Method and system for wavelet based detection of colon polyps

• US 20050105801A1
• Filed: 08/10/2004
• Published: 05/19/2005
• Est. Priority Date: 08/13/2003
• Status: Active Grant

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Periaswamy, Senthil
• Granted Patent: US 7,397,936 B2
• Time in Patent Office: Days
• US Class Current: 382/192
• CPC Class Codes: G06T 2207/30028 Colon; Small intestine
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A method of identifying colon polyps in a digital volume, wherein the volume includes a plurality of values corresponding to a domain of points in a 3D space, is provided. The method includes selecting a mother wavelet scaling function that corresponds to a polyp; performing a forward wavelet transformation on the volume to obtain a set of wavelet coefficients, wherein each wavelet coefficient is associated with a length scale; determining, for each length scale, a transformation magnitude; and forming, for each length scale, a transformed set of wavelet coefficients associated with the length scale. An inverse wavelet transform is performed on the transformed wavelet coefficients for each length scale to obtain a reconstructed volume, and the reconstructed volume is analyzed for the existence of polyps.

Claims

1. A method of identifying colon polyps in a digital volume, wherein said volume comprises a plurality of values corresponding to a domain of points in a 3D space, said method comprising the steps of:

selecting a mother wavelet scaling function that corresponds to a polyp;

performing a forward wavelet transformation on said volume to obtain a set of wavelet coefficients, wherein each wavelet coefficient is associated with a length scale;

determining, for each length scale, a transformation magnitude; and

forming, for each length scale, a transformed set of wavelet coefficients associated with said length scale.

2. The method of claim 1, further comprising the steps of performing, for each length scale, an inverse wavelet transform on the transformed wavelet coefficients corresponding to said length scale, to obtain a reconstructed volume;

and

analyzing said reconstructed volume for the existence of polyps.

3. The method of claim 2, wherein the transformation magnitude comprises a multiplier magnitude, wherein the step of forming a transformed set of wavelet coefficients comprises multiplying, for each length scale, the wavelet coefficients by the multiplier for that length scale to from a multiplied coefficient set for said length scale and wherein said step of performing an inverse wavelet transform is performed on said multiplied coefficient set for said length scale.

4. The method of claim 2, wherein the transformation magnitude comprises a threshold magnitude, wherein the step of forming a transformed set of wavelet coefficients comprises selecting, for each length scale, those wavelet coefficients whose magnitudes exceeds the threshold for said length scale, to form a subset of wavelet coefficients associated with said length scale, and wherein said step of performing an inverse wavelet transform is performed on said coefficient subset for said length scale.

5. The method of claim 1, wherein the transformation magnitude comprises a threshold magnitude, wherein the step of forming a transformed set of wavelet coefficients comprises selecting, for each length scale, those wavelet coefficients whose magnitudes exceeds the threshold for said length scale, to form a template subset of wavelet coefficients associated with said length scale, and further comprising the steps of:

performing a forward wavelet transform of a second volume, to obtain a second set of wavelet coefficients;

comparing, for each length scale, coefficients of the second set of wavelet coefficients with coefficients of a template subset for said length scale, wherein a match between said coefficients of said template subset and coefficients of said second set of wavelet coefficients indicates the presence of a polyp corresponding to said length scale.

6. The method of claim 3, further comprising, prior to performing the forward wavelet transform, calculating a first derivative of said volume, and selecting the mother wavelet function to be similar to a first derivative of a polyp.

7. The method of claim 4, further comprising, prior to performing the forward wavelet transform, calculating a first derivative of said volume, and selecting the mother wavelet function to be similar to a first derivative of a polyp.

8. The method of claim 5, further comprising, prior to performing the forward wavelet transform, calculating a first derivative of said volume, and selecting the mother wavelet function to be similar to a first derivative of a polyp.

