Image region segmentation system and method
View Patent ↗The invention relates to a method of segmenting an image of a structure stored as a set of spatially related data points representing variations in a predetermined parameter, said method comprising the steps of selecting a seed point within the structure to be segmented, assigning to each of the data points a value of connectivity indicative of the confidence that respective areas of the data points are part of the same structure as said seed point, said value of connectivity including a function of the distance of the respective point from said seed point, establishing a threshold value for said level of connectivity and selecting for display data points meeting said threshold value.
1. A method of segmenting a structure from a dataset stored as a set of spatially related data points said method comprising: using processor to perform the steps of:
a) obtaining said dataset from a database;
b) enabling selection of a seed point within a desired structure to be segmented from said dataset;
c) for each data point in said dataset:
i) determining a path from said seed point to said data point;
ii) examining variations in a predetermined characteristic of other data points along said path; and
iii) defining a confidence level representing connectivity from a respective data point to said seed point based on a function of said variations in said predetermined characteristic and a distance from said seed point along said path;
d) determining if additional paths exist between said seed point and said data point and, if so, repeating step c) and selecting a path with a highest confidence level; and
e) segmenting said structure by selecting data points with a required confidence level that said data points selected belong to the same structure as said seed point.
2. A method according to claim 1 wherein the length of said path is combined with said function employing variations of said parameter to obtain said value of connectivity.
3. A method according to claim 2 wherein a value of said predetermined parameter is mapped to a table to assign one of a plurality of values thereto and said evaluation of variation of said parameter is evaluated on the basis of values obtained from said table.
4. A method according to claim 2 wherein values of said predetermined parameter are compared to a predefined range of values and those outside said range modified.
5. A method according to claim 4 wherein said values are modified by reducing said values to zero.
6. A method according to claim 2 wherein an area of said image is selected and values of said predetermined parameter in said area are modified.
7. A method according to claim 6 wherein said values are modified by reducing said values to zero.
8. A method according to claim 2 wherein values of said predetermined characteristic are exceeding said threshold are changed to a common value.
9. A method according to claim 2 wherein values of said predetermined parameters are compared to said seed point and those within predefined limits of that of said seed point are selected for further processing.
10. A method according to claim 1 wherein a weighting factor is applied to said length of said path to vary the effect of said distance and said value of connectivity.
11. A method according to claim 10 wherein said weighting factor is variable.
12. A method according to claim 1 wherein said required confidence establishes a threshold value for said connectivity and is adjustable to vary the data points selected for display.
13. A method according to claim 1 wherein said function employing variations in said parameter is an indication of the maximum variation in said parameter.
14. A method according to claim 1 wherein a plurality of paths are evaluated and limited by application of a volume size value.
15. A method according to claim 1 wherein the number of the selected data points is compared to an anticipated value.
16. A method according to claim 15 wherein a threshold for said required confidence is adjusted to bring said number of selected data points into conformity with said anticipated value.
17. A method according to claim 15 wherein said set of selected data points represents a volume of a structure.
18. A method according to claim 1 wherein said seed point is selected from a set of data points other than those being segmented.
19. A method according to claim 18 wherein said seed point is selected based upon characteristics in said data set indicative of a particular feature.
20. A method according to claim 1 wherein said seed point is selected from examination of a set of data points to identify a characteristic indicative of a particular feature.
21. A computer readable medium comprising computer executable instructions that when executed cause a computing device to:
a) obtain said dataset from a database;
b) enable selection of a seed point within a desired structure to be segmented from said dataset;
c) for each data point in said dataset;
i) determine a path from said seed point to said data point;
ii) examine variations in a predetermined characteristic of other data points along said path; and
iii) define a confidence level representing connectivity from a respective data point to said seed point based on a function of said variations in said predetermined characteristic and a distance from said seed point along said path;
d) determining if additional paths exist between said seed point and said data point and, if so, repeating step c) and select a path with a highest confidence level; and
e) segment said structure by selecting data points with a required confidence level that said data points selected belong to the same structure as said seed point.
22. The computer readable medium according to claim 21 wherein the length of said path is combined with said function employing variations of said parameter to obtain said value of connectivity.
23. The computer readable medium according to claim 22 wherein a value of said predetermined parameter is mapped to a table to assign one of a plurality of values thereto and said evaluation of variation of said parameter is evaluated on the basis of values obtained from said table.
24. The computer readable medium according to claim 22 wherein values of said predetermined parameters are compared to said seed point and those within predefined limits of that of said seed point are selected for further processing.
25. The computer readable medium according to claim 21 wherein a weighting factor is applied to said length of said path to vary the effect of said distance and said value of connectivity.
26. The computer readable medium according to claim 21 wherein said required confidence establishes a threshold value for said connectivity and is adjustable to vary the data points selected for display.
27. The computer readable medium according to claim 21 wherein said seed point is selected from examination of a set of data points to identify a characteristic indicative of a particular feature.
28. A method of establishing a threshold value for segmenting an image of a structure stored as a set of spatially related data points representing variations in a predetermined parameter, said method comprising: using processor to perform the steps of :establishing a first number of data points expected to represent a segmented feature of said image; establishing said threshold value of a characteristic of said data set; determining a second number of data points meeting said threshold; comparing said first and second number of data points; and adjusting said threshold if said first and second numbers do not correlate.
29. A method according to claim 28 wherein said threshold is adjusted until said first and second numbers substantially correspond to each other.
30. A computer readable medium comprising computer executable instructions that when executed cause a computing device to perform the method according to claim 28 .