Method of segmenting anatomic entities in 3d digital medical images

Abstract

For each of a number of landmarks in an image an initial position of the landmark is defined. Next a neighborhood around the initial position comprising a number of candidate locations of the landmark, is sampled and a cost is associated with each of the candidate locations. A cost function expressing a weighted sum of overall gray level cost and overall shape cost for all candidate locations is optimized. A segmented anatomic entity is defined as a path through a selected combination of candidate locations for which combination the cost function is optimized. During optimization towards the optimal segmented surface / volume graph traversal methods are exploited.

Description

technical field

[0001] The present invention relates to a method for segmenting entities in digital images, more specifically to a method for segmenting anatomical entities in digital medical images. The segmentation process according to the invention is based on the use of a geometric model and a photometric model of the image.

[0002] The method of the invention is particularly useful as a process for performing geometric measurements on digital images, more particularly on digital medical images. Background technique

[0003] In radiology practice, geometric measurements are frequently used to help diagnose abnormalities. In order to perform these measurements, key user points must be placed at their corresponding anatomical landmark locations in the image (eg, displayed on a display device). Measurements such as the distance between two points or the angle between lines are based on the location of these key user points. It is also possible to assess the normality or...

Images