Fast preprocessing algorithms for digital mammography CAD and workstation
A method and apparatus are disclosed for an image preprocessing device that automatically detects chestwall laterality; removes border artifacts; and segments breast tissue and pectoral muscle from digital mammograms. The algorithms in the preprocessing device utilize the computer cache, a vertical Sobel filter and a probabilistic Hough transform to detect curved edges. The preprocessing result, along with a pseudo-modality normalized image, can be used as input to a CAD (computer-aided detection) server or to a mammography image review workstation. In the case of workstation input, the preprocessing results improve the protocol for chestwall-to-chestwall image hanging, and support optimal image contrast display of each segmented region.
1. A method for detecting the laterality of a mammogram image, the method comprising:
receiving a mammogram image;
downsampling the mammogram image;
performing a normalization process on the mammogram image;
applying a probabilistic Hough transform (PHT) to a right half and a left half of the image to generate a number of right PHT lines in the right half of the image and to generate a number of left PHT lines in the left half of the image; and
comparing the number of right PHT lines with the number of left PHT lines to determine a laterality of the mammogram image.
2. The method of claim 1 , wherein performing the normalization process on the mammogram image comprises:
performing a dynamic range normalization on the mammogram image; and
truncating a set of bright pixels of the mammogram image using a maximum dynamic range value, the set of bright pixels being a set of pixels of the mammogram image with values greater than a predetermined threshold.
3. The method of claim 2 , further comprising:
summing a set of pixels of the mammogram image on the left half of the mammogram image to form a left sum and a set of pixels on the right half of the mammogram image to form a right sum; and
comparing the left sum and the right sum to determine the laterality of the mammogram image.
4. The method of claim 3 , further comprising providing the laterality of the mammogram image to a mammography image preprocessing system for use in preparing mammography images for viewing by a user of the system.
5. A method for detecting and determining the positions of borders in a mammogram image, the method comprising:
receiving a mammogram image;
duplicating a left portion, a right portion, a top portion, and a bottom portion of the mammogram image, each portion defined as a percentage of the mammogram image;
applying a Sobel filter with a first kernel to the right and left portions and with a second kernel to the top and bottom portions;
normalizing the filtered left, right, top, and bottom portions;
utilizing a probabilistic Hough transform to detect a left edge, a right edge, a top edge, and a bottom edge in the left, right, top, and bottom portions respectively; and
forming a border for the mammogram image by using the detected left, right, top, and bottom edges.
6. The method of claim 5 , further comprising flipping the mammogram image about a central, vertical axis if the mammogram image has a right chestwall laterality so that the chestwall laterality becomes a left chestwall laterality.
7. The method of claim 5 , wherein the left, right, top, and bottom portions are each defined as 10% of the mammogram image.
8. The method of claim 5 , wherein the first kernel is a vertical 3×3 Sobel kernel and the second kernel is a horizontal 3×3 Sobel kernel.
9. A method for segmenting a pectoral muscle region from a full breast segmentation of a mammogram image, the method comprising:
applying a Gaussian blur to the mammogram image to smooth a pixel profile representing a breast skinline;
applying a Sobel filter to the image;
eroding the image to remove the pixel profile representing the breast skinline;
selecting the upper two-thirds of the image;
utilizing a probabilistic Hough transform to detect a plurality of pectoral muscle edges; and
forming and smoothing a pectoral muscle segmentation border utilizing the plurality of pectoral muscle edges.
10. The method of claim 9 , wherein applying the Sobel filter comprises applying a Sobel filter with a vertical 3×3 Sobel kernel.
11. The method of claim 9 , further comprising intersecting the pectoral muscle segmentation border with a full breast tissue segmentation to form a breast tissue segmentation and a pectoral muscle region segmentation.
12. The method of claim 11 , wherein the full breast tissue segmentation is created by growing lines from a set of all pixels that are located on the chestwall side of the image.