Systems and methods for automated screening and prognosis of cancer from whole-slide biopsy images
First Claim
1. A method for automatically detecting classifying, and grading cancerous regions of one or more biopsy images comprising:
- performing an image color standardization procedure on the one or more biopsy images to produce one or more color standardized biopsy images;
performing edge detection on the one or more color standardized biopsy images;
generating sets of texture-based feature vectors from the one or more color standardized biopsy images, wherein generating sets of texture-based feature vectors comprises extraction of a set of features from Fourier and wavelet transforms and fractal analysis of the one or more color standardized biopsy images;
training a classifier by using the generated sets of texture-based feature vectors;
classifying the one or more biopsy images according to the Gleason grading system;
using the result of the classification to determine the Gleason score of the one or more biopsy images;
wherein fractal analysis of the one or more color standardized biopsy images comprises;
performing image filtering on the one or more color standardized biopsy images depending on nature of noise in the one or more color standardized biopsy image;
binarizing the one or more color standardized biopsy images to produce binary images of the one or more color standardized biopsy images,calculating fractal dimension of the binary images by using different grid sizes based on a Differential Box Counting (DBC) algorithm; and
fusing resulting fractal dimensions; and
wherein binarizing the one or more color standardized biopsy images comprises;
performing image filtering on the one or more color standardized biopsy images to produce one or more filtered images;
smoothing the one or more filtered images using shape-dependent filters;
calculating gradient vectors in the one or more filtered images using different kernels;
selecting an edge angle in the one or more filtered images;
determining threshold values within a local dynamic range in the one or more filtered images,generating several edge maps in the one or more filtered images, andfusing the generated edge maps together to form one or more binary images of the one or more color standardized biopsy images.
0 Assignments
0 Petitions
Accused Products
Abstract
Systems and methods for detection, grading, scoring and tele-screening of cancerous lesions are described. A complete scheme for automated quantitative analysis and assessment of human and animal tissue images of several types of cancers is presented. Various aspects of the invention are directed to the detection, grading, prediction and staging of prostate cancer on serial sections/slides of prostate core images, or biopsy images. Accordingly, the invention includes a variety of sub-systems, which could be used separately or in conjunction to automatically grade cancerous regions. Each system utilizes a different approach with a different feature set. For instance, in the quantitative analysis, textural-based and morphology-based features may be extracted at image- and (or) object-levels from regions of interest.
-
Citations
3 Claims
-
1. A method for automatically detecting classifying, and grading cancerous regions of one or more biopsy images comprising:
-
performing an image color standardization procedure on the one or more biopsy images to produce one or more color standardized biopsy images; performing edge detection on the one or more color standardized biopsy images; generating sets of texture-based feature vectors from the one or more color standardized biopsy images, wherein generating sets of texture-based feature vectors comprises extraction of a set of features from Fourier and wavelet transforms and fractal analysis of the one or more color standardized biopsy images; training a classifier by using the generated sets of texture-based feature vectors; classifying the one or more biopsy images according to the Gleason grading system; using the result of the classification to determine the Gleason score of the one or more biopsy images; wherein fractal analysis of the one or more color standardized biopsy images comprises; performing image filtering on the one or more color standardized biopsy images depending on nature of noise in the one or more color standardized biopsy image; binarizing the one or more color standardized biopsy images to produce binary images of the one or more color standardized biopsy images, calculating fractal dimension of the binary images by using different grid sizes based on a Differential Box Counting (DBC) algorithm; and fusing resulting fractal dimensions; and wherein binarizing the one or more color standardized biopsy images comprises; performing image filtering on the one or more color standardized biopsy images to produce one or more filtered images; smoothing the one or more filtered images using shape-dependent filters; calculating gradient vectors in the one or more filtered images using different kernels; selecting an edge angle in the one or more filtered images; determining threshold values within a local dynamic range in the one or more filtered images, generating several edge maps in the one or more filtered images, and fusing the generated edge maps together to form one or more binary images of the one or more color standardized biopsy images. - View Dependent Claims (2, 3)
-
Specification