Method and system to enhance robust identification of abnormal regions in radiographs

• US 5,838,815 A
• Filed: 02/22/1996
• Issued: 11/17/1998
• Est. Priority Date: 12/01/1994
• Status: Expired due to Fees

• Current Assignee: University Of Pittsburgh (University of Pittsburgh of The Commonwealth System of Higher Education)
• Original Assignee: University Of Pittsburgh (University of Pittsburgh of The Commonwealth System of Higher Education)
• Inventors: Chang, Yuan-Hsiang, Zheng, Bin, Gur, David
• Primary Examiner(s): Couso, Jose L.
• Application Number:
  US08/605,754
• Time in Patent Office: 999 Days
• Field of Search: 382/128, 382/130, 382/132, 382/170, 382/203, 346/413.13, 346/413.17, 346/413.19, 346/413.22, 346/413.23, 378/99, 358/111, 128/634, 128/754
• US Class Current: 382/128
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts

Abstract

A method and apparatus for detecting abnormal regions in living tissue depicted in a digital radiograph includes identifying suspected abnormal regions depicted in the radiograph and then, for each identified suspected abnormal region, extracting multiple topographic layers of the region from the digital radiograph; determining features of the region in each of the layers; and applying inter-layer multivariate non-linear criteria to the features to determine whether the suspected abnormal region is to be classified as an abnormal region. The distribution of digital values in the digitized radiograph is modified and a probabalistic determination is made as to whether suspected abnormal regions are actually abnormal from multiple analyses. An image is processed using at least two partially correlated detection schemes to produce at least two corresponding result sets of potentially abnormal regions; and the results of the detection schemes are combined using logical "AND" or logical "OR" combining rules to detect abnormal regions in the radiograph.

Claims

1. A method of detecting an abnormal region in living tissue depicted in a digital radiograph, the method comprising:

identifying a suspected abnormal region depicted in the radiograph;

extracting multiple topographic layers of the suspected abnormal region from the digital radiograph, the multiple topographic layers being extracted from the same radiograph;

determining features of the region in each of said layers; and

applying an inter-layer multivariate criterion to said features to determine whether said suspected abnormal region is an abnormal region.

2. A method as in claim 1, wherein said criterion is non-linear.

3. A method as in claim 1, wherein said features of the region include at least one of:

the size of the region, the shape of the region, the longest axis of the region;

the contrast of the region;

the standard deviation of the region;

the skewness of the region;

the size growth ratio of the region; and

the central position shift of the region.

4. A method as in claim 3, wherein said criterion relates to at least one of:

(a) the contrast and the standard deviation in a third layer;

(b) the shape factor and longest axis in the third layer;

(c) the size and the contrast in a first layer;

(d) the growth ratio between the first layer and a second layer compared to the growth ratio between the second and third layers;

(e) the shape factor in the second layer and the central position shift between the first and second layers;

(f) the contrast in the first layer and central position shift between the first and second layer;

(g) the growth ratio between the second and third layers compared to the skewness in the second layer;

(h) the growth ratio between the second and third layers compared to the standard deviation in the first layer;

(i) the shape factor in the second layer compared to the skewness in the third layer;

(j) the shape factor in the second layer compared to the shape factor in the third layer;

(k) the standard deviation in the first layer compared to the longest axis in the second layer; and

(l) the contrast in the first layer compared to the skewness in the second layer.

5. A method as in claim 2, wherein said features of the region include:

the size of the region, the shape of the region, the longest axis of the region;

the contrast of the region;

the standard deviation of the region;

the skewness of the region;

the size growth ratio of the region; and

the central position shift of the region.

6. A method as in claim 5, wherein said criterion relates to:

(a) the contrast and the standard deviation in a third layer;

(b) the shape factor and longest axis in the third layer;

(c) the size and the contrast in a first layer;

(d) the growth ratio between the first layer and a second layer compared to the growth ratio between the second and third layers;

(e) the shape factor in the second layer and the central position shift between the first and second layers;

(f) the contrast in the first layer and central position shift between the first and second layer;

(g) the growth ratio between the second and third layers compared to the skewness in the second layer;

(h) the growth ratio between the second and third layers compared to the standard deviation in the first layer;

(i) the shape factor in the second layer compared to the skewness in the third layer;

(j) the shape factor in the second layer compared to the shape factor in the third layer;

(k) the standard deviation in the first layer compared to the longest axis in the second layer; and

(l) the contrast in the first layer compared to the skewness in the second layer.

