PACS viewer and a method for identifying patient orientation
View Patent ↗A PACS viewer and an associated method and computer program product are provided for identifying patient orientation. With respect to a PACS viewer, the PACS viewer includes processing circuitry configured to register an image onto a template. The processing circuitry is also configured to determine a representative seed point within each of the left and right lungs as represented by the image. The processing circuitry is further configured to apply an active contour model to each of the left and right lungs to generate a binary image of the left and right lungs. In this regard, the processing circuitry is configured to apply the active contour model by initializing the active contour model with a mask constructed using the representative seed points. Further, the processing circuitry is configured to detect at least one feature from the binary image to permit identification of the left and right lungs.
1. A Picture Archiving and Communications System (PACS) viewer comprising processing circuitry configured to:
register an image onto a template;
determine a representative seed point within each of left and right lungs as represented by the image;
apply an active contour model to each of the left and right lungs to generate a binary image of the left and right lungs, wherein the processing circuitry is configured to apply the active contour model by initializing the active contour model with a mask constructed using the representative seed points; and
detect at least one feature from the binary image to permit identification of the left and right lungs.
2. A PACS viewer according to claim 1 wherein the processing circuitry is configured to detect at least one feature from the binary image by:
determining an area of the left and right lungs from the binary image; and
identifying the right lung to have the larger area.
3. A PACS viewer according to claim 1 wherein the processing circuitry is configured to detect at least one feature from the binary image by:
defining a line between the left and right lungs as represented by the binary image, wherein the line is located based upon a diaphragm of one of the lungs;
dividing the line into first and second line segments extending between a midline between the left and right lungs and a respective lung;
comparing a length of each of the first and second line segments; and
identifying the left lung to be on the same side of the midline as the longer of the first and second line segments.
4. A PACS viewer according to claim 1 wherein the processing circuitry is configured to detect at least one feature from the binary image by:
making a first comparison of the binary image to the template and making a second comparison of a mirrored version of one of the binary image or the template to the other one of the binary image or the template, wherein the template represents a predefined orientation with respect to the left and right lungs;
determining which of the first comparison or the second comparison is indicative of more similarity; and
identifying the left and right lungs within the binary image based upon the predefined orientation of the template and the one of the first or second comparison that is indicative of more similarity.
5. A PACS viewer according to claim 1 wherein the processing circuitry is configured to determine a representative seed point by:
defining a plurality of lines extending across the image;
within each of the left and right lungs, determining a seed point on each of the plurality of lines; and
determining the representative seed point within each of the left and right lungs from among the seed points determined on each of the plurality of lines within each of the left and right lungs.
6. A PACS viewer according to claim 1 wherein the processing circuitry is further configured to subject the image to anisotropic diffusion prior to applying the active contour model.
7. A PACS viewer according to claim 1 wherein the processing circuitry is further configured to predict patient orientation based upon the identification of the left and right lungs utilizing a support vector machine classifier.
8. A PACS viewer according to claim 7 wherein the processing circuitry is further configured to:
compare the patient orientation that is predicted to an indication of orientation that is associated with the image; and
cause a notification to be provided in an instance in which the patient orientation that is predicted is different than the indication of orientation that is associated with the image.
9. A method for identifying patient orientation, the method comprising:
registering an image onto a template;
determining a representative seed point within each of left and right lungs as represented by the image;
applying an active contour model to each of the left and right lungs to generate a binary image of the left and right lungs, wherein applying the active contour model comprises initializing the active contour model with a mask constructed using the representative seed points; and
detecting at least one feature from the binary image to permit identification of the left and right lungs.
10. A method according to claim 9 wherein detecting at least one feature from the binary image comprises:
determining an area of the left and right lungs from the binary image; and
identifying the right lung to have the larger area.
11. A method according to claim 9 wherein detecting at least one feature from the binary image comprises:
defining a line between the left and right lungs as represented by the binary image, wherein the line is located based upon a diaphragm of one of the lungs;
dividing the line into first and second line segments extending between a midline between the left and right lungs and a respective lung;
comparing a length of each of the first and second line segments; and
identifying the left lung to be on the same side of the midline as the longer of the first and second line segments.
12. A method according to claim 9 wherein detecting at least one feature from the binary image comprises:
making a first comparison of the binary image to the template and making a second comparison of a mirrored version of one of the binary image or the template to the other one of the binary image or the template, wherein the template represents a predefined orientation with respect to the left and right lungs;
determining which of the first comparison or the second comparison is indicative of more similarity; and
identifying the left and right lungs within the binary image based upon the predefined orientation of the template and the one of the first or second comparison that is indicative of more similarity.
13. A method according to claim 9 wherein determining a representative seed point comprises:
defining a plurality of lines extending across the image;
within each of the left and right lungs, determining a seed point on each of the plurality of lines; and
determining the representative seed point within each of the left and right lungs from among the seed points determined on each of the plurality of lines within each of the left and right lungs.
14. A method according to claim 9 further comprising subjecting the image to anisotropic diffusion prior to applying the active contour model.
15. A method according to claim 9 further comprising predicting patient orientation based upon the identification of the left and right lungs utilizing a support vector machine classifier.
16. A method according to claim 15 further comprising:
comparing the patient orientation that is predicted to an indication of orientation that is associated with the image; and
causing a notification to be provided in an instance in which the patient orientation that is predicted is different than the indication of orientation that is associated with the image.
17. A computer program product for identifying patient orientation, the computer program product comprising a non-transitory computer readable storage medium having program code portions stored thereon, the program code portions configured, upon execution, to:
register an image onto a template;
determine a representative seed point within each of left and right lungs as represented by the image;
apply an active contour model to each of the left and right lungs to generate a binary image of the left and right lungs, wherein applying the active contour model comprises initializing the active contour model with a mask constructed using the representative seed points; and
detect at least one feature from the binary image to permit identification of the left and right lungs.
18. A computer program product according to claim 17 wherein the program code portions configured to detect at least one feature from the binary image comprise program code portions configured to:
determine an area of the left and right lungs from the binary image; and
identify the right lung to have the larger area.
19. A computer program product according to claim 17 wherein the program code portions configured to detect at least one feature from the binary image comprise program code portions configured to:
define a line between the left and right lungs as represented by the binary image, wherein the line is located based upon a diaphragm of one of the lungs;
divide the line into first and second line segments extending between a midline between the left and right lungs and a respective lung;
compare a length of each of the first and second line segments; and
identify the left lung to be on the same side of the midline as the longer of the first and second line segments.
20. A computer program product according to claim 17 wherein the program code portions configured to detect at least one feature from the binary image comprise program code portions configured to:
make a first comparison of the binary image to the template and making a second comparison of a mirrored version of one of the binary image or the template to the other one of the binary image or the template, wherein the template represents a predefined orientation with respect to the left and right lungs;
determine which of the first comparison or the second comparison is indicative of more similarity; and
identify the left and right lungs within the binary image based upon the predefined orientation of the template and the one of the first or second comparison that is indicative of more similarity.