Claims
- 1. A method of detecting at least a candidate abnormality in a sonographic image, the method comprising:
calculating plural skewness values at respective plural locations in the sonographic image; and determining an area in the sonographic image to be the candidate abnormality, based at least in part on the skewness values.
- 2. The method of claim 1, further comprising:
merging the skewness values with other pixel values determined in accordance with other analytic features, so as to form plural merged pixels; forming a merged image from the plural merged pixels; and comparing the merged values in the merged image to a threshold, so as to arrive at comparison results that are used in the candidate abnormality area determining step.
- 3. The method of claim 2, wherein:
the other analytic features are derived from the sonographic image.
- 4. The method of claim 1, further comprising:
forming a skewness image from the plural skewness values; and comparing the skewness values in the skewness image to a threshold, so as to arrive at comparison results that are used in the candidate abnormality area determining step.
- 5. The method of claim 4, wherein the calculating step comprises:
convoluting the sonographic image with a mask by moving the mask over plural locations in the sonographic image; and calculating the plural skewness values at respective locations of the mask.
- 6. The method of claim 5, wherein the skewness image forming step comprises:
assigning the plural skewness values to respective mask center points.
- 7. The method of claim 4, wherein the candidate abnormality area determining step comprises:
determining a particular skewness value to indicate part of a candidate abnormality when the particular skewness value exceeds the threshold.
- 8. The method of claim 7, further comprising:
calculating a standard deviation of skewness values in the skewness image; and determining the threshold as a mathematical function of the standard deviation.
- 9. The method of claim 8, wherein the threshold determining step comprises:
determining the threshold as being directly proportional to a first power of the standard deviation of the skewness values in the skewness image.
- 10. The method of claim 1, wherein the calculating step comprises:
calculating the skewness values as a mathematical function of a standard deviation of a gray-value distribution of pixels in the sonographic image.
- 11. The method of claim 10, wherein the calculating step comprises calculating the skewness values according to a formula:
- 12. The method of claim 1, further comprising:
superimposing an emphasis symbol on the sonographic image so as to indicate the area that was determined to be the candidate abnormality.
- 13. The method of claim 1, further comprising:
forming a histogram of gray values of pixels in the sonographic image to form a gray value histogram; and adding white noise to the gray value histogram to form a modified gray value histogram that is configured for use in the skewness value calculating step.
- 14. The method of claim 1, further comprising:
repeatedly executing the steps of claim 1 to detect the candidate abnormality based on a sequence of sonographic images in real time.
- 15. An automated method of diagnosing a candidate abnormality in a sonographic image, the method comprising:
determining an area of the candidate abnormality in the sonographic image using the candidate abnormality detecting method of any of claims 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14; calculating an abnormality skewness value of the area that was determined to be the candidate abnormality; and determining a likelihood of malignancy of the candidate abnormality based at least in part on the abnormality skewness value.
- 16. The method of claim 15, wherein the likelihood determining step comprises:
comparing the abnormality skewness value to a threshold; and determining the candidate abnormality to be malignant if the abnormality skewness value exceeds the threshold, and to be benign if the abnormality threshold exceeds the abnormality skewness value.
- 17. A system implementing the method of claim 16.
- 18. A computer program product storing program instructions for execution on a computer system, which when executed by the computer system, cause the computer system to perform the method recited in claim 16.
- 19. A system implementing the method of claim 15.
- 20. A computer program product storing program instructions for execution on a computer system, which when executed by the computer system, cause the computer system to perform the method recited in claim 15.
- 21. A method of diagnosing a designated candidate abnormality in an area of a sonographic image, the method comprising:
calculating an abnormality skewness value of the area; and determining a likelihood of malignancy of the candidate abnormality based at least in part on the abnormality skewness value.
- 22. The method of claim 21, wherein the likelihood determining step comprises:
comparing the abnormality skewness value to a threshold; and determining the candidate abnormality to be malignant if the abnormality skewness value exceeds the threshold, and to be benign if the abnormality threshold exceeds the abnormality skewness value.
- 23. A system implementing the method of any of claims 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 21 or 22.
- 24. A computer program product storing program instructions for execution on a computer system, which when executed by the computer system, cause the computer system to perform the method of any of claims 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 21 or 22.
Government Interests
[0001] The present invention was made in part with U.S. Government support under grant number CA89452 and CA09649 from the USPHS, and U.S. Army Medical Research and Materiel Command grant number 97-2445. The U.S. Government may have certain rights to this invention.