Image Classification Optimization of High Resolution Tissue Images

  • M. Kozlovszky
  • K. Hegedűs
  • G. Windisch
  • L. Kovács
  • G. Pintér
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8353)

Abstract

Generic image classification methods are not performing well on tissue images. Such software solutions are producing high number of false negative and positive results, which prevents their clinical usage. We have created the MorphCeck high resolution tissue image processing framework, which enables us to collect morphological and morphometrical parameter values of the examined tissues. Size of such tissue images can easily reach the order of 100 MB–1 GB. Therefore, the image processing speed and effectiveness is an important factor. Our main goal is to accurately evaluate high resolution H-E (hematoxilin-eozin) stained colon tissue sample images, and based on the parameters classify the images into differentiated sets according to the structure and the surface manifestation of the tissues. We have interfaced our MorphCheck tissue image measurement software framework with the WND-CHARM general purpose image classifier and tried to classify high resolution tissue images with this combined software solution. The classification is by default initiated with a large training set and three main classes (healthy, adenoma, carcinoma), however the new image classification process’ wall-clock time was intolerably high on single core PC. The processing time is depending on the size/resolution of the image and the size of the training set. Due to the tissue specific image parameters the classification effectiveness was promising. So we have started a development process to decrease the processing time and further increase the accuracy of the classification. We have developed a workflow based parallel version of the MorphCheck and WND-CHARM classifier software. In collaboration with the MTA SZTAKI Application Porting Centre the WND-CHARM has been ported to some distributed computing infrastructure (DCI). The paper introduces the steps that were taken to optimize WND-CHARM applications running faster using DCIs and some performance results of the tissue image classification process.

Keywords

Application porting Medical image processing workflow HP-SEE gUSE Scalability MorphCheck WND-CHARM 

References

  1. 1.
    3DHistech. http://www.3DHistech.com. Accessed 09 Feb 2013
  2. 2.
    Aperio. http://www.aperio.com. Accessed 09 Jan 2013
  3. 3.
    Definiens. http://www.definiens.com. Accessed 09 Jan 2013
  4. 4.
    Visiopharm. http://www.visiopharm.com. Accessed 15 Feb 2013
  5. 5.
    Kacsuk, P., Farkas, Z., Sipos, G., Hermann, G., Kiss, T.: Supporting workflow-level PS applications by the P-GRADE grid portal. In: Towards Next Generation Grids Proceedings of the CoreGRID Symposium (2007)Google Scholar
  6. 6.
    Orlov, N., Shamir, L., Macura, T., Johnston, J., Mark Eckley, D., Goldberg, I.G.: WND-CHARM: Multi-purpose image classification using compound image transforms. Pattern Recognit Lett. 29(11), 1684–1693 (2008). (PMCID: PMC2573471, NIHMSID: NIHMS53719)CrossRefGoogle Scholar
  7. 7.
    Szénási, S., Vámossy, Z., Kozlovszky, M.: Evaluation and comparison of cell nuclei detection algorithms. In: 16th International Conference on Intelligent Engineering Systems (INES), Lisbon, July 2012, pp. 469–475 (2012). ISBN: 978-1-4673-2694-0Google Scholar
  8. 8.
  9. 9.
  10. 10.
    HP-SEE. http://www.hp-see.eu. Accessed 09 Feb 2013
  11. 11.
    Libtiff. http://www.libtiff.org. Accessed 09 Feb 2013
  12. 12.
    FFTW. http://www.fftw.org/download.html. Accessed 09 Feb 2013

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • M. Kozlovszky
    • 1
  • K. Hegedűs
    • 1
  • G. Windisch
    • 1
  • L. Kovács
    • 1
  • G. Pintér
    • 1
  1. 1.John von Neumann Faculty of InformaticsÓbuda UniversityBudapestHungary

Personalised recommendations