Investigative Magnetic Resonance Imaging

Korean Society of Magnetic Resonance in Medicine (대한자기공명의과학회)
권호 표지
DOI QR코드

DOI QR Code

Convolutional Neural Network-Based Automatic Segmentation of Substantia Nigra on Nigrosome and Neuromelanin Sensitive MR Images

  • Kang, Junghwa (Division of Biomedical Engineering, Hankuk University of Foreign Studies) ;
  • Kim, Hyeonha (Department of Intelligent Precision Healthcare Convergence, Sungkyunkwan University) ;
  • Kim, Eunjin (Department of Electrical and Computer Engineering, Sungkyunkwan University) ;
  • Kim, Eunbi (Division of Biomedical Engineering, Hankuk University of Foreign Studies) ;
  • Lee, Hyebin (Department of Radiology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea) ;
  • Shin, Na-young (Department of Radiology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea) ;
  • Nam, Yoonho (Division of Biomedical Engineering, Hankuk University of Foreign Studies)
  • Received : 2021.06.07
  • Accepted : 2021.08.08
  • Published : 2021.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, neuromelanin and nigrosome imaging techniques have been developed to evaluate the substantia nigra in Parkinson's disease. Previous studies have shown potential benefits of quantitative analysis of neuromelanin and nigrosome images in the substantia nigra, although visual assessments have been performed to evaluate structures in most studies. In this study, we investigate the potential of using deep learning based automatic region segmentation techniques for quantitative analysis of the substantia nigra. The deep convolutional neural network was trained to automatically segment substantia nigra regions on 3D nigrosome and neuromelanin sensitive MR images obtained from 30 subjects. With a 5-fold cross-validation, the mean calculated dice similarity coefficient between manual and deep learning was 0.70 ± 0.11. Although calculated dice similarity coefficients were relatively low due to empirically drawn margins, selected slices were overlapped for more than two slices of all subjects. Our results demonstrate that deep convolutional neural network-based method could provide reliable localization of substantia nigra regions on neuromelanin and nigrosome sensitive MR images.

Keywords

Acknowledgement

This research work was supported by the scientific research grant funded by the Korean Society of Magnetic Resonance in Medicine. This research was supported by the Korean Society of Magnetic Resonance in Medicine and a grant (NRF-2020R1F1A1070517) of the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Republic of Korea.

References

  1. Louis ED, Benito-Leon J. Alzheimer's disease, Parkinson's disease and essential tremor: three common degenerative diseases with shared mechanisms? Eur J Neurol 2010;17:765-766 https://doi.org/10.1111/j.1468-1331.2010.02976.x
  2. Ungerstedt U. 6-Hydroxy-dopamine induced degeneration of central monoamine neurons. Eur J Pharmacol 1968;5:107-110 https://doi.org/10.1016/0014-2999(68)90164-7
  3. Reimao S, Pita Lobo P, Neutel D, et al. Substantia nigra neuromelanin magnetic resonance imaging in de novo Parkinson's disease patients. Eur J Neurol 2015;22:540-546 https://doi.org/10.1111/ene.12613
  4. Cho ZH. Review of recent advancement of ultra high field magnetic resonance imaging: from anatomy to tractography. Investig Magn Reson Imaging 2016;20:141-151 https://doi.org/10.13104/imri.2016.20.3.141
  5. French ED, Dillon K, Wu X. Cannabinoids excite dopamine neurons in the ventral tegmentum and substantia nigra. Neuroreport 1997;8:649-652 https://doi.org/10.1097/00001756-199702100-00014
  6. Sofic E, Paulus W, Jellinger K, Riederer P, Youdim MB. Selective increase of iron in substantia nigra zona compacta of parkinsonian brains. J Neurochem 1991;56:978-982 https://doi.org/10.1111/j.1471-4159.1991.tb02017.x
  7. Noh H, Hong S, Han B. Learning deconvolution network for semantic segmentation. In Proceedings of the IEEE International Conference on Computer Vision, 2015:1520-1528
  8. Lee D, Lee J, Ko J, Yoon J, Ryu K, Nam Y. Deep learning in MR image processing. Investig Magn Reson Imaging 2019;23:81-99 https://doi.org/10.13104/imri.2019.23.2.81
  9. Garcia-Garcia A, Orts-Escolano S, Oprea S, Villena-Martinez V, Garcia-Rodriguez J. A review on deep learning techniques applied to semantic segmentation. arXiv preprint. arXiv:1704.06857, 2017
  10. Nam Y, Gho SM, Kim DH, Kim EY, Lee J. Imaging of nigrosome 1 in substantia nigra at 3T using multiecho susceptibility map-weighted imaging (SMWI). J Magn Reson Imaging 2017;46:528-536 https://doi.org/10.1002/jmri.25553
  11. Sung YH, Lee J, Nam Y, et al. Differential involvement of nigral subregions in idiopathic parkinson's disease. Hum Brain Mapp 2018;39:542-553 https://doi.org/10.1002/hbm.23863
  12. Milletari F, Navab N, Ahmadi SA. V-net: Fully convolutional neural networks for volumetric medical image segmentation. In 2016 Fourth International Conference on 3D Vision (3DV) IEEE, 2016:565-571
  13. Sudre CH, Li W, Vercauteren T, Ourselin S, Jorge Cardoso M. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. Deep Learn Med Image Anal Multimodal Learn Clin Decis Support (2017) 2017;2017:240-248