System and method to detect a transition in an image stream
A system and method may detect a transition in an image stream captured within a body lumen. One or more parameters may be computed at points corresponding to a time scale of the image stream. Transition in the values of the parameters may be identified and may indicate a location along the body lumen. The image stream may be segmented at the transition points identified.
1. A method for segmenting an image stream captured within a gastrointestinal tract by an in vivo device inserted into said gastrointestinal tract, said device comprising an imager and being designed to pass through said gastrointestinal tract and to capture a stream of images thereof, the method comprising:
receiving a first image and a second image of the image stream captured using said imager during a single pass of the device through the gastrointestinal tract, wherein the first image and the second image correspond to first and second chronological points in a time scale of the image stream, respectively;
computing, using a computer, average pixel color values of the first image and of the second image;
computing average tissue color based on RGB images with a predetermined number of pixels that pass at least one condition selected from: pixel ratio of red pixels to green pixels which passes a predetermined threshold, and pixel ratio of green pixels to blue pixels which passes a predetermined threshold;
detecting, using a computer, a transition between organs of the gastrointestinal tract based on a change in the average pixel color values; and
segmenting the image stream at the points in the time scale of the image stream that correspond to transitions between the organs of the gastrointestinal tract.
2. The method according to claim 1 wherein the organs of the gastrointestinal tract include the esophagus, stomach, small intestine, colon, rectum and the transverse colon.
3. The method according to claim 1 further comprising determining a confidence level of at least one of the transitions.
4. The method according to claim 1 wherein determining a confidence level comprises computing a deviation of the average pixel color values corresponding to a group of image frames from the image stream.
5. The method according to claim 1 further comprising computing intensity values of pixels in the first image and in the second image; and identifying a change in the intensity values between the first image and the second image.
6. The method according to claim 1 further comprising computing a speed or acceleration parameter to detect the movement pattern of the device through the gastrointestinal tract.
7. The method according to claim 1 further comprising displaying a segmented portion of the image stream.
8. The method according to claim 7 further comprising displaying a segmented image stream.
9. The method according to claim 1 comprising computing a location parameter to detect the movement pattern of the device through the gastrointestinal tract.
10. The method according to claim 1 further comprising identifying, using a computer, a change in the characteristics of a movement pattern of the device through the gastrointestinal tract to detect a transition through the transverse colon.
11. A system for segmenting an image stream captured within a gastrointestinal tract comprising:
an in vivo device comprising an imager to capture the image stream from within the gastrointestinal tract during a single pass of said device through the gastrointestinal tract, wherein the image stream comprises a first image and a second image that correspond to first and second chronological points in a time scale of the image stream, respectively;
a parameter computing unit to compute average pixel color values of the first image and the second image and to compute average tissue color based on RGB images with a predetermined number of pixels that pass at least one condition selected from: pixel ratio of red pixels to green pixels which passes a predetermined threshold, and pixel ratio of green pixels to blue pixels which passes a predetermined threshold;
a transition detecting unit to detect a transition between organs of the gastrointestinal tract by identifying a change in the average pixel color values between the first image and the second image; and
a segmentation unit to segment the image stream at the points in the time scale of the image stream that correspond to transitions between organs of the gastrointestinal tract.
12. The system according to claim 11 wherein the transition detecting unit indicates a point along the time scale that corresponds to the transition.
13. The system according to claim 11 wherein the transition detecting unit determines a confidence level of at least one of the transitions.
14. The system according to claim 13 wherein the transition detecting unit determines a confidence level based on a deviation of the average pixel color values corresponding to a group of image frames from the image stream.
15. The system according to claim 13 wherein the transition detecting unit further computes intensity values of pixels in the first image and the second image; and identifies a change in the intensity values between the first image and the second image.
16. The system according to claim 11 comprising a display unit, wherein the display unit displays at least a segment of the image stream.
17. The system according to claim 16 comprising a display unit, wherein the display unit displays a segmented image stream.
18. The system according to claim 11 wherein the transition detecting unit detects a transition through the transverse colon by identifying a change in the characteristics of a movement pattern of the device through the gastrointestinal tract.
19. The system according to claim 18 comprising a location sensor wherein the location sensor senses a location of the in-vivo device, wherein the transition detecting unit identifies the change in the characteristics of the movement pattern of the device based on the sensed location of the in-vivo device.