15. A program storage device readable by a computer, tangibly embodying a program of instructions executable by the computer to perform the method steps of claim 1 for identifying colon polyps in a digital volume, wherein said volume comprises a plurality of values corresponding to a domain of points in a 3D space.

16. The program storage device of claim 15, wherein said program of instructions executable by the computer performs the method steps of claim 2.

17. The program storage device of claim 16, wherein said program of instructions executable by the computer performs the method steps of claim 3.

18. The program storage device of claim 16, wherein said program of instructions executable by the computer performs the method steps of claim 4.

19. The program storage device of claim 15, wherein said program of instructions executable by the computer performs the method steps of claim 5.

20. The program storage device of claim 17, wherein said program of instructions executable by the computer performs the method steps of claim 6.

21. The program storage device of claim 18, wherein said program of instructions executable by the computer performs the method steps of claim 7.

22. The program storage device of claim 19, wherein said program of instructions executable by the computer performs the method steps of claim 8.

9. A method of identifying colon polyps in a digital volume, wherein said volume comprises a plurality of values corresponding to a domain of points in a 3D space, said method comprising the steps of:

selecting a mother wavelet scaling function that corresponds to a polyp;

performing a forward wavelet transformation on said volume to obtain a set of wavelet coefficients, wherein each wavelet coefficient is associated with a length scale;

determining, for each length scale, a multiplier magnitude;

multiplying, for each length scale, the wavelet coefficients by the multiplier for that length scale to form a multiplied coefficient set for said length scale;

performing, for each length scale, an inverse wavelet transform on the multiplied coefficient set corresponding to said length scale, to obtain a reconstructed volume; and

analyzing said reconstructed volume for the existence of polyps.

10. The method of claim 9, further comprising, prior to performing the forward wavelet transform, calculating a first derivative of said volume, and selecting the mother wavelet function to be similar to a first derivative of a polyp.

11. A method of identifying colon polyps in a digital volume, wherein said volume comprises a plurality of values corresponding to a domain of points in a 3D space, said method comprising the steps of:

selecting a mother wavelet scaling function that corresponds to a polyp;

performing a forward wavelet transformation on said volume to obtain a set of wavelet coefficients, wherein each wavelet coefficient is associated with a length scale;

determining, for each length scale, a threshold magnitude;

selecting, for each length scale, those wavelet coefficients whose magnitudes exceeds the threshold for said length scale, to form a subset of wavelet coefficients associated with said length scale;

performing, for each length scale, an inverse wavelet transform on the coefficient subset corresponding to said length scale, to obtain a reconstructed volume; and

analyzing said reconstructed volume for the existence of polyps.

12. The method of claim 11, further comprising, prior to performing the forward wavelet transform, calculating a first derivative of said volume, and selecting the mother wavelet function to be similar to a first derivative of a polyp.

13. A method of identifying colon polyps in a digital volume, wherein said volume comprises a plurality of values corresponding to a domain of points in a 3D space, said method comprising the steps of:

selecting a mother wavelet scaling function that corresponds to a polyp;

performing a forward wavelet transformation on said volume to obtain a set of wavelet coefficients, wherein each wavelet coefficient is associated with a length scale;

determining, for each length scale, a threshold magnitude;

selecting, for each length scale, those wavelet coefficients whose magnitudes exceeds the threshold for said length scale, to form a template subset of wavelet coefficients associated with said length scale;

performing a forward wavelet transform of a second volume, to obtain a second set of wavelet coefficients;

comparing, for each length scale, coefficients of the second set of wavelet coefficients with coefficients of a template subset for said length scale, wherein a match between said coefficients of said template subset and coefficients of said second set of wavelet coefficients indicates the presence of a polyp corresponding to said length scale.

14. The method of claim 13, further comprising, prior to performing the forward wavelet transform, calculating a first derivative of said volume, selecting the mother wavelet function to be similar to a first derivative of a polyp, and, prior to performing the forward wavelet transform of the second volume, calculating a first derivative of said second volume.