skip to main content
10.1145/3299815.3314995acmconferencesArticle/Chapter ViewAbstractPublication Pagesacm-seConference Proceedingsconference-collections
short-paper

Fast Training of a Convolutional Neural Network for Brain MRI Classification

Published: 18 April 2019 Publication History

Abstract

The increase in AD population calls for an automated and accurate method for efficiently diagnosing AD through MRI, preferably at the stage of Early MCI (EMCI). Graphics Processing Unit (GPU), ReLU, Dropout contribute to better performance of Artificial Neural Network (ANN), resulting in a considerable decline in error rates for computer vision tasks like MRI-based diagnosis through classification. CNN-based architectures such as AlexNet have become dominant in image classification challenges. VGG-like architectures emerged as improved variants of AlexNet but the large numbers of kernels and layers in VGG tend to slow down the training process. VoxCNN has been designed for binary AD/LMCI/EMCI/Normal classification of 3D MRI images, incorporating batch normalization, Dropout and Adam for training. Although VoxCNN output decent results with a relatively small dataset as input, the fact that 150 epochs have to be performed for each run of classification is prone to yield a lengthy training process. We tweaked Dropout probability and managed to reduce training time of VoxCNN by 49% while maintaining the same level of prediction performance in binary AD/LMCI/EMCI/Normal classification. Both ROC AUC and accuracy exhibited minor loss, i.e., 0.3% and 1.4%, respectively. Standard deviation of both ROC AUC and accuracy at Epoch 75 was improved on the whole after adjustment of Dropout probability.

References

[1]
Alzheimer's Association. 2018. 2018 Alzheimer's Disease Facts and Figures. Alzheimer's & Dementia. 14, 3 (2018), pp. 367--429.
[2]
N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov. 2014. Dropout: A Simple Way to Prevent Neural Networks from Overfitting. The Journal of Machine Learning Research. 15, 1 (2014), pp. 1929--1958.
[3]
A. Krizhevsky, I. Sutskever, and G.E. Hinton. 2012. Imagenet Classification with Deep Convolutional Neural Networks. In Advances in Neural Information Processing Systems, pp. 1097--1105.
[4]
J. Liu, Y. Pan, M. Li, Z. Chen, L. Tang, C. Lu, and J. Wang. 2018. Applications of Deep Learning to MRI Images: A Survey. Big Data Mining and Analytics. 1, 1 (2018), pp. 1--18.
[5]
H. Suk, S. Lee, and D. Shen. 2014. Hierarchical Feature Representation and Multimodal Fusion with Deep Learning for AD/MCI Diagnosis. Neuroimage 101, (2014), pp. 569--582.
[6]
H. Suk, S. Lee, and D. Shen. 2015. Latent Feature Representation with Stacked Auto-encoder for AD/MCI Diagnosis. Brain Structure and Function. 220, 2 (2015), pp. 841--859.
[7]
S. Sarraf, D. D. DeSouza, J. Anderson, and G. Tofighi. 2017. DeepAD: Alzheimer's Disease Classification via Deep Convolutional Neural Networks Using MRI and fMRI. BioRxiv (2017), p. 70441.
[8]
A. Payan and G. Montana. 2015. Predicting Alzheimer's Disease: A Neuroimaging Study with 3D Convolutional Neural Networks. In ICPRAM.
[9]
E. Hosseini-Asl, M. Ghazal, A. Mahmoud, A. Aslantas, A. M. Shalaby, M. F. Casanova, G. N. Barnes, G. Gimel'farb, R. Keynton, and A. El-Baz. 2018. Alzheimer's Disease Diagnostics by a 3D Deeply Supervised Adaptable Convolutional Network. Frontiers in Bioscience (Landmark Edition). 23, (2018), p. 584.
[10]
S. Korolev, A. Safiullin, M. Belyaev, and Y. Dodonova. 2017. Residual and Plain Convolutional Neural Networks for 3D Brain MRI Classification. In 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017), pp. 835--838.
[11]
K. Simonyan and A. Zisserman. 2014. Very Deep Convolutional Networks for Large-scale Image Recognition. arXiv Preprint. arXiv1409.1556 (2014).
[12]
S. Ioffe and C. Szegedy. 2015. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. arXiv Preprint. arXiv1502.03167 (2015).
[13]
R. Pascanu, T. Mikolov, and Y. Bengio. 2013. On the Difficulty of Training Recurrent Neural Networks. In International Conference on Machine Learning, pp. 1310--1318.
[14]
D.P. Kingma and J. Ba. 2014. Adam: A Method for Stochastic Optimization. arXiv Preprint. arXiv1412.6980 (2014).
[15]
C. R. Jack Jr, M. A. Bernstein, N. C. Fox, P. Thompson, G. Alexander, D. Harvey, B. Borowski, P. J. Britson, J. L. Whitwell, and C. Ward. 2008. The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI Methods. Journal of Magnetic Resonance Imaging: An Official Journal of the International Society for Magnetic Resonance in Medicine. 27, 4 (2008), pp. 685--691.

