Semi-automated image segmentation system and method

Patent metadata

Field	Value
Publication number	US-10275881-B2
Application number	US-201615393430-A
Country	US
Kind code	B2
Filing date	Dec 29, 2016
Priority date	Dec 31, 2015
Publication date	Apr 30, 2019
Grant date	Apr 30, 2019

Abstract

Image segmentation can include a pre-initialization image analysis of image data using an image analysis algorithm to generate a modified image, and the modified image can be presented on a display. An initialization can be performed on the modified image that includes user input on the modified image. The modified image can be segmented using a segmentation algorithm that evaluates the user input. Upon evaluating the user input, the segmentation algorithm can cause a segmented image to be produced which can be presented on the display.

First claim

The invention claimed is:

1. A method of segmenting an image, the method comprising the steps of:
performing a pre-initialization image analysis of image data using an image analysis algorithm to generate a modified image and presenting the modified image on a display, wherein the image data represents a plurality of image elements;
performing an initialization of the modified image generated by the pre-initialization image analysis, wherein performing the initialization includes receiving a user input on the modified image on the display; and
segmenting the modified image using a segmentation algorithm that evaluates the received user input of the performed initialization and displaying a segmented image on the display,
wherein the modified image includes a proposed control contour and the proposed control contour on the modified image on the display comprises a first visual indicator along a first portion of the proposed control contour and a second visual indicator along a second portion of the proposed control contour, wherein the first and second visual indicators are different and one of the first and second visual indicators designates that the respective portion of the proposed control contour is below a threshold level of information determined by the pre-initialization image analysis, and
wherein the second visual indicator along the second portion of the proposed control contour designates that the second portion of the proposed control contour is below a threshold level of information, and wherein the second visual indicator comprises one of a color different from a color of the first visual indicator and a line pattern different from a line pattern of the first visual indicator.

2. The method of claim 1 , further comprising:
generating the image data used in the pre-initialization image analysis representing the plurality of image elements with an intravascular ultrasound imaging system.

3. The method of claim 2 , wherein generating the image data with the intravascular ultrasound imaging system comprises:
collecting cross-sectional image data using a catheter assembly that includes an intravascular imaging device having an imaging module for emitting and receiving energy.

4. The method of claim 3 , wherein the intravascular imaging device comprises an ultrasound transducer configured to emit and receive ultrasound energy.

5. The method of claim 1 , wherein the image analysis algorithm includes calculating a probability density function of the image data.

6. The method of claim 1 , wherein the image analysis algorithm includes calculating a gradient function of the image data.

7. The method of claim 1 , wherein the modified image is clearer relative to an image generated using the image data before the pre-initialization image analysis of the image data is performed.

8. The method of claim 1 , wherein receiving the user input on the modified image on the display comprises receiving a confirmation of a proposed control contour, or a portion thereof, presented on the modified image on the display.

9. The method of claim 1 , wherein receiving the user input on the modified image on the display comprises receiving a modification of a proposed control contour, or a portion thereof, presented on the modified image on the display.

10. The method of claim 1 , wherein receiving the user input on the modified image on the display comprises receiving a control contour placed on the modified image on the display.

11. The method of claim 10 , wherein the control contour is placed at or near a location of a lumen-intima interface on the modified image on the display.

12. The method of claim 10 , wherein the control contour is placed at or near a location of a media-adventitia interface on the modified image on the display.

13. The method of claim 10 , wherein the control contour is placed on a longitudinal image on the display and a first control point is generated on a first cross-sectional image on the display, wherein the first control point corresponds to a location of the control contour on the first cross-sectional image.

14. The method of claim 13 , wherein the first control point is generated on the longitudinal image on the display and a second control point is generated on the longitudinal image on the display, wherein the second control point corresponds to a location of the control contour on a second cross-sectional image, and wherein the first cross-sectional image and the second cross-sectional image are spaced apart along the longitudinal image by a predetermined interval of cross-sectional images.

15. The method of claim 1 , wherein the threshold level of information determined by the pre-initialization image analysis comprises a signal-to-noise ratio of the image data, and wherein the second visual indicator designates that the respective portion of the proposed control contour is below the threshold level signal-to-noise ratio of the image data.

16. The method of claim 1 , wherein the image analysis algorithm increase contrast between neighboring image elements in the image data, wherein the modified image is presented on the display with the increased contrast between neighboring image elements, wherein the initialization is performed after the pre-initialization image analysis, and wherein the segmenting is performed after the initialization.

17. A method of segmenting an image, the method comprising the steps of:
performing a pