Variable exposure portable perfusion monitor

1. A method of imaging a target, comprising:

acquiring, by a processor of an imaging apparatus, multiple images of the target, wherein the multiple images have different exposure values;

determining temporal variances for the multiple images, wherein the temporal variances are determined over pixel regions having predetermined dimensions over a pre-defined number of images over a period of time, and wherein the temporal variance of a given pixel region represents a difference between a square of mean square values of the pixels in the given pixel region over a pre-defined number of frames, N, where N is a positive integer, and a mean value of square values of the pixels over the pre-defined number of frames;

determining spatial variances for the multiple images, wherein the spatial variances are determined over the pixel regions having predetermined dimensions, and wherein a spatial variance of a given pixel region represents a difference between a mean value of squares of pixel values and a square of a mean value of pixels in the region; and

generating a perfusion image of the target by combining the temporal variances and the spatial variances such that a local flow rate in the perfusion image at a given pixel is a function of changes in the spatial variances and the temporal variances as a function of exposure values;

wherein the generating the perfusion image comprises evaluating an average according to:

I

P

⁢

E

⁢

R

⁢

F

=

∑

i

=

1

i

=

N

-

1

❘

"\[LeftBracketingBar]"

w

E

i

⁢

I

E

i

SPAT

/

TEMP

-

w

E

i

+

1

⁢

I

E

i

+

1

SPAT

/

TEMP

❘

"\[RightBracketingBar]"

N

-

1

,

wherein N is a total number of exposures collected, I PERF represents pixel value at each pixel location of the perfusion image, I E i SPAT/TEMP and I E i+i SPAT/TEMP represent spatial or temporal variances at corresponding pixel location for exposure times E i and E i+i , W E i and W E i+i are real numbers representing a relative weight for each exposure.

2. The method of claim 1 , wherein

the spatial variance is determined using a local neighborhood variance calculation; and

the temporal variance is determined by computing variance of each pixel of the target over the N frames, wherein N is a positive integer.

3. The method of claim 2 , wherein N is greater than 1, and wherein the temporal variance is determined by storing a single frame in a memory, determining a contribution of the single frame to the temporal variance, and repeating the storing and the determination of the contribution over each of the N frames.

4. The method of claim 3 , wherein images from different exposure values are combined by a difference operation and categorized based on the values resulting from the difference operation and are related to motion of the target being imaged.

5. An apparatus comprising a processor and one or more cameras,

wherein the processor is configured to:

control the one or more cameras to obtain multiple images of a target at different exposure values;

determine temporal variances for the multiple images, wherein the temporal variances are determined over pixel regions having predetermined dimensions over a pre-defined number of images over a period of time, and wherein the temporal variance of a given pixel region represents a difference between a square of mean square values of the pixels in the given pixel region over a pre-defined number of frames, N, where N is a positive integer, and a mean value of square values of the pixels over the pre-defined number of frames;

determine spatial variances for the multiple images, wherein the spatial variances are determined over the pixel regions having predetermined dimensions, and wherein a spatial variance of a given pixel region represents a difference between a mean value of squares of pixel values and a square of a mean value of pixels in the region;

generate a perfusion image of the target by combining the temporal variances and the spatial variances such that a local flow rate in the perfusion image at a given pixel is a function of changes in the spatial variances and the temporal variances as a function of exposure values;

wherein the generating the perfusion image comprises evaluating an average according to:

I PREF =Σ i=1 i=N-1 |W E i I E i SPAT/TEMP −W E i+1 I E i+1 SPAT/TEMP ,

wherein N is a total number of exposures collected, I PERF represents pixel value at each pixel location of the perfusion image, I E i SPAT/TEMP and I E i+i SPAT/TEMP represent spatial or temporal variances at corresponding pixel location for exposure times E i and E i+i , W E i and W E i+i are real numbers representing a relative weight for each exposure.

6. The apparatus of claim 5 , wherein the spatial variance is determined using a local neighborhood variance calculation.

