Advanced search
1 file | 2.30 MB

Convolutional neural network pruning to accelerate membrane segmentation in electron microscopy

Joris Roels (UGent) , Jonas De Vylder (UGent) , Jan Aelterman (UGent) , Yvan Saeys (UGent) and Wilfried Philips (UGent)
Author
Organization
Abstract
Biological membranes are one of the most basic structures and regions of interest in cell biology. In the study of membranes, segment extraction is a well-known and difficult problem because of impeding noise, directional and thickness variability, etc. Recent advances in electron microscopy membrane segmentation are able to cope with such difficulties by training convolutional neural networks. However, because of the massive amount of features that have to be extracted while propagating forward, the practical usability diminishes, even with state-of-the-art GPU's. A significant part of these network features typically contains redundancy through correlation and sparsity. In this work, we propose a pruning method for convolutional neural networks that ensures the training loss increase is minimized. We show that the pruned networks, after retraining, are more efficient in terms of time and memory, without significantly affecting the network accuracy. This way, we manage to obtain real-time membrane segmentation performance, for our specific electron microscopy setup.
Keywords
electron microscopy, membrane, segmentation, pruning, convolutional neural networks, MITOCHONDRIA

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 2.30 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Roels, Joris, Jonas De Vylder, Jan Aelterman, Yvan Saeys, and Wilfried Philips. 2017. “Convolutional Neural Network Pruning to Accelerate Membrane Segmentation in Electron Microscopy.” In 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017), 633–637. New York, NY, USA: IEEE.
APA
Roels, Joris, De Vylder, J., Aelterman, J., Saeys, Y., & Philips, W. (2017). Convolutional neural network pruning to accelerate membrane segmentation in electron microscopy. 2017 IEEE 14th international symposium on biomedical imaging (ISBI 2017) (pp. 633–637). Presented at the 14th International symposium on Biomedical Imaging (ISBI 2017): From nano to macro, New York, NY, USA: IEEE.
Vancouver
1.
Roels J, De Vylder J, Aelterman J, Saeys Y, Philips W. Convolutional neural network pruning to accelerate membrane segmentation in electron microscopy. 2017 IEEE 14th international symposium on biomedical imaging (ISBI 2017). New York, NY, USA: IEEE; 2017. p. 633–7.
MLA
Roels, Joris, Jonas De Vylder, Jan Aelterman, et al. “Convolutional Neural Network Pruning to Accelerate Membrane Segmentation in Electron Microscopy.” 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017). New York, NY, USA: IEEE, 2017. 633–637. Print.
@inproceedings{8528682,
  abstract     = {Biological membranes are one of the most basic structures and regions of interest in cell biology. In the study of membranes, segment extraction is a well-known and difficult problem because of impeding noise, directional and thickness variability, etc. Recent advances in electron microscopy membrane segmentation are able to cope with such difficulties by training convolutional neural networks. However, because of the massive amount of features that have to be extracted while propagating forward, the practical usability diminishes, even with state-of-the-art GPU's. A significant part of these network features typically contains redundancy through correlation and sparsity. In this work, we propose a pruning method for convolutional neural networks that ensures the training loss increase is minimized. We show that the pruned networks, after retraining, are more efficient in terms of time and memory, without significantly affecting the network accuracy. This way, we manage to obtain real-time membrane segmentation performance, for our specific electron microscopy setup.},
  author       = {Roels, Joris and De Vylder, Jonas and Aelterman, Jan and Saeys, Yvan and Philips, Wilfried},
  booktitle    = {2017 IEEE 14th international symposium on biomedical imaging (ISBI 2017)},
  isbn         = {9781509011728},
  keyword      = {electron microscopy,membrane,segmentation,pruning,convolutional neural networks,MITOCHONDRIA},
  language     = {eng},
  location     = {Melbourne, VIC, Australia},
  pages        = {633--637},
  publisher    = {IEEE},
  title        = {Convolutional neural network pruning to accelerate membrane segmentation in electron microscopy},
  url          = {http://dx.doi.org/10.1109/isbi.2017.7950600},
  year         = {2017},
}

Altmetric
View in Altmetric
Web of Science
Times cited: