Noise resistance analysis of wavelet-based channel energy feature for breast lesion classification on ultrasound images

  • Authors:

  • Yueh-Ching Liao;Shu-Mei Guo;King-Chu Hung;Po-Chin Wang;Tsung-Lung Yang

  • Affiliations:

  • Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C.;Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C.;Department of Computer and Communication Engineering, National Kaohsiung, First University of Science and Technology, Taiwan, R.O.C.;Department of Radiology, Kaohsiung Veterans General Hospital, Taiwan, R.O.C.;Department of Radiology, Kaohsiung Veterans General Hospital, Taiwan, R.O.C.

  • Venue:
  • PCM'10 Proceedings of the Advances in multimedia information processing, and 11th Pacific Rim conference on Multimedia: Part II
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Wavelet-based channel energy with low cost and high efficacy is a valuable feature for the differential diagnoses between benign and malignant breast lesions. The new feature is a contour approach that generally suffers from lacking a reliable contour detection algorithm with convincing results due to extreme noise. For investigating a procedure suitable for clinical application, noise resistance capability of the new feature is evaluated in this study. The evaluation system consists of two snake-based contour detection algorithms associated with two pre-processes. These combinations can produce four test datasets of contour sonogram. Classification performance evaluation is based on a probabilistic neural network and a genetic algorithm used for distribution parameter determination.