7. A method as in claim 3, wherein said criterion relates to at least one of:

the contrast and the standard deviation in a third layer;

the shape factor in a second layer compared to the skewness in the third layer; and

the contrast in a first layer compared to the skewness in the second layer.

8. A method as in claim 1, wherein said criterion relates to at least one of:

contrast in one of said layers and central position shift between said one of said layers and another of said layers;

growth ratio between two of said layers compared to skewness in one of said two of said layers;

growth ratio between two of said layers compared to standard deviation in one of said layers;

shape factor in one of said layers compared to skewness in another of said layers;

shape factor in one of said layers compared to shape factor in another of said layers;

standard deviation in one of said layers compared to longest axis in another of said layers; and

contrast in one of said layers compared to skewness another of said layers.

15. A method as in claim 1 wherein the radiograph is a mammogram and wherein the abnormal region depicts a mass or clustered microcalcifications.

18. A method as in claim 1 wherein the radiograph is produced from computed tomography of the liver and wherein the abnormal region depicts a mass.

21. A method as in claim 1 wherein the radiograph is a chest x-ray and wherein the abnormal region depicts a nodule.

9. A method of detecting abnormal regions in living tissue depicted in a digital radiograph, the method comprising:

identifying suspected abnormal regions depicted in the radiograph;

for each identified suspected abnormal region,

(a) extracting multiple topographic layers of the region from the digital radiograph, the multiple topographic layers being extracted from the same radiograph;

(b) determining features of the region in each of said layers; and

(c) applying inter-layer multivariate criteria to said features to determine whether said suspected abnormal region is an abnormal region.

16. A method as in claim 9 wherein the radiograph is a mammogram and wherein the abnormal regions depict masses or clustered microcalcifications.

19. A method as in claim 9 wherein the radiograph is produced from computed tomography of the liver and wherein the abnormal regions depict masses.

22. A method as in claim 9 wherein the radiograph is a chest x-ray and wherein the abnormal regions depict nodules.

10. An apparatus for detecting abnormal regions in living tissue depicted in a digital radiograph, the apparatus comprising:

means for identifying suspected abnormal regions depicted in the radiograph;

means for extracting multiple topographic layers of the region from the digital radiograph for each suspected abnormal region identified by the means for identifying, the multiple topographic layers all being extracted from the same radiograph;

means for determining features in each of said layers extracted by the means for extracting; and

means for applying inter-layer multivariate criteria to said features to determine whether said suspected abnormal region is an abnormal region.

11. An apparatus as in claim 10, wherein said means for applying inter-layer multivariate criteria comprises means for applying non-linear criteria.

12. An apparatus as in claim 10, wherein said features of the region include at least one of:

the size of the region, the shape of the region, the longest axis of the region;

the contrast of the region;

the standard deviation of the region;

the skewness of the region;

the size growth ratio of the region; and

the central position shift of the region.

13. An apparatus as in claim 10, wherein said criteria relate to at least one of:

(a) the contrast and the standard deviation in a third layer;

(b) the shape factor and longest axis in the third layer;

(c) the size and the contrast in a first layer;

(d) the growth ratio between the first layer and a second layer compared to the growth ratio between the second and third layers;

(e) the shape factor in the second layer and the central position shift between first and second layer;

(f) the contrast in the first layer and central position shift between the first and second layer;

(g) the growth ratio between the second and third layers compared to the skewness in the second layer;

(h) the growth ratio between the second and third layers compared to standard deviation in the first layer;

(i) the shape factor in the second layer compared to the skewness in the third layer;

(j) the shape factor in the second layer compared to the shape factor in the third layer;

(k) the standard deviation in the first layer compared to the longest axis in the second layer; and

(l) the contrast in the first layer compared to the skewness in the second layer.

14. An apparatus as in claim 10, wherein said criteria relate to at least one of:

contrast in one of said layers and central position shift between said one of said layers and another of said layers;

growth ratio between two of said layers compared to skewness in one of said two of said layers;

growth ratio between two of said layers compared to standard deviation in one of said layers;

shape factor in one of said layers compared to skewness in another of said layers;

shape factor in one of said layers compared to shape factor in another of said layers;

standard deviation in one of said layers compared to longest axis in another of said layers; and

contrast in one of said layers compared to skewness another of said layers.

17. An apparatus as in claim 10 wherein the radiograph is a mammogram and wherein the abnormal regions depict masses or clustered microcalcifications.

20. An apparatus as in claim 10 wherein the radiograph is produced from computed tomography of the liver and wherein the abnormal regions depict masses.

23. An apparatus as in claim 10 wherein the radiograph is a chest x-ray and wherein the abnormal regions depict nodules.