Time-resolved diffusion tomographic imaging in highly scattering turbid media

• US 5,813,988 A
• Filed: 03/18/1996
• Issued: 09/29/1998
• Est. Priority Date: 02/03/1995
• Status: Expired due to Fees

• Current Assignee: The Research Foundation of State University of New York
• Original Assignee: Rutgers University
• Inventors: Cai, Wei, Liu, Feng, Alfano, Robert R., Lax, Melvin, Das, Bidyut B.
• Primary Examiner(s): Lateef, Marvin M.
• Assistant Examiner(s): SHAW, SHAWNA JEANNINE
• Application Number:
  US08/618,471
• Time in Patent Office: 925 Days
• Field of Search: 128/664, 128/665, 128/633, 356/446, 356/237, 356/239, 356/240, 356/432, 356/436, 356/441, 250/341.1, 250/358.1, 600/310, 600/473, 600/476
• US Class Current: 600/476
• CPC Class Codes: A61B 5/0073 by tomography, i.e. reconstruction of 3D images from 2D projections A61B5/0066 takes precedence
• CPC Class Codes: A61B 5/0091 for mammography
• CPC Class Codes: G01N 21/4795 spatially resolved investigating of object in scattering medium in vivo A61B

Abstract

A method for imaging objects in highly scattering turbid media. According to one embodiment of the invention, the method involves using a plurality of intersecting source/detectors sets and time-resolving equipment to generate a plurality of time-resolved intensity curves for the diffusive component of light emergent from the medium. For each of the curves, the intensities at a plurality of times are then inputted into the following inverse reconstruction algorithm to form an image of the medium:

X.sup.(k+1).spsp.T = Y.sup.T W+X.sup.(k).spsp.T Λ! W.sup.T

W+Λ!^-1

wherein W is a matrix relating output at detector position r_d, at time t, to source at position r_s, Λ is a regularization matrix, chosen for convenience to be diagonal, but selected in a way related to the ratio of the noise, <nn> to fluctuations in the absorption (or diffusion) X_j that we are trying to determine:

Λ.sub.ij =λ.sub.j δ.sub.ij with λ.sub.j

=<nn>/<ΔXjΔXj>

Here Y is the data collected at the detectors, and X^k is the kth iterate toward the desired absoption information.

Claims

1. A method of imaging an object located in a highly scattering turbid medium, said method comprising the steps of:

(a) illuminating the object through the highly scattering turbid medium with a pulse of light, whereby the light emergent from the highly scattering turbid medium consists of a ballistic component, a snake-like component and a diffusive component;

(b) determining the intensity of said diffusive component at a plurality of points in time; and

(c) using said intensity determinations to form an image of the object in the highly scattering turbid medium, said using step comprising using a mathematical inversion algorithm to generate an image of the highly scattering turbid medium, said mathematical inversion algorithm being

space="preserve" listing-type="equation">X.sup.(K+1).spsp.T = Y.sup.T W+X.sup.(K).spsp.T Λ

! W.sup.T W+Λ

!.sup.-1

wherein W is a matrix relating output at detector position r_d, at time t, to source at position r_s, Λ

is a regularization matrix, chosen for convenience to be diagonal, but selected in a way related to the ratio of the noise, <

nn>

to fluctuations in the absorption (or diffusion) X_j that we are trying to determine;

space="preserve" listing-type="equation">Λ

.sub.ij =λ

.sub.j δ

.sub.ij with λ

.sub.j =<

nn>

/<

Δ

X.sub.j Δ

X.sub.j >

Here Y is the data collected at the detector, and X^k is the kth iterate toward the desired absorption information.

2. The method as claimed in claim 1 wherein said pulse of light has a wavelength in the visible to near infrared region of the spectrum.

3. The method as claimed in claim 2 wherein said pulse of light has a wavelength of about 700 nm to about 1500 nm.

4. The method as claimed in claim 1 wherein said pulse of light is an ultrashort pulse of light.

5. The method as claimed in claim 1 wherein said pulse of light is a laser pulse emitted from a laser selected from the group consisting of Ti:

Sapphire lasers, Cr⁴⁺ Forsterite lasers, Cr⁴⁺ YAG lasers, semiconductor lasers and Nd;

YAG lasers.

6. The method as claimed in claim 1 wherein said highly scattering turbid medium is a biological tissue.

7. The method as claimed in claim 6 wherein said biological tissue is selected from the group consisting of human breast tissue, human brain tissue, human neck tissue and human prostate tissue and wherein said object is a tumor.

8. The method as claimed in claim 6 further comprising the step of detecting cancerous tumors in said biological tissue using fluorescence spectroscopy.

