ALGORITHMS FOR SELECTING MASS DENSITY CANDIDATES FROM DIGITAL MAMMOGRAMS

Abstract

The present invention provides a method for selecting mass density candidates from mammograms for computer-aided lesion detection, review and diagnosis. The method has two steps: a Gaussian difference filter to enhance the intensity and a Canny detector to find potential mass density contours. For circumscribed masses, an additional Hough circle detector is used. This invention makes use of both intensity and morphology information and only processes each image at a single gray-level, so both sensitivity and processing time are improved. The selection algorithm can be also used to select mass candidates from ultrasound images, from 3D tomosynthesis mammography images and from breast MRI images.

Description

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

Not Applicable.

REFERENCE TO SEQUENCE LISTING, A TABLE, OR A COMPUTER PROGRAM LISTING COMPACT DISC APPENDIX

Not Applicable.

BACKGROUND OF THE INVENTION

The present invention relates generally to the field of medical imaging analysis. Particularly, the present invention relates to a method and system for candidates selection of mass density from digital mammography images in conjunction with computer-aided detection, review and diagnosis (CAD) for mammography CAD server and digital mammography workstation.

The U.S. patent Classification Definitions: 382/254 (class 382, Image Analysis, subclass 254 Image Enhancement or Restoration); 382/128 (class 382, Image Analysis, subclass 128 Biomedical applications).

Mass density candidates are the locations on mammograms that are used as initial regions of interest to detect potential breast cancers that present abnormal signs of mass densities or architectural distortions. Most existing candidate selection algorithms are based on the intensity of the images, such as, a combination of the global maximum and local maximum (see reference 1), multi-gray-level thresholding on a subtracted image (see reference 2), peak selection from multiple Fourier band-pass images (see reference 3). Using a limited discrete number of levels or bands to select the mass densities, which have a continuous range of intensity levels and sizes, requires ad hoc adjustment of a large number of parameters. Intensity-based methods also usually perform calculations on multiple images, which results in expensive computation.

BRIEF SUMMARY OF THE INVENTION

This invention makes use of both intensity and morphologic algorithms to process each image at a single gray-level to select the candidates. The detailed algorithm is shown in FIG. 2. Because both intensity and morphological information are used, the selection sensitivity is better than algorithms that use intensity alone. Since each image is processed at only one gray level, the processing time is fast. The typical time to generate around 25 candidates from one mammogram image is less than 500 ms, which is much faster than a comparable band-pass method, which typically takes more than 5 seconds for a single mammogram.

The presented candidate selection algorithm can be also used to select mass candidates from ultrasound images, from 3D tomosynthesis mammography images and from breast MRI images.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

FIG. 1 provides overview of mass candidate selection algorithm.

FIG. 2 details the mass candidate selection algorithm.

FIG. 3 shows an example of the algorithm result.

DETAILED DESCRIPTION OF THE INVENTION

As shown in FIG. 1, the input to the mass candidate selection algorithm is a digital mammogram image, or a breast image from other modality. The image is preprocessed to remove artifacts outside the breast tissue. The image resolution of a digital mammogram is usually between 50 um to 100 um. The image therefore can be down-sampled to a lower resolution, i.e., 300 um, in order to improve processing speed without compromising processing quality. The algorithm to select mass density candidate uses this down-sampled image.

As shown in FIG. 2, the down-sampled image is smoothed along the edge of the skinline. A Gaussian difference filter is applied to the smoothed image (see FIG. 3 the original image and the Gaussian difference filtered image). The first Gaussian filter kernel size is selected as 56; and the second Gaussian filter kernel size is selected as 12. The filtered image is masked by breast mask to remove border artifacts. Next step is to use Canny edge detector to find contours of the candidates. The first threshold of the Canny edge detector is selected as 10; and the second threshold of the Canny edge detector is selected as 600. The thresholds are used for edge linking. The aperture parameter for Sobel operator in the implementation of the Canny detector is 3. Finally the candidates are generated from the Canny contours. Those contours that the size either is smaller (<5 mm) or larger (>50 mm) than the mass density criteria are culled from the final results (see circles overlaid over the cancer in FIG. 3).

Claims

1. A method to select the mass density candidates from mammograms for computer-aided cancer detection, review and diagnosis, which comprises of: down-sample the digital mammogram image to a low resolution;smoothing edge along the skinline;applying the Gaussian difference filter to enhance the intensity;masking the filtered image using breast mask;using Canny detector to find potential mass density contours;generating candidate list from canny contours.

2. The method of claim 1, wherein the Gaussian difference filer has parameters of the first kernel size 56 and the second kernel size 12, respectively.

3. The method of claim 1, wherein the Canny detector has parameters of the first threshold 10 and the second threshold 600; and the aperture size 3.

4. The method of claim 1, wherein the Canny detector for circumscribed masses, an additional Hough circle detector is used.

5. The selection algorithm of claim 1 can be used to select mass candidates from mammography images.

6. The selection algorithm of the claim 1 can also be used to select mass candidates from ultrasound images.

7. The selection algorithm of the claim 1 can also be used to select mass candidates from 3D tomosynthesis mammography images,

8. The selection algorithm of the claim 1 can also be used to select mass candidates from breast MRI images.