Detection of arterial lumen in sonographic images based on active contours and diffusion filters

  • Authors:
  • Amr R. Abdel-Dayem

  • Affiliations:
  • Department of Mathematics and Computer Science, Laurentian University, Sudbury, Ontario, Canada

  • Venue:
  • ICIAR'10 Proceedings of the 7th international conference on Image Analysis and Recognition - Volume Part II
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a scheme for extracting carotid artery contours from ultrasound images using a modified active contour model. The scheme uses a single seed point as an input. A complex diffusion filter is used to provide a robust estimation of the image's edge map. This edge map is used to define the external energy function for the proposed active contour. The scheme produces accurate results compared to the gold standard images. Moreover, the proposed snake model was compared to two snake models found in literature. While the first model uses Canny edge detector, the second employs the Sobel operator to calculate the image's edge map. Experimental results over a set of 40 images show that the proposed model outperforms the other two models. Finally, sensitivity analysis over the entire set of test images revealed that the scheme is insensitive to the seed point location, as long as it is located inside the artery area.