Back to results
US 10,096,110 B2Granted
System and method for automated stereology of cancer
AssigneesUniversity of South Florida, Stereology Resource Center, Inc.
InventorsPeter Randolph Mouton, Dmitry Goldgof, Lawrence O. Hall, Baishali Chaudhury
Life of this patent
Estimating the term…
Abstract
This invention relates to a system and method for applying an ensemble of segmentations to microscopy images of a tissue sample to determine if the tissue sample is representative of cancerous tissue. The ensemble of segmentations is applied to a plurality of greyscale or color microscopy images to generate a final image level segmentation and a final blob level segmentation. The final image level segmentation and final blob level segmentation are used to calculate a mean nuclear volume to determine if the tissue sample is representative of cancerous tissue.
Claims20
A method for determining if a tissue sample is representative of cancerous tissue, the method comprising:
- applying an ensemble of segmentations to a plurality of greyscale images to generate a set of segmented images for each of the plurality of greyscale images;
- determining if a segmented image of the set of segmented images comprises a blob that is larger than a predetermined maximum blob size, and rejecting the set of segmented images if a segmented image of the set of segmented images comprises a blob that is larger than a predetermined maximum blob size;
- determining if at least half of the segmented images of the set of segmented images are similar, rejecting the set of segmented images if at least half of the segmented images of the set of segmented images are not similar and accepting the set of segmented images if at least half of the segmented images of the set of segmented images are similar to generate a set of accepted image level segmented images;
- determining if a blob is present in at least half of the segmented images of the set of segmented images, rejecting the set of segmented images if the blob is not present in at least half of the segmented images of the set of segmented images and accepting the set of segmented images if the blob is present in at least half of the segmented images of the set of segmented images to generate a set of accepted blob level segmented images;
- combining the segmented images of the set of accepted image level segmented images using a consensus function to generate a final image level segmentation;
- combining the segmented images of the set of accepted blob level segmented images using a consensus function to generate a final blob level segmentation;
- and
- calculating the mean nuclear volume of the final image level segmentation and the final blob level segmentation to determine if the tissue sample is representative of cancerous tissue.
5 more elements
A method for determining if a tissue sample is representative of cancerous tissue, the method comprising:
- receiving a plurality of color microscopy images of at least one stained tissue sample;
- converting the plurality of color microscopy images to a plurality of greyscale microscopy images;
- performing screening of each of the plurality of greyscale images to determine if the pixel intensity of each of the plurality of greyscale images is acceptable and rejecting the greyscale images if the pixel intensity of the greyscale image is not acceptable;
- applying an ensemble of segmentations to a plurality of greyscale images to generate a set of segmented images for each of the plurality of greyscale images;
- determining if a segmented image of the set of segmented images comprises a blob that is larger than a predetermined maximum blob size, and rejecting the set of segmented images if a segmented image of the set of segmented images comprises a blob that is larger than a predetermined maximum blob size;
- determining if at least half of the segmented images of the set of segmented images are similar, rejecting the set of segmented images if at least half of the segmented images of the set of segmented images are not similar and accepting the set of segmented images if at least half of the segmented images of the set of segmented images are similar to generate a set of accepted image level segmented images;
- determining if a blob is present in at least half of the segmented images of the set of segmented images, rejecting the set of segmented images if the blob is not present in at least half of the segmented images of the set of segmented images and accepting the set of segmented images if the blob is present in at least half of the segmented images of the set of segmented images to generate a set of accepted blob level segmented images;
- combining the segmented images of the set of accepted image level segmented images using a consensus function to generate a final image level segmentation;
- combining the segmented images of the set of accepted blob level segmented images using a consensus function to generate a final blob level segmentation;
- and
- calculating the mean nuclear volume of the final image level segmentation and the final blob level segmentation to determine if the tissue sample is representative of cancerous tissue.
8 more elements
One or more non-transitory computer-readable media having computer-executable instructions for performing a method of running a software program on a computing device, the computing device operating under an operating system, the method including issuing instructions from the software program comprising:
- applying an ensemble of segmentations to a plurality of greyscale images to generate a set of segmented images for each of the plurality of greyscale images;
- determining if a segmented image of the set of segmented images comprises a blob that is larger than a predetermined maximum blob size, and rejecting the set of segmented images if a segmented image of the set of segmented images comprises a blob that is larger than a predetermined maximum blob size;
- determining if at least half of the segmented images of the set of segmented images are similar, rejecting the set of segmented images if at least half of the segmented images of the set of segmented images are not similar and accepting the set of segmented images if at least half of the segmented images of the set of segmented images are similar to generate a set of accepted image level segmented images;
- determining if a blob is present in at least half of the segmented images of the set of segmented images, rejecting the set of segmented images if the blob is not present in at least half of the segmented images of the set of segmented images and accepting the set of segmented images if the blob is present in at least half of the segmented images of the set of segmented images to generate a set of accepted blob level segmented images;
- combining the segmented images of the set of accepted image level segmented images using a consensus function to generate a final image level segmentation;
- combining the segmented images of the set of accepted blob level segmented images using a consensus function to generate a final blob level segmentation;
- and
- calculating the mean nuclear volume of the final image level segmentation and the final blob level segmentation to determine if the tissue sample is representative of cancerous tissue.
5 more elements
Dependencies are inferred from the claim text.