International Conference on Medical Image Computing and Computer-Assisted Intervention

Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015 pp 531-538

Why Does Synthesized Data Improve Multi-sequence Classification?

  • Gijs van Tulder
  • Marleen de Bruijne
Conference paper

DOI: 10.1007/978-3-319-24553-9_65

Part of the Lecture Notes in Computer Science book series (LNCS, volume 9349)
Cite this paper as:
van Tulder G., de Bruijne M. (2015) Why Does Synthesized Data Improve Multi-sequence Classification?. In: Navab N., Hornegger J., Wells W., Frangi A. (eds) Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015. Lecture Notes in Computer Science, vol 9349. Springer, Cham

Abstract

The classification and registration of incomplete multi-modal medical images, such as multi-sequence MRI with missing sequences, can sometimes be improved by replacing the missing modalities with synthetic data. This may seem counter-intuitive: synthetic data is derived from data that is already available, so it does not add new information. Why can it still improve performance? In this paper we discuss possible explanations. If the synthesis model is more flexible than the classifier, the synthesis model can provide features that the classifier could not have extracted from the original data. In addition, using synthetic information to complete incomplete samples increases the size of the training set.

We present experiments with two classifiers, linear support vector machines (SVMs) and random forests, together with two synthesis methods that can replace missing data in an image classification problem: neural networks and restricted Boltzmann machines (RBMs). We used data from the BRATS 2013 brain tumor segmentation challenge, which includes multi-modal MRI scans with T1, T1 post-contrast, T2 and FLAIR sequences. The linear SVMs appear to benefit from the complex transformations offered by the synthesis models, whereas the random forests mostly benefit from having more training data. Training on the hidden representation from the RBM brought the accuracy of the linear SVMs close to that of random forests.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Gijs van Tulder
    • 1
  • Marleen de Bruijne
    • 1
    • 2
  1. 1.Biomedical Imaging Group RotterdamErasmus MC University Medical CenterRotterdamThe Netherlands
  2. 2.Department of Computer ScienceUniversity of CopenhagenCopenhagenDenmark

Personalised recommendations