Medical volume segmentation using bank of Gabor filters

ABSTRACT

In this paper, we will present an unsupervised approach for segmenting medical volume images based on texture properties. The texture properties of the volume data are defined based on spatial frequencies as implemented using a statistical method known as Gabor filters. Each Gabor filter in the bank is tuned to detect patterns of a specific frequency and orientation when convolved with a medical volume. The convolution is performed in the Fourier domain and the resulting response image is a feature which is added to our feature vector. The feature vector is thus passed into a classification/segmentation algorithm.

REFERENCES

• 1: A. C. Bovik , M. Clark , W. S. Geisler, Multichannel Texture Analysis Using Localized Spatial Filters, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.12 n.1, p.55-73, January 1990 [doi>10.1109/34.41384]
• 2: J. Bigun. Speed, frequency, and orientation tuned 3-d gabor filter banks and their design. In IAPR Int'l Conf. on Pattern Recognition, volume c, pages 184--187, 1994.
• 3: A. M. M. F. D. M. D. Chute and J. Tomaszewski. A novel stochastic combination of 3d texture features for automated segmentation of prostatic adenocarcinoma from high resolution mri. Medical Imaging Computing and Computer-Assisted Intervention, 2: 581--591, 2003.
• 4: J. G. Daugman. Uncertainty relation for resolution in space, spatial frequency and orientation optimized by two-dimensional visual cortical filters. Journal of Optical Society of America, 2: 1160--1169, 1985.
• 5: James C. Duncan , Nicholas Ayache, Medical Image Analysis: Progress over Two Decades and the Challenges Ahead, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.22 n.1, p.85-106, January 2000 [doi>10.1109/34.824822]
• 6: M. Feng and T. Reed. Dense motion field estimation by 3-d gabor representation. In ICIP International Conference on Image Processing, volume 4, pages 2555--2558, October 2004.
• 7: A. Jain and F. Farrokhnia. Unsupervised texture segmentation using gabor filters.
• 8: S. Lakare. 3D Segmentation Techniques for Medical Volumes. December 2000.
• 9: Adriana Mavilio Núñez, Texture Segmentation of a 3D Seismic Section with Wavelet Transform and Gabor Filters, Proceedings of the International Conference on Pattern Recognition, p.3358, September 03-08, 2000
• 10: S. Oh and C. Yoonsuck. Texture segmentation in 2d vs. 3d: Did 3d developmentally precede 2d?
• 11: D. M. Z. Qian and L. Axel. Extraction and tracking of mri tagging sheets using a 3d gabor filter bank. In Proceedings of Intl Conf. of the Engineering in Medicine and Biology Society, 2006.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.6 Segmentation

Additional Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.7 Feature Measurement
      I.4.9 Applications

  J. Computer Applications
  J.3 LIFE AND MEDICAL SCIENCES

General Terms:
Algorithms, Design, Experimentation

Keywords:
Gabor filters, K-means, classification, clustering, medical volume images, segmentation, texture analysis