Multi-scale active contour segmentation

摘要

A system and method for active contour segmentation where an image for a structure and an initial position on the image are received, a multi-scale image representation comprising successive image levels each having associated therewith a representation of the image is computed, a representation of the image at a given level having a different image resolution than that of a representation of the image at a subsequent level, a given one of the levels at which noise in the image is removed is identified, the initial position is set as a current contour and the given level as a current level, the current contour is deformed at the current level to expand into an expanded contour matching a shape of the structure, the expanded contour is set as the current contour and the subsequent level as the current level, and the steps are repeated until the last level is reached.

主权项

1. A computer-implemented method for active contour segmentation of imaging data, the method comprising:
(a) receiving at least one image for a given structure and obtaining an initial position on the image; for each one of the at least one image, (b) computing a multi-scale image representation comprising a plurality of successive image levels each having associated therewith a representation of the image, a representation of the image at a given one of the plurality of successive image levels having a different image resolution than that of a representation of the image at a subsequent one of the plurality of successive image levels; (c) identifying a given one of the image levels at which noise in the image is removed; (d) setting the initial position as a current contour and the given image level as a current image level; (e) deforming the current contour at the current image level to expand into an expanded contour matching a shape of the given structure; (f) setting the expanded contour as the current contour and the subsequent one of the plurality of image levels as the current image level; and (g) repeating steps (e) and (f) until a last one of the plurality of image levels is reached, wherein identifying the given one of the image levels comprises: (h) identifying a selected representation of the image associated with a selected one of the plurality of image levels and at which the image resolution is highest; (i) defining a region of interest within the selected representation of the image; (j) performing image sample analysis to determine whether noise is present within the region of interest; (k) if no noise is present within the region of interest, determining that the selected image level is adequate for initiating contour deformation thereat and setting the selected image level as the given one of the image levels; and (l) otherwise, selecting another one of the plurality of image levels having a lower image resolution than that of the selected image level, setting the other one of the plurality of image levels as the selected image level, and repeating steps (i) to (l) until the last one of the plurality of image levels is reached.