9. The method as claimed in claim 1 wherein said determining step comprises measuring the intensity of light at a plurality of points in the time frame spanning from 50 ps to 10 ns after illumination.

10. The method as claimed in claim 9 wherein said determining step comprises measuring the intensity of light at hundreds of points in the time frame spanning from 50 ps to 10 ns after illumination.

11. The method as claimed in claim 10 wherein said hundreds of points are determined by slicing the time frame into equal portions.

12. A method of forming a map of a highly scattering turbid medium, said method comprising the steps of:

(a) illuminating the highly scattering turbid medium with a first pulse of light along a first axis of incidence, whereby the light emergent from the highly scattering turbid medium due to said first pulse of light consists of a ballistic component, a snake component and a diffusive component;

(b) determining, at a plurality of times, the intensity of said diffusive component of the first pulse of light emergent from the highly scattering turbid medium at a plurality of locations;

(c) illuminating the highly scattering turbid medium with a second pulse of light along a second axis of incidence, said second axis of incidence intersecting with said first axis of incidence, whereby the light emergent from the highly scattering turbid medium due to said second pulse of light consists of a ballistic component, a snake component and a diffusive component;

(d) determining, at a plurality of times, the intensity of said diffusive component of the second pulse of light emergent from the highly scattering turbid medium at a plurality of locations; and

(e) using the intensity determinations from steps (b) and (d) to generate a map of the highly scattering turbid medium, said using step comprising using a mathematical inversion algorithm to generate a map of the highly scattering turbid medium, said mathematical inversion algorithm being

space="preserve" listing-type="equation">X.sup.(K+1).spsp.T = Y.sup.T W+X.sup.(K).spsp.T Λ

! W.sup.T W+Λ

!.sup.-1

wherein W is a matrix relating output at detector position r_d, at time t, to source at position r_s, Λ

is a regularization matrix, chosen for convenience to be diagonal, but selected in a way related to the ratio of the noise, <

nn>

to fluctuations in the absorption (or diffusion) X_j that we are trying to determine;

space="preserve" listing-type="equation">Λ

.sub.ij =λ

.sub.j δ

.sub.ij with λ

.sub.j =<

nn>

/<

Δ

X.sub.j Δ

X.sub.j >

Here Y is the data collected at the detectors, and X^k is the kth iterate toward the desired absorption information.

13. The method as claimed in claim 12 wherein each of said first and said second pulses of light has a wavelength in the visible to near infrared region of the spectrum.

14. The method as claimed in claim 12 wherein each of said first and said second pulses of light has a wavelength of about 700 nm to about 1500 nm.

15. The method as claimed in claim 12 wherein each of said first and said second pulses of light is an ultrashort pulse of light.

16. The method as claimed in claim 12 wherein each of said first and said second pulses of light is a laser pulse emitted from a laser selected from the group consisting of Ti:

Sapphire lasers, Cr⁴⁺ Forsterite lasers, Cr⁴⁺ YAG lasers, semiconductor lasers and Nd;

YAG lasers.

17. The method as claimed in claim 12 wherein said highly scattering turbid medium is a biological tissue.

18. The method as claimed in claim 12 wherein said biological tissue is selected from the group consisting of human breast tissue, human brain tissue, human neck tissue and human prostate tissue and wherein said object is a tumor.

19. The method as claimed in claim 12 wherein said determining steps each comprise time-resolving light emergent from the highly scattering turbid medium.

20. The method as claimed in claim 12 further comprising the steps of:

(f) illuminating the highly scattering turbid medium with a third pulse of light along a third axis of incidence, said third axis of incidence intersecting a plane defined by said first and second axes of incidence, whereby the light emergent from the highly scattering turbid medium due to said third pulse consists of a ballistic component, a snake component and a diffusive component; and

(g) determining, at a plurality of times, the intensity of said diffusive component of the third pulse of light emergent from the highly scattering turbid medium at a plurality of locations;

(h) wherein said using step comprises using the intensity determinations from steps (b), (d) and (g) to generate a map of the highly scattering turbid medium.

21. The method as claimed in claim 17 further comprising the step of detecting cancerous tumors in said biological tissue using fluorescence spectroscopy.

22. The method as claimed in claim 12 wherein said determining step comprises measuring the intensity of light at a plurality of points in the time frame spanning from 50 ps to 10 ns after illumination.

23. The method as claimed in claim 12 wherein said determining step comprises measuring the intensity of light at hundreds of points in the time frame spanning from 50 ps to 10 ns after illumination.

24. The method as claimed in claim 23 wherein said hundreds of points are determined by slicing the time frame at regular intervals.

25. The method as claimed in claim 12 wherein said map is an absorption map.

26. The method as claimed in claim 12 wherein said map is a scattering map.