| CPC A61B 6/502 (2013.01) [A61B 6/5211 (2013.01); G06T 5/77 (2024.01); G06T 7/0012 (2013.01); G16H 30/40 (2018.01)] | 19 Claims |

|
1. A method of processing a mammogram image, comprising:
receiving a digital mammography image of a breast of a patient;
separating the digital mammography image into three channels, red, green and blue, to obtain three channel images;
removing at least a portion of noise and uneven illumination from each of the three channel images to form a first set of three uniform background images;
separating a luminance channel and a chrominance channel from the first set of three uniform background images with a transfer function to form a second set of three uniform background images;
applying principal component analysis (PCA) to the second set of the three uniform background images to determine an eigenvalue and corresponding eigenvector;
assembling a single grayscale image from a weight of the eigenvalues and corresponding eigenvectors and applying a first subspace projection;
removing at least a portion of a pectoral muscle from the grayscale image with a seed region growth technique to obtain a muscle removed image;
normalizing the muscle removed image using a second order Gaussian Laplacian (LoG) and an oriented diffusion filter to form a processed image; and
identifying cancerous and/or noncancerous regions in the processed image and diagnosing the patient.
|