Multiple Sclerosis Identification by 14-Layer Convolutional Neural Network With Batch Normalization, Dropout, and Stochastic Pooling.pdf (2.92 MB)
Download file

Multiple Sclerosis Identification by 14-Layer Convolutional Neural Network With Batch Normalization, Dropout, and Stochastic Pooling.

Download (2.92 MB)
journal contribution
posted on 20.05.2019, 13:13 by S-H Wang, C Tang, J Sun, J Yang, C Huang, P Phillips, Y-D Zhang
Aim: Multiple sclerosis is a severe brain and/or spinal cord disease. It may lead to a wide range of symptoms. Hence, the early diagnosis and treatment is quite important. Method: This study proposed a 14-layer convolutional neural network, combined with three advanced techniques: batch normalization, dropout, and stochastic pooling. The output of the stochastic pooling was obtained via sampling from a multinomial distribution formed from the activations of each pooling region. In addition, we used data augmentation method to enhance the training set. In total 10 runs were implemented with the hold-out randomly set for each run. Results: The results showed that our 14-layer CNN secured a sensitivity of 98.77 ± 0.35%, a specificity of 98.76 ± 0.58%, and an accuracy of 98.77 ± 0.39%. Conclusion: Our results were compared with CNN using maximum pooling and average pooling. The comparison shows stochastic pooling gives better performance than other two pooling methods. Furthermore, we compared our proposed method with six state-of-the-art approaches, including five traditional artificial intelligence methods and one deep learning method. The comparison shows our method is superior to all other six state-of-the-art approaches.

Funding

This paper is supported by Natural Science Foundation of China (61602250), National key research and development plan (2017YFB1103202), Henan Key Research and Development Project (182102310629), Open Fund of Guangxi Key Laboratory of Manufacturing System & Advanced Manufacturing Technology (17-259-05-011K).

History

Citation

Frontiers in Neuroscience, 2018, 12:818

Author affiliation

/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/Department of Informatics

Version

VoR (Version of Record)

Published in

Frontiers in Neuroscience

Publisher

Frontiers Media

issn

1662-4548

Acceptance date

19/10/2018

Copyright date

2018

Available date

20/05/2019

Publisher version

https://www.frontiersin.org/articles/10.3389/fnins.2018.00818/full

Language

en