Automated macromolecular crystal detection system and method
View Patent ↗An automated macromolecular method and system for detecting crystals in two-dimensional images, such as light microscopy images obtained from an array of crystallization screens. Edges are detected from the images by identifying local maxima of a phase congruency-based function associated with each image. The detected edges are segmented into discrete line segments, which are subsequently geometrically evaluated with respect to each other to identify any crystal-like qualities such as, for example, parallel lines, facing each other, similarity in length, and relative proximity. And from the evaluation a determination is made as to whether crystals are present in each image.
1. A method of detecting macromolecular crystals in images comprising:
detecting edges in said images by identifying local maxima of a phase congruency-related function associated with each image;
segmenting the detected edges into discrete line segments;
evaluating the geometric relationships that the line segments have with each other to identify any crystal-like qualities; and
determining the presence of crystals in each image based on said evaluation.
2. The method of claim 1 ,
wherein the crystal-like qualities include at least one of parallel lines, facing each other, similarity in length, and relative proximity.
3. The method of claim 1 ,
wherein said evaluation includes rating pairs of line segments based on the respective crystal-like qualities thereof.
4. The method of claim 3 ,
wherein said evaluation includes grouping the line segments based on the crystal-like quality ratings of each combination of pairs of line segments.
5. The method of claim 4 ,
wherein said evaluation includes, for each group, classifying each edge pixel by a local edge orientation.
6. The method of claim 5 ,
wherein said evaluation includes, for each orientation, calculating auto-correlation, rating the quality of the auto-correlation, and calculating a global quality score from the auto-correlation quality rating.
7. The method of claim 6 ,
wherein the quality of the auto-correlation is rated according to the number of features in the auto-correlation above a predetermined threshold, and length and location of said features.
8. The method of claim 7 ,
wherein the determination of the presence of crystals in each image is based on the global quality score for each orientation.
9. The method of claim 1 ,
wherein the phase congruency-related function is the phase congruency function itself.
10. The method of claim 1 ,
wherein the phase congruency-related function is a local energy function.
11. The method of claim 10 ,
further comprising reducing noise by applying a noise threshold on the local energy function.
12. The method of claim 1 ,
further comprising producing a skeletonal structure from the detected edges by removing non-maximal values of the edges.
13. The method of claim 1 ,
further comprising removing spur lines from the detected edges.
14. The method of claim 1 ,
further comprising smoothing the image prior to edge detection.
15. The method of claim 14 ,
wherein anisotropic diffusion is used for smoothing.
16. A computerized system for detecting macromolecular crystals from images comprising:
a digital conversion component that converts said light microscopy images into corresponding phase-based digital image data using the Fourier transform;
an edge detection component that detects edges from the image data by computing local maxima of a phase congruency-related function associated with each image;
a segmentation component that divides the detected edges into discrete line segments; and
a geometric analyzer component that evaluates the geometric relationships that the line segments have with each other to identify any crystal-like qualities, and determines whether crystals are present in each image based on said evaluation.
17. The computerized system of claim 16 ,
wherein the crystal-like qualities include at least one of parallel lines, facing each other, similarity in length, and relative proximity.
18. The computerized system of claim 16 ,
wherein said geometric analyzer rates pairs of line segments based on the respective crystal-like qualities thereof.
19. The computerized system of claim 18 ,
wherein said geometric analyzer groups the line segments based on the crystal-like quality ratings of each combination of pairs of line segments.
20. The computerized system of claim 19 ,
wherein for each group, said geometric analyzer classifies each edge pixel by a local edge orientation.
21. The computerized system of claim 20 ,
wherein for each orientation, said geometric analyzer calculates auto-correlation, rates the quality of the auto-correlation, and calculates a global quality score from the auto-correlation quality rating.
22. The computerized system of claim 21 ,
wherein the quality of the auto-correlation is rated according to the number of features in the auto-correlation above a predetermined threshold, and length and location of said features.
23. The computerized system of claim 22 ,
wherein said geometric analyzer determines the presence of crystals in each image based on the global quality score for each orientation.
24. The computerized system of claim 16 ,
wherein the phase congruency-related function is the phase congruency function itself.
25. The computerized system of claim 16 ,
wherein the phase congruency-related function is a local energy function.
26. The computerized system of claim 25 ,
further comprising a noise reduction component that applies a noise threshold on the local energy function.
27. The computerized system of claim 16 ,
further comprising producing a skeletonal structure from the detected edges by removing non-maximal values of the edges.
28. The computerized system of claim 16 ,
further comprising a spur removal component for removing spur lines from the detected edges.
29. The computerized system of claim 16 ,
further comprising a smoothing component for smoothing the image prior to edge detection.
30. The computerized system of claim 29 ,
wherein the smoothing component utilizes anisotropic diffusion.
31. A computerized system for detecting macromolecular crystals from images comprising:
means for digitally converting said images into corresponding phase-based digital image data using the Fourier transform;
means for detecting edges from the image data by computing local maxima of a phase congruency-related function associated with each image;
means for dividing the detected edges into discrete line segments;
means for evaluating the geometric relationships that the line segments have with each other to identify any crystal-like qualities; and
means for determining the presence of crystals in an image from said evaluation.
32. A computer program product comprising:
a computer readable medium having a computer readable code embodied therein for causing the detection of macromolecular crystals in images, said computer program product having:
computer readable program code means for causing a computer to detect edges in said images by identifying local maxima of a phase congruency-related function associated with each image;
computer readable program code means for causing said computer to segment the detected edges into discrete line segments;
computer readable program code means for causing said computer to evaluate the geometric relationships that the line segments have with each other to identify any crystal-like qualities; and
computer readable program code means for causing said computer to determine the presence of crystals in each image based on said evaluation.
33. An article of manufacture comprising:
a computer readable medium having a computer readable code means embodied therein for causing the detection of macromolecular crystals in images, said computer readable code means in said article of manufacture comprising:
computer readable program code means for causing a computer to detect edges in said images by identifying local maxima of a phase congruency-related function associated with each image;
computer readable program code means for causing said computer to segment the detected edges into discrete line segments;
computer readable program code means for causing said computer to evaluate the geometric relationships that the line segments have with each other to identify any crystal-like qualities; and
computer readable program code means for causing said computer to determine the presence of crystals in each image based on said evaluation.
34. A program storage device readable by a computer, tangibly embodying a program of instructions executable by the computer to perform method steps for detecting macromolecular crystals from images, said method steps comprising:
detecting edges in said images by identifying local maxima of a phase congruency-related function associated with each image;
segmenting the detected edges into discrete line segments;
evaluating the geometric relationships that the line segments have with each other to identify any crystal-like qualities; and
determining the presence of crystals in each image based on said evaluation.