Systems and methods for analyzing target contrast features in images of biological samples
AssigneeAmersham Biosciences AB
InventorsRichard L. Hansen, William J. Karsh
Life of this patent
Estimating the term…
Abstract
Systems and methods for acquiring images of biological samples and analyzing contrast features are provided. The present invention may determine whether a contrast feature in a image of a biological example qualifies as a grain (textural feature of interest). The size, intensity of contrast features, or other parameters may be tuned to find contrast features that correspond to features having significance in a given application. In addition, the total number of grains per image, the number of grains per sample, the number of grains per unit area, the area fraction occupied by grains, the ratio of intensity of grains to the intensity of the image, or other suitable characteristics may be determined.
Claims30
An automated method for analyzing target contrast features in a digital image of a biological sample, comprising:
- partitioning at least a portion of the digital image into search areas of similar size with the target contrast features;
- locating a first pixel within each of the search areas based on pixel intensity level;
- locating a test region around each of the first pixels, wherein the test regions and the target contrast features are similar in size;
- and
- calculating intensity localization values for the test regions by comparing, for each test region, the intensity of a plurality of interior pixels located within the test region to the intensity of a plurality of perimeter pixels located at the perimeter of the test region.
An automated system for analyzing target contrast features in a digital image of a biological sample, comprising:
- an image acquisition system adapted to acquire the digital image of the biological sample;
- and
- a processor adapted to partition at least a portion of the digital image into search areas of similar size with the target contrast features, locate a first pixel within each of the search areas based on pixel intensity level, locate a test region around each of the first pixels, wherein the test regions and the target contrast features are similar in size, and calculate intensity localization values for the test regions by comparing, for each test region, the intensity level of a plurality of interior pixels located within the test region to the intensity level of a plurality of perimeter pixels located at the perimeter of the test region.
Dependencies are inferred from the claim text.