Method and apparatus for correction of mammograms for non-uniform breast thickness
View Patent ↗A method and apparatus for correcting for a distortion in a mammogram. The method comprises classifying pixels as either likely fat or likely not-fat. The method further comprises identifying a candidate distortion and calculating a histogram of the likely fat and the likely not-fat pixels at the candidate distortion. The method further comprises evaluating a quality of the candidate distortion based on features of the histograms of pixel values in the fat and dense tissue classes. The distortion may be a two dimensional tilt of the plates or a three dimensional deformation of the plates as in a bend.
1. A computer-implemented method, comprising:
receiving, at one or more computer processing units, first information representative of a mammogram of a breast compressed between two compression plates, the compression plates departing from an ideal parallel configuration by an anomalous tilt;
operating the one or more processing units to generate second information representative of a degree of said tilt based on said first information; and
operating the one or more processing units to transform said first information and said second information into third information, representative of a more ideal version of said mammogram that would have been obtained if said compression plates were closer to said ideal parallel configuration; and
storing the third information in a storage device.
2. The computer-implemented method of claim 1 , wherein said operating the one or more processing units to generate second information comprises:
classifying pixels as either likely fat or likely not-fat; identifying a plurality of candidate tilts;
calculating a histogram of the likely fat and the likely not-fat pixels at each candidate tilt;
evaluating a quality of each candidate tilt based on features of the histograms of pixel values in the fat and the not-fat tissue classes; and
computing said second information based at least in part on the candidate tilt having the highest quality.
3. The computer-implemented method of claim 2 , wherein said evaluating the quality of the candidate tilt is based on a skew and variance of the histograms.
4. The computer-implemented method of claim 3 , wherein evaluating the quality of the candidate tilt comprises calculating a weighted Mahalanobis distance between the likely fat and the likely not-fat distributions.
5. The computer-implemented method of claim 4 , wherein said evaluating the quality of the candidate tilt further comprises giving more weight to the variance of the fat distribution in the weighted Mahalanobis distance calculation.
6. The method of claim 4 , wherein evaluating the quality of the candidate tilt further comprises using the skewness of the fat histogram to give a preference to distributions that are not skewed to the negative side.
7. A tilt correcting system comprising:
an input to receive first information representative of a mammogram of a breast compressed between two compression plates, the compression plates departing from an ideal parallel configuration by an anomalous tilt;
a tilt quality evaluator to generate second information representative of a degree of said tilt based on said first information; and
an image adjustor to generate third information using said first information and said second information, said third information being representative of a more ideal version of said mammogram that would have been obtained if said compression plates were closer to said ideal parallel configuration.
8. The tilt correcting system of claim 7 further comprising:
a pixel classifier to classify pixels as either likely fat or likely not-fat;
a candidate tilt identifier to identify a plurality of candidate tilts;
a histogram creator to calculate a histogram of the likely fat and the likely not-fat pixels at each candidate tilt;
a quality calculator to evaluate a quality of each candidate tilt based on features of the histograms of pixel values in the fat and the not-fat tissue classes; and
a tilt selector to compute said second information based at least in part on the candidate tilt having the highest quality.
9. The tilt correcting system of claim 8 , wherein the tilt quality evaluator for evaluating the quality of the candidate tilt comprises a skewness calculator to evaluate skew of the histograms and a variance calculator to evaluate variance of the histograms.
10. The tilt correcting system of claim 9 , wherein the tilt quality evaluator further comprises a Mahalanobis calculator to calculate a weighted Mahalanobis distance between the likely fat and the likely not-fat distributions.
11. The tilt correcting system of claim 10 , wherein the quality calculator gives more weight to the variance of the fat distribution in the weighted Mahalanobis distance calculation.
12. The method of claim 10 , wherein the quality calculator uses the skewness of the fat histogram to give a preference to distributions that are not skewed to the negative side.
13. A machine readable medium having stored thereon data representing sequences of instructions, which when executed by a computer system, cause said computer system to perform the steps of:
receiving a first information representative of a mammogram of a breast compressed between two compression plates, the compression plates departing from an ideal parallel configuration by an anomalous tilt;
generating a second information representative of a degree of said tilt based on said first information; and
generating third information using said first information and said second information, said third information being representative of a more ideal version of said mammogram that would have been obtained if said compression plates were closer to said ideal parallel configuration.
14. The machine readable medium of claim 13 , wherein the generation of the second information comprises the steps of:
classifying pixels as either likely fat or likely not-fat; identifying a plurality of candidate tilts;
calculating a histogram of the likely fat and the likely not-fat pixels at each candidate tilt;
evaluating a quality of each candidate tilt based on features of the histograms of pixel values in the fat and the not-fat tissue classes; and
computing said second information based at least in part on the candidate tilt having the highest quality.
15. The machine readable medium of claim 14 , wherein the evaluation the quality of the candidate tilt is based on a skew and variance of the histograms.
16. The machine readable medium of claim 15 , wherein evaluating the quality of the candidate tilt comprises calculating a weighted Mahalanobis distance between the likely fat and the likely not-fat distributions.
17. The machine readable medium of claim 16 , wherein the evaluating the quality of the candidate tilt further comprises giving more weight to the variance of the fat distribution in the weighted Mahalanobis distance calculation.
18. The machine readable medium of claim 16 , wherein evaluating the quality of the candidate tilt further comprises using the skewness of the fat histogram to give a preference to distributions that are not skewed to the negative side.