Cited By

View all
  • (2025)Alzheimer’s Disease: Exploring Pathophysiological Hypotheses and the Role of Machine Learning in Drug DiscoveryInternational Journal of Molecular Sciences10.3390/ijms2603100426:3(1004)Online publication date: 24-Jan-2025
  • (2022)A CAD System for Alzheimer’s Disease Classification Using Neuroimaging MRI 2D SlicesComputational and Mathematical Methods in Medicine10.1155/2022/86807372022(1-11)Online publication date: 9-Aug-2022
  • (2022)A comprehensive study on early detection of Alzheimer disease using convolutional neural networkINTERNATIONAL CONFERENCE ON ADVANCES IN MATERIALS, COMPUTING AND COMMUNICATION TECHNOLOGIES: (ICAMCCT 2021)10.1063/5.0071058(050012)Online publication date: 2022
  • Show More Cited By

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
ACMSE '19: Proceedings of the 2019 ACM Southeast Conference
April 2019
295 pages
ISBN:9781450362511
DOI:10.1145/3299815
  • Conference Chair:
  • Dan Lo,
  • Program Chair:
  • Donghyun Kim,
  • Publications Chair:
  • Eric Gamess
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 18 April 2019

Permissions

Request permissions for this article.

Check for updates

Author Tags

  1. Alzheimer's Disease
  2. Convolutional Neural Network
  3. Deep Learning
  4. Magnetic Resonance Imaging
  5. Mild Cognitive Impairment
  6. Neuroimaging

Qualifiers

  • Short-paper
  • Research
  • Refereed limited

Conference

ACM SE '19
Sponsor:
ACM SE '19: 2019 ACM Southeast Conference
April 18 - 20, 2019
GA, Kennesaw, USA

Acceptance Rates

Overall Acceptance Rate 502 of 1,023 submissions, 49%

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)9
  • Downloads (Last 6 weeks)0
Reflects downloads up to 15 Feb 2025

Other Metrics

Citations

Cited By

View all
  • (2025)Alzheimer’s Disease: Exploring Pathophysiological Hypotheses and the Role of Machine Learning in Drug DiscoveryInternational Journal of Molecular Sciences10.3390/ijms2603100426:3(1004)Online publication date: 24-Jan-2025
  • (2022)A CAD System for Alzheimer’s Disease Classification Using Neuroimaging MRI 2D SlicesComputational and Mathematical Methods in Medicine10.1155/2022/86807372022(1-11)Online publication date: 9-Aug-2022
  • (2022)A comprehensive study on early detection of Alzheimer disease using convolutional neural networkINTERNATIONAL CONFERENCE ON ADVANCES IN MATERIALS, COMPUTING AND COMMUNICATION TECHNOLOGIES: (ICAMCCT 2021)10.1063/5.0071058(050012)Online publication date: 2022
  • (2021)Enhancing Learnability of Classification Algorithms Using Simple Data Preprocessing in fMRI Scans of Alzheimer’s DiseaseAdvances in Automation, Signal Processing, Instrumentation, and Control10.1007/978-981-15-8221-9_98(1055-1063)Online publication date: 5-Mar-2021

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media