7. The apparatus of claim 5 , wherein the temporal variance is determined by computing variance of each pixel of the target over N frames, wherein N is a positive integer.

8. The apparatus of claim 5 , wherein N is greater than 1, and wherein the temporal variance is determined by storing a single frame in a memory, determining a contribution of the single frame to the temporal variance, and repeating the storing and the determination of the contribution over each of the N frames.

9. The apparatus of claim 8 , wherein images from different exposure values are combined by a difference operation and categorized using resulting values and related to motion of the target under observation.

10. The apparatus of claim 5 , wherein the one or more cameras include:

multiple cameras configured to acquire images simultaneously, where the cameras can be a combination of monochrome and color cameras; and

a processor configured to process the images.

11. The apparatus of claim 10 , wherein monochrome cameras have an intrinsic infrared filter that is removed, and a band-pass optical filter is added and configured to allow only laser light to pass through while filtering all other wavelengths.

12. The apparatus of claim 10 , wherein the color cameras have a laser blocking filter attached to filter out laser wavelength and allowing all other wavelengths to passthrough.

13. The apparatus of claim 10 , wherein the multiple cameras comprise a CCD/CMOS based image sensor with exposure times ranging from 0.1 ms-1 s and camera with higher near-infrared sensitivity.

14. The apparatus of claim 10 , wherein the cameras have optical image stabilization and are capable of variable focusing with different optical magnifications.

15. The apparatus of claim 10 , wherein the multiple cameras have an optical magnification in a range of 0.5×-10×.

16. The apparatus of claim 10 , wherein at least some of the multiple cameras configured to capture frames at frame rates in a range of 10-1000 frames per second.

17. The apparatus of claim 10 , wherein at least some of the multiple cameras are configured to operate at 8 bits or greater resolution per pixel.

18. The apparatus of claim 10 , wherein the processor is configured to overlay normalized perfusion images onto the camera images.

19. A non-transitory computer-readable medium having code stored thereon; the code, upon execution, causing a processor to implement a method comprising:

acquiring multiple images of a target, wherein the multiple images have different exposure values;

determining temporal variances for the multiple images, wherein the temporal variances are determined over pixel regions having predetermined dimensions over a pre-defined number of images over a period of time, and wherein the temporal variance of a given pixel region represents a difference between a square of mean square values of the pixels in the given pixel region over a pre-defined number of frames, N, where N is a positive integer, and a mean value of square values of the pixels over the pre-defined number of frames;

determining spatial variances for the multiple images, wherein the spatial variances are determined over the pixel regions having predetermined dimensions, and wherein a spatial variance of a given pixel region represents a difference between a mean value of squares of pixel values and a square of a mean value of pixels in the region; and

generating a perfusion image of the target by combining the temporal variances and the spatial variances such that a local flow rate in the perfusion image at a given pixel is a function of changes in the spatial variances and the temporal variances as a function of exposure values;

wherein the generating the perfusion image comprises evaluating an average according to:

I PERF =Σ i=1 i=N-1 |W E i I E i SPAT/TEMP −W E i+1 I E i+1 SPAT/TEMP |/N− 1,

wherein N is a total number of exposures collected, I PERF represents pixel value at each pixel location of the perfusion image, I E i SPAT/TEMP and I E i+i SPAT/TEMP represent spatial or temporal variances at corresponding pixel location for exposure times E i and E i+i , W E i and W E i+i are real numbers representing a relative weight for each exposure.

20. The non-transitory computer-readable medium of claim 19 , wherein the spatial variance is determined using a local neighborhood variance calculation; and

the temporal variance is determined by computing variance of each pixel of the target over the N frames, wherein N is a positive integer;

wherein N is greater than 1, and wherein the temporal variance is determined by storing a single frame in a memory, determining a contribution of the single frame to the temporal variance, and repeating the storing and the determination of the contribution over each of the N frames; and

wherein images from different exposure values are combined by a difference operation and categorized based on the values resulting from the difference operation and are related to motion of the target being imaged.