Real time detection of gastrointestinal sections and transitions of an in-vivo device therebetween

• US 8,965,079 B1
• Filed: 09/28/2011
• Issued: 02/24/2015
• Est. Priority Date: 09/28/2010
• Status: Active Grant

• Current Assignee: Given Imaging Limited (Medtronic Plc)
• Original Assignee: Given Imaging Limited (Medtronic Plc)
• Inventors: Zinaty, Ofra, Rudyk, Eduard
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): HEIDEMANN, JASON E
• Application Number:
  US13/247,046
• Time in Patent Office: 1,245 Days
• Field of Search: None
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/10068 Endoscopic image
• CPC Class Codes: G06T 2207/30028 Colon; Small intestine
• CPC Class Codes: G06T 2207/30092 Stomach; Gastric
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/254 involving subtraction of images

Abstract

An in-vivo device captures images of the GI system and transfers frames to an external system for analysis. As frames are transferred from the in-vivo device to the external system frames may be classified as belonging to a particular GI section. Based on statistical analysis of the classes, a temporary determination may be made, that the in-vivo device has transitioned to a “target section”, and, as a result of the temporary determination, the operation mode of the in-vivo device may temporarily change from a first mode to a second mode. If the temporary determination is followed by a determination which is based on analysis of additional classes, the determination that the in-vivo device is in the target section may be made final and the in-vivo device may remain in the second mode. Otherwise, the first operation mode may be resumed and the whole transition determination process may start anew.

Claims

1. A method for detecting the movement of an in-vivo device between sections of a gastrointestinal (GI) tract, comprising:

for a stream of image frames, each of the frames comprising an image captured in vivo by an in in-vivo device traversing the GI tract and related non-image data;

sequentially classifying frames from the stream of frames, as being associated with a GI region based on the image data, the classifying producing frame classes;

analyzing the frame classes; and

determining whether the in-vivo device is in a particular GI section based on the analysis.

2. The method as in claim 1, comprising, for each frame class:

storing the frame class in a storage device;

if the number of frame classes stored in the storage device is smaller than a statistically sufficient number of frame classes, adding to the storage device additional one or more frame classes; and

if the number of classes stored in the storage device is equal to or greater than the statistically sufficient number of frame classes, analyzing the stored frame classes and temporarily determining whether the in-vivo device is in a GI section based on the analysis result.

3. The method of claim 1, comprising classifying frames from the stream of frames as not being associated with any particular GI region.

4. A method for detecting transition of an in-vivo device between sections of the gastrointestinal (GI) system, comprising:

a) receiving a stream of frames, each of the frames comprising image data corresponding to an image captured in vivo by an in in-vivo device traversing the GI tract and related non-image data;

b) classifying frames from the stream of frames, one frame at a time, as related to a particular GI section or as not related to any particular GI section based on image data, wherein each frame to be classified is classified by the time the next frame to be classified is received, the classifying producing frame classes;

c) analyzing the frame classes statistically; and

d) determining whether the in-vivo device is in a particular GI section or has transitioned to a particular GI section based on the analysis.

5. The method as in claim 4, comprising:

if it is determined that the in-vivo device is in a particular GI section or has transitioned to a particular GI section, generating a corresponding indication, and

if no determination can be made regarding the location of the in-vivo device in the GI system or regarding the transition of the in-vivo device to the particular GI section, repeating operations b) to d) with additional one or more frames.

6. The method as in claim 5, comprising selection of frames for classification is contingent on a rate at which the in-vivo device captures images or transfers the stream of frames.

7. The method as in claim 4, wherein the determination includes determining whether the in-vivo device transitions from a particular section of the GI tract to an adjacent section of the GI tract.

8. The method as in claim 4, wherein the class of a frame is selected from the group consisting of invalid frame, stomach frame, small bowel frame, colon frame and rejected frame.

9. The method as in claim 4, wherein classifying a particular frame is based on any one of:

a feature derived from imaging parameters pertaining to the particular frame, imaging parameters pertaining to a classified frame preceding the particular frame and imaging parameters pertaining to a classified frame following the particular frame;

features derived from color information pertaining to the pertinent image, and features derived from non-image information related to the frame.

10. The method as in claim 4, wherein operations b) to d) are repeated with additional one or more frames until a determination is made regarding the location of the in-vivo device in a particular GI section or transition thereof to a particular GI section, or until a timeout is reached.

11. The method as in claim 4, comprising, for each frame class:

(a) buffering the frame class in a class buffer;

(b) if the number of frame classes buffered in the class buffer is smaller than a statistically sufficient number of frame classes, updating the class buffer with additional one or more frame classes;

(c) if the number of classes buffered in the classes buffer is equal to or greater than the statistically sufficient number of frame classes,

analyzing the buffered frame classes and temporarily determining whether the in-vivo device has transitioned to a target GI section based on the analysis result, and if no determination can be made regarding the transition, repeating operations (a) and (c) for the next classified frame.

12. The method as in claim 11, further comprising:

(d) updating the class buffer with additional one or more frame classes, and after each update,

analyzing the frame classes in the updated class buffer and, based on the analysis result,

(i) validating the temporary determination, or

(ii) invalidating the temporary determination and repeating operation (d).

13. The method as in claim 12, comprising:

transitioning the in-vivo device from a first image capturing rate to a second image capturing rate if a transition of the in-vivo device from a first GI section to a second GI section is determined.

14. The method as in claim 13, wherein the frame class associated with each image frame is buffered in a first database of the class buffer if the pertinent image is captured using the first image capturing rate, and in a second database of the class buffer if the pertinent image is captured using the second image capturing rate.

15. The method as in claim 14, wherein transitioning of the in-vivo device from the first image capturing rate to the second image capturing rate is temporary if the transition of the in-vivo device to the second GI section is determined based on statistical analysis of the first database, and permanent if it is determined based on statistical analysis of the second database.

16. An in vivo imaging system for detecting transition of an in-vivo device to a target GI section, comprising:

a memory to store a stream of frames, each of the frames comprising image data corresponding to an image captured in vivo by an in in-vivo device traversing the GI tract and related non-image data; and

a processor to;

classify frames from the stream of frames, one frame at a time, as related to a particular GI section or as not related to any particular GI section based on image data, wherein each frame to be classified is classified by the time the next frame to be classified is received, the classifying producing frame classes;

analyze the frame classes statistically; and

determine whether the in-vivo device is in a particular GI section or has transitioned to a particular GI section based on the analysis.

17. The system of claim 16, wherein:

if it is determined that the in-vivo device is in a particular GI section or has transitioned to a particular GI section, the processor is to generate a corresponding indication, and

if no determination can be made regarding the location of the in-vivo device in the GI system or regarding the transition of the in-vivo device to the particular GI section, the processor is to repeat the classification, analysis and determining operations.

18. The system of claim 16, comprising selection of frames for classification contingent on a rate at which the in-vivo device captures images or transfers the stream of frames.

19. The system of claim 16, wherein the determination includes determining whether the in-vivo device transitions from a particular section of the GI tract to an adjacent section of the GI tract.

20. The system of claim 16, wherein the class of a frame is selected from the group consisting of invalid frame, stomach frame, small bowel frame, colon frame and rejected frame.