Method and apparatus for detecting missed areas during endoscopy

Patent metadata

Field	Value
Publication number	US-11219358-B2
Application number	US-202117177167-A
Country	US
Kind code	B2
Filing date	Feb 16, 2021
Priority date	Mar 2, 2020
Publication date	Jan 11, 2022
Grant date	Jan 11, 2022

Abstract

A method of processing images captured using an endoscope comprising a camera is disclosed. According to this method, regular images captured by the camera are received while the endoscope travels through a human gastrointestinal (GI) tract. The regular images are mosaicked to determine any missed or insufficiently imaged area in a section of the human GI tract already travelled by the endoscope. If any missed or insufficiently imaged area is detected, information regarding any missed or insufficiently imaged area is provided. When a target area in the regular images is lack of parallax, the target area is determined as one missed area and an edge corresponding to a structure of the human lumen is highlighted. For a capsule endoscope, the endoscope can be configured to be controlled or steered to move so as to re-image the missed or insufficiently imaged area.

First claim

The invention claimed is:

1. A method of processing images captured using an endoscope comprising a camera, the method comprising:
receiving regular images captured by the camera while the endoscope travels through a human lumen;
mosaicking the regular images into a mosaicked image to determine any missed area in a section of the human lumen travelled by the endoscope, wherein distance information of the regular images is used to assist said mosaicking the regular images, and the distance information is derived based on structured light images associated with the regular images, and wherein the regular images are normalized according to the distance information of the regular images and optical magnification information to facilitate said mosaicking the regular images; and
if any missed area is detected, providing information regarding any missed area.

2. The method of claim 1 , wherein the distance information is used to determine whether a target area in one regular image is out of focus or not and if the target area is out of focus in all regular images covering the target area, information of the target area is provided.

3. The method of claim 1 , wherein the endoscope further comprises a motion sensing device to measure camera motion inside the human lumen.

4. The method of claim 3 , wherein the motion sensing device corresponds to an accelerometer or a gyrator.

5. The method of claim 3 , wherein the motion sensing device is used to determine camera movement, camera trajectory, camera orientation or any combination thereof.

6. The method of claim 5 , wherein said mosaicking the regular images is performed in a space based on the camera trajectory.

7. The method of claim 1 , wherein a target area in the regular images is lack of a surface area in the mosaicked image, the target area is determined as one missed area.

8. The method of claim 1 , wherein said providing the information regarding any missed area comprises displaying the mosaicked image with highlight on any missed area.

9. The method of claim 1 further comprising generating a 2D or 3D mosaicked image and displaying the 2D or 3D mosaicked image on a display device with any missed area highlighted.

10. The method of claim 1 , wherein the endoscope further comprises a motion sensing device to measure camera motion inside the human lumen.

11. The method of claim 10 , wherein the motion sensing device corresponds to an accelerometer or a gyrator.

12. The method of claim 10 , wherein the motion sensing device is used to determine camera movement, camera trajectory, camera orientation or any combination thereof.

13. The method of claim 12 , wherein said mosaicking the regular images is performed in a space based on the camera trajectory.

14. The method of claim 1 further comprising generating a 2D or 3D mosaicked image and storing the 2D or 3D mosaicked image.

15. The method of claim 14 , wherein the endoscope corresponds to a tethered endoscope and information of the 2D or 3D mosaicked image stored is used by an operator during withdraw process of the endoscope to re-image any missed area.

16. The method of claim 14 , wherein the endoscope corresponds to a tethered endoscope and information of the 2D or 3D mosaicked image stored is used in a subsequent colonoscopy of a same patient.

17. The method of claim 1 , wherein when a target area in the regular images is lack of parallax, the target area is determined as one missed area and an edge corresponding to a structure of the human lumen is highlighted.

18. The method of claim 17 , wherein the endoscope is configured to be controlled or steered to move so as to re-image other side of the edge.

19. The method of claim 1 , wherein when a target area in the regular images is lack of parallax, the target area is determined as one missed area and a related structure in the human lumen is highlighted.

20. The method of claim 19 , wherein the endoscope is configured to be controlled or steered to move so as to re-image the target area.

21. The method of claim 1 , wherein when a target area in the regular images is lack of a surface area in the mosaicked image, the target area is determined as one missed area.

22. The method of claim 1 further comprises deriving camera position, camera movement, camera orientation or a combination thereof inside the human lumen by using motion estimation based on the regular images.

23. The method of claim 1 , wherein the endoscope corresponds to a tethered endoscope.

24. The method of claim 1 , wherein the endoscope corresponds to a capsule endoscope.

25. The method of claim 24 , wherein the endoscope is configured to be controlled or steered to move so as to re-image a detected missed area.

26. The method of claim 1 further comprising determining aggregated redundant areas in the regular images and reducing frame rate of the regular images if the aggregated redundant areas exceed a threshold.

27. The method of claim 1 further comprising increasing frame rate of the regular images if detected missed areas exceed a threshold.

28. A non-transitory computer-readable medium having stored thereon a computer-readable code executable by a processor to cause the processor to:
receive regular images captured by a camera while an endoscope travels through a human lumen;
mosaic the regular images into a mosaicked image to determine any missed area in a section of the human lumen travelled by the endoscope, wherein distance information of the regular images is used to assist said mosaicking the regular images, and the distance information is derived based on structured light images associated with the regular images, and wherein the regular images are normalized according to the distance information of the regular images and optical magnification information to facilitate said mosaicking the regular images; and
if any missed area is detected, provide information regarding any missed area.

Assignees

• Capsovision Inc
• Capso Vision Inc

Inventors

• Wang Kang-Huai
• Wu Chenyu
• Hadley Mark
• Wilson Gordon C

Classifications

• A61B1/00009 Primary

  of image signals during a use of endoscope · CPC title

• A61B1/00194

  adapted for three-dimensional imaging · CPC title

• A61B1/00097

  Sensors · CPC title

• A61B1/041 Primary

  Capsule endoscopes for imaging · CPC title

• A61B1/00045

  Display arrangement · CPC title