and reproducibility study of automatic MRI brain tissue segmentation methods,” NeuroImage, vol. 51, no. 3, pp. 1047–1056, 2010.
  3. R. Karakis, K. Gurkahraman, G. D. Mitsis, and M. H. Boudrias, “Deep learning prediction of motor performance in stroke individuals using neuroimaging data,” J. Biomed. Inform., vol. 141, Art. no. 104357, 2023.
  4. F. Isensee et al., “Automated brain extraction of multisequence MRI using artificial neural networks,” Hum. Brain Mapp., vol. 40, no. 17, pp. 4952–4964, 2019.
  5. M. E. Brummer et al., “Automatic detection of brain contours in MRI datasets,” IEEE Trans. Image Process., vol. 12, no. 2, pp. 153–166, 1993.
  6. D. W. Shattuck, S. R. Sandor-Leahy, K. A. Schaper, D. A. Rottenberg, and R. M. Leahy, “Magnetic resonance image tissue classification using a partial volume model,” NeuroImage, vol. 13, no. 5, pp. 856–876, 2001.
  7. D. W. Shattuck and R. M. Leahy, “BrainSuite: An automated cortical surface identification tool,” Med. Image Anal., vol. 6, no. 2, pp. 129–142, 2002.
  8. S. M. Smith, “Fast robust automated brain extraction,” Hum. Brain Mapp., vol. 17, pp. 143–155, 2002.
  9. M. Jenkinson, M. Pechaud, and S. Smith, “BET2 - MR-based estimation of brain, skull and scalp surfaces,” Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB), Oxford, 2005.
  10. M. Jenkinson, C. F. Beckmann, T. E. Behrens, M. W. Woolrich, and S. M. Smith, “FSL,” NeuroImage, vol. 62, no. 2, pp. 782–790, 2012.
  11. A. M. Dale, B. Fischl, and M. I. Sereno, “Cortical surface-based analysis: I. Segmentation and surface reconstruction,” NeuroImage, vol. 9, no. 2, pp. 179–194, 1999.
  12. B. Fischl, “Freesurfer,” NeuroImage, vol. 62, no. 2, pp. 774–781, 2012.
  13. S. F. Eskildsen et al., “BEaST: Brain extraction based on non-local segmentation technique,” NeuroImage, vol. 59, no. 3, pp. 2362–2373, 2012.
  14. R. Souza et al., “An open, multi-vendor, multi-field-strength brain MR dataset and analysis of publicly available skull stripping methods agreement,” NeuroImage, vol. 170, pp. 482–494, 2018.
  15. S. K. Warfield, K. H. Zou, and W. M. Wells, “Simultaneous truth and performance level estimation (STAPLE): An algorithm for the validation of image segmentation,” IEEE Trans. Med. Imaging, vol. 23, no. 7, pp. 903–921, 2004.
  16. F. Segonne et al., “A hybrid approach to the skull stripping problem in MRI,” NeuroImage, vol. 22, no. 3, pp. 1060–1075, 2004.
  17. M. Yapici, R. Karakis, and K. Gurkahraman, “Improving brain tumor classification with deep learning using synthetic data,” Comput. Mater. Continua, vol. 74, no. 3, pp. 5049–5067, 2023.
  18. N. Siddique, S. Paheding, C. P. Elkin, and V. Devabhaktuni, “U-Net and its variants for medical image segmentation: A review of theory and applications,” IEEE Access, vol. 9, pp. 82031–82057, 2021.
  19. J. Kleesiek et al., “Deep MRI brain extraction: A 3D convolutional neural network for skull stripping,” NeuroImage, vol. 129, pp. 460–469, 2016.
  20. H. Hwang, H. Z. U. Rehman, and S. Lee, “3D U-Net for skull stripping in brain MRI,” Appl. Sci., vol. 9, no. 3, Art. no. 569, 2019.
  21. Q. Zhang et al., “FRNET: Flattened residual network for infant MRI skull stripping,” arXiv preprint https://arxiv.org/abs/1904.05578, 2019.
  22. K. Gurkahraman and C. Dasgin, “Brain extraction from magnetic resonance images using UNet modified with residual and dense layers,” Turk Doga ve Fen Dergisi, unpublished.
  23. A. Hoopes, J. S. Mora, A. V. Dalca, B. Fischl, and M. Hoffmann, “SynthStrip: Skull-stripping for any brain image,” NeuroImage, vol. 260, Art. no. 119474, 2022.
  24. X. Feng, C. Wang, S. Cheng, and L. Guo, “Automatic liver and tumor segmentation of CT based on cascaded U-Net,” in Proc. Chin. Intell. Syst. Conf., 2019, pp. 155–164.
  25. Y. Ding et al., “A stacked multi-connection simple reducing net for brain tumor segmentation,” IEEE Access, vol. 7, pp. 104011–104024, 2019.
  26. S. K. R. Chinnam, V. Sistla, and V. K. K. Kolli, “Multimodal attention-gated cascaded U-Net model for automatic brain tumor detection and segmentation,” Biomed. Signal Process. Control, vol. 78, Art. no. 103907, 2022.
  27. Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, “UNet++: A nested U-Net architecture for medical image segmentation,” in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, ser. Lecture Notes in Computer Science, vol. 11045, Springer, 2018, pp. 3–11.
  28. G. Li, K. Wang, P. Lv, et al., “Multistage feature fusion knowledge distillation,” Scientific Reports, vol. 14, article no. 13373, 2024.