Automated method and system for the differentiation of bone disease on radiographic images

• US 20030133601A1
• Filed: 11/22/2002
• Published: 07/17/2003
• Est. Priority Date: 11/23/2001
• Status: Abandoned Application

• Current Assignee: University of Chicago The (University of Chicago)
• Original Assignee: University of Chicago
• Inventors: Chinander, Michael R., Favus, Murray, Vokes, Tamara, Giger, Maryellen L.
• Application Number:
  US10/301,836
• Publication Number:
  US 20030133601A1
• Time in Patent Office: Days
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/30008 Bone
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A method, system, and computer program product for analyzing a medical image to determine a measure of bone strength, comprising identifying plural regions of interest (ROIs) in the medical image; calculating at least one texture feature value for each ROI; averaging the at least one texture feature value calculated for each ROI to obtain at least one average texture feature value; and determining the measure of bone strength based on the at least one average texture feature value using a classifier. Alternatively, the image data in each ROI is first transformed into the frequency domain and averaged to obtain an average image. This process reduces noise and improves the performance of the system. The assessment of bone strength and/or osteoporosis is used as a predictor of risk of fracture.

Claims

1. A method of analyzing a medical image to determine a measure of bone strength, comprising:

identifying plural regions of interest (ROIs) in the medical image;

calculating at least one texture feature value for each ROI;

averaging the at least one texture feature value calculated for each ROI to obtain at least one average texture feature value; and

determining the measure of bone strength based on the at least one average texture feature value.

2. The method of claim 1, wherein the calculating step comprises:

calculating as the at least one texture feature value, at least one of a root-mean-square value, a first moment of a power spectrum value, and a Minkowski dimension.

3. The method of claim 1, wherein the determining step comprises:

determining the measure of bone strength by merging the at least one texture feature with feature-related data using a classifier,

said feature-related data including at least one of bone geometry, bone structure, bone mass data, and clinical data.

4. The method of claim 3, wherein the determining step comprises:

determining the measure of bone strength by merging the at least one texture feature with feature-related data using at least one of an artificial neural network and a linear discriminant.

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 any one of claims 1-19.

21. A system configured to analyze a medical image by performing the steps recited in any one of claims 1-19.

5. A method of analyzing a medical image to determine a measure of bone strength, comprising:

identifying plural regions of interest (ROIs) in the medical image;

transforming image data in each of said ROIs into respective frequency domain image data;

averaging the respective frequency domain image data to obtain average image data;

calculating at least one texture feature value from the average image data; and

determining the measure of bone strength based on the at least one texture feature value.

6. The method of claim 5, wherein the transforming step comprises:

transforming image data in each of said ROIs into the respective frequency domain image data using a two-dimensional Fourier transform.

7. The method of claim 5, wherein the calculating step comprises:

calculating as the at least one texture feature value, at least one of a root-mean-square value, a first moment of a power spectrum value, and a Minkowski dimension.

8. The method of claim 5, wherein the determining step comprises:

determining the measure of bone strength by merging the at least one texture feature with feature-related data using a classifier,

said feature-related data including at least one of bone geometry, bone structure, bone mass data, and clinical data.

9. The method of claim 8, wherein the determining step comprises:

determining the measure of bone strength by merging the at least one texture feature with feature-related data using at least one of an artificial neural network and a linear discriminant.

10. A method of analyzing plural medical images to determine a measure of bone strength, comprising:

identifying a region of interest (ROI) having a corresponding center pixel in each medical image;

transforming image data in the ROI of each medical image into respective frequency domain image data;

averaging the respective frequency domain image data to obtain average image data;

calculating at least one texture feature value from the average image data; and

determining the measure of bone strength based on the at least one texture feature value.

11. The method of claim 10, wherein the transforming step comprises:

transforming image data in each of said ROIs into the respective frequency domain image data using a two-dimensional Fourier transform.

12. The method of claim 10, wherein the calculating step comprises:

calculating as the at least one texture feature value, at least one of a root mean square value, a first moment of a power spectrum value, and a Minkowski dimension.

13. The method of claim 10, wherein the determining step comprises:

determining the measure of bone strength by merging the at least one texture feature with feature-related data using a classifier,

said feature-related data including at least one of bone geometry, bone structure, bone mass data, and clinical data.

14. The method of claim 13, wherein the determining step comprises:

determining the measure of bone strength by merging the at least one texture feature with feature-related data using at least one of an artificial neural network and a linear discriminant.

15. The method of claim 10, further comprising:

repeating the identifying, transforming, averaging, and calculating steps for a plurality of ROIs having a corresponding plurality of center pixels;

associating the at least one feature value calculated in each calculating step with a center pixel in the corresponding plurality of center pixels to form at least one texture feature image.

16. The method of claim 15, further comprising:

displaying each of the at least one texture feature image as a color image on a display unit.

17. A method of analyzing a medical image to determine a measure of bone strength, comprising:

identifying plural regions of interest (ROIs) in the medical image, each ROI having a corresponding center pixel;

transforming image data in each of said ROIs into respective frequency domain image data;

calculating at least one texture feature value for each ROI using the respective frequency domain image data; and

determining the measure of bone strength based on the at least one texture feature value.

18. The method of claim 17, further comprising:

repeating the identifying, transforming, and calculating steps for a plurality of ROIs having a corresponding plurality of center pixels; and

associating the at least one feature value calculated for each ROI with the corresponding center pixel to form at least one texture feature image.

19. A method of analyzing plural medical images to form at least one texture feature image, comprising:

identifying a region of interest (ROI) having a corresponding center pixel in each medical image;

calculating at least one texture feature value for the ROI in each medical image;

averaging the at least one texture feature value of each medical image in the plural medical images;

repeating the identifying, calculating, and averaging steps for a plurality of ROIs having a corresponding plurality of center pixels;

associating the at least one feature value calculated in each calculating step with a center pixel in the corresponding plurality of center pixels to form the at least one texture feature image.