Browse > Article
http://dx.doi.org/10.9717/kmms.2021.24.8.979

Multiple Sclerosis Lesion Detection using 3D Autoencoder in Brain Magnetic Resonance Images  

Choi, Wonjune (Dept. of Computer Science and Engineering, Pusan National University)
Park, Seongsu (Major of AI., Dept. of Information Convergence Engineering, Pusan National University)
Kim, Yunsoo (Major of AI., Dept. of Information Convergence Engineering, Pusan National University)
Gahm, Jin Kyu (Dept. of Computer Science and Engineering, Pusan National University, Major of AI., Dept. of Information Convergence Engineering, Pusan National University)
Publication Information
Journal of Korea Multimedia Society / v.24, no.8, 2021 , pp. 979-987 More about this Journal
Abstract
Multiple Sclerosis (MS) can be early diagnosed by detecting lesions in brain magnetic resonance images (MRI). Unsupervised anomaly detection methods based on autoencoder have been recently proposed for automated detection of MS lesions. However, these autoencoder-based methods were developed only for 2D images (e.g. 2D cross-sectional slices) of MRI, so do not utilize the full 3D information of MRI. In this paper, therefore, we propose a novel 3D autoencoder-based framework for detection of the lesion volume of MS in MRI. We first define a 3D convolutional neural network (CNN) for full MRI volumes, and build each encoder and decoder layer of the 3D autoencoder based on 3D CNN. We also add a skip connection between the encoder and decoder layer for effective data reconstruction. In the experimental results, we compare the 3D autoencoder-based method with the 2D autoencoder models using the training datasets of 80 healthy subjects from the Human Connectome Project (HCP) and the testing datasets of 25 MS patients from the Longitudinal multiple sclerosis lesion segmentation challenge, and show that the proposed method achieves superior performance in prediction of MS lesion by up to 15%.
Keywords
MRI; Autoencoder; Anomaly Segmentation; Multiple Sclerosis; Skip Connection;
Citations & Related Records
연도 인용수 순위
  • Reference
1 M. Jenkinson, and S. Smith, "A Global Optimisation Method for Robust Affine Registration of Brain Images," Medical Image Analysis, Vol. 5, No. 2, pp. 143-156, 2001   DOI
2 M. Jenkinson, C.F. Beckmann, T.E.J Beherens, M.W. Woolrich, and S.M. Smith, "FSL," NeuroImage, Vol. 62, No. 2, pp. 782-790, 2012.   DOI
3 M. Jenkinson, P. Bannister, M. Brady, and S. Smith, "Improved Optimization for the Robust and Accurate Linear Registration and Motion Correction of Brain Images," NeuroImage, Vol. 17, No. 2, pp. 825-841, 2002.   DOI
4 A. Carass, S. Roy, A. Jog, J.L. Cuzzocreo, E. Magrath, A. Gherman et al., "Longitudinal Multiple Sclerosis Lesion Segmentation Data Resource," Data in Brief, Vol. 12, pp. 346-350, 2017.   DOI
5 R. Hussian, H. Zubair, S. Pursell, and M. Shahab, "Neurodegenerative Diseases: Regenerative Mechnisms and Novel Therapeutic Approaches," Brain Science, Vol. 8, No. 9, p. 177, 2018.   DOI
6 S. Faissner and R. Gold, "Oral Therapies for Multiple Sclerosis," Cold Spring Harbor Perspectives in Medicine, Vol. 9, No. 1, p. a032011, 2019.   DOI
7 M.S. To, I.G. Sarno, C. Chong, M. Jenkinson, and G. Carneiro, "Self-supervised Lesion Change Detection and Localisation in Longitudinal Multiple Sclerosis Braim Imaging," arXiv preprint arXiv:2106.00919, 2021.
8 C. Baur, S. Denner, B. Wiestler, N. Navab, and S. Albarqouni, "Autoencoders for Unsupervised Anomaly Segmentation in Brain MR Images: A Comparative Study," Medical Image Analysis, 101952, 2021.
9 B. Zong, Q. Song, M.R. Min, W. Cheng, C. Lumezanu, D. Cho, and H. Chen, "Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection," The International Conference on Learning Representations (ICLR), 2018.
10 H.E. Atlason, A. Love, S. Sigurdsson, V. Gudnason, and L.M. Ellingsen, "SegAE: Unsupervised White Matter Lesion Segmentation from Brain MRIs using a CNN Autoencoder," NeuroImage: Clinical, Vol. 24, pp. 102085, 2019.   DOI
11 T. Tong, G. Li, X. Liu, and Q. Gao, "Image Super-Resolution Using Dense Skip Connections," Proceeding of the IEEE international conference on computer vision, pp. 4799-4807, 2017.
12 N.J. Kwak, H.J. Shin, J.S. Yang, and T.S. Song, "CNN Applied Modified Residual Block Structure," Journal of Korea Multimedia Society, Vol. 23, No. 7, pp. 803-811, 2020.   DOI
13 K. He, X. Zhang, S. Ren, and J. Sun. "Deep Residual Learning for Image Recognition," Proceeding of the IEEE conference on computer vision and pattern recognition, pp. 770-778, 2016.
14 M. Stangel, T. Kuhlmann, P.M. Matthews, and T.J. Kilpatrick, "Achievements and Obstacles of Remyelinating Therapies in Multiple Sclerosis," Nature Reviews Neurology, Vol. 13, No. 12, pp. 742-754, 2017.   DOI
15 S. Park, Y. Kim, and J.K. Gahm, "MRI Image Super Resolution through Filter Learning Based on Surrounding Gradient Information in 3D Space," Journal of Korea Multimedia Society, Vol. 24, No. 2, pp. 178-185, 2021.   DOI
16 H. Arai, Y. Chayama, H. Iyatomi, and K. Oishi, "Significant Dimension Reduction of 3D Brain MRI using 3D Convolutional Autoencoders," Proceeding of the 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society(EMBC), pp. 5162-5165, 2018.
17 M.F. Glasser, S.N. Sotiropoulos, J.A. Wilson, T.S. Coalson, B. Fischl, J.L. Andersson et al., "The Minimal Preprocessing Pipelines for the Human Connectome Porject," NeuroImage, Vol. 80, pp. 105-124, 2013.   DOI