Three dimensional tissue imaging system and method

1. A method of non-destructive imaging of an anomalous tissue in a region of interest, the method comprising:

a) illuminating the region of interest with light in multiple spectral bands, each of the spectral bands being associated with a respective predetermined depth value in the tissue;

b) capturing, with a camera, a two-dimensional image of the region of interest under illumination by the light in each of the spectral bands;

c) identifying a first anomaly area containing the anomalous tissue within the image of a topmost spectral band;

d) applying the first anomaly area to each of the images of the spectral bands that are deeper than the topmost;

e) segmenting each of the images of the deeper spectral bands, one by one, wherein the segmenting automatically identifies pixels belonging to the anomalous tissue, isolating a shape of the anomalous tissue;

f) halting the segmentation of the images of the deeper spectral bands when the anomalous tissue is no longer detected;

g) determining a set of data points for each of the segmented images, wherein each set of data points has lateral position data defining a position within a respective spectral plane at a respective depth value;

h) transforming each set of data points for each segmented image into a single three-dimensional anomaly volume by additionally applying the predetermined depth value to each of the respective set of data points for each segmented image, yielding transformed sets of data points, respectively, each data point within each of the transformed sets of data points thereby assigning lateral position and depth data values;

i) rendering an image of the three-dimensional anomaly volume by plotting the transformed sets of data points; and

j) displaying the rendered image to a user.

2. The method of claim 1 , wherein the segmentation comprises segmenting the image of the deeper spectral bands into a binary image.

3. The method of claim 1 , wherein the camera includes optical components and a sensor.

4. The method of claim 3 , wherein of each of the predetermined depth values is based on the respective spectral band of light and optical calibration of the camera.

5. The method of claim 3 , wherein the sensor is a complementary metal oxide semiconductor-type sensor.

6. The method of claim 1 , wherein the multiple spectral bands of light are provided by one or more light-emitting diodes.

7. The method of claim 1 , further comprising rendering texture within the image of the three-dimensional anomaly volume by additionally processing at least one of the spectral band anomaly areas with a pre-defined computation, in order to illustrate differences in spatial gradient.

8. The method of claim 1 , wherein:

the topmost spectral band is a wavelength of light of about 430 nm and the first depth value is about 0.4 mm.

9. The method of claim 1 wherein identifying the region of interest comprises the following steps:

performing an Expectation Maximization (EM) and Gaussian Mixture Model (GMM) clustering of the pixels;

automatically dividing all of the pixels into the anomalous tissue or a background and assigning a conditional probability at every pixel for being in an anomalous tissue class or a non-anomalous tissue class;

selecting a threshold on a final probability;

accumulating a Cumulative Distribution Function; and

cutting off the threshold to produce a bounding box for the anomalous tissue.

10. The method of claim 9 further comprising adding a Gaussian weighting kernel to the EM/GMM model to eliminate spurious pixels.

11. A system for non-destructive imaging of an anomalous tissue in a region of interest, comprising:

a) a source of illumination of light in multiple spectral bands;

b) a camera for acquiring digital images of the region of interest based on the light re-emitted from the region of interest when illuminated at multiple each of the at least two spectral bands, the digital image comprising digital signals whose values are a function of a condition of the region of interest;

c) a memory for storing the digital images;

d) a digital processor adapted and configured for

i) processing the digital image of a topmost spectral band stored in memory and computing an estimated value of parameters which are a function of the processed topmost image to define a first anomaly area;

ii) processing and applying the first anomaly area to each of the deeper spectral band images;

iii) segmenting each of the images of the deeper spectral bands, one by one, wherein the segmenting automatically identifies pixels belonging to the anomalous tissue, isolating a shape of the anomalous tissue;

iv) halting the segmentation of the images of the deeper spectral bands when the anomalous tissue is no longer detected;

v) determining a set of data points for each of the segmented images, wherein each set of data points has lateral position data defining a position within a respective spectral plane at a respective depth value;

vi) transforming the each set of data points for each segmented image into a single three-dimensional anomaly volume; and

vii) rendering an image of the three-dimensional anomaly; and

e) a display adapted and configured for displaying the rendered image to a user.

12. A non-transitory computer-readable storage medium comprising a plurality of instructions configured to execute on at least one computer processor to enable the computer processor to:

a) initiate illumination of a region of interest with light in multiple spectral bands, each of the spectral bands being associated with a respective predetermined depth value in the tissue;

b) initiate capture, with a camera, a two-dimensional image of the region of interest under illumination by the light in each of the spectral bands;

c) identify a first anomaly area containing the anomalous tissue within the image of a topmost spectral band;

d) apply the first anomaly area to each of the images of the spectral bands that are deeper than the topmost;

e) segment each of the images of the deeper spectral bands, one by one, wherein the segmenting automatically identifies pixels belonging to the anomalous tissue, isolating a shape of the anomalous tissue;

f) halt the segmentation of the images of the deeper spectral bands where the anomalous tissue is no longer detected;

g) determine a set of data points for each of the segmented images, wherein each set of data points has lateral position data defining a position within a respective spectral plane at a respective depth value;

h) transform each set of data points for each segmented image into a single three-dimensional anomaly volume by additionally applying the predetermined depth value to each of the respective setoff data points for each segmented image, yielding transformed sets of data points, respectively, each data point within each of the transformed sets of data points thereby assigning lateral position and depth data values;

i) render an image of the three-dimensional anomaly volume by plotting the transformed sets of data points; and

j) display the rendered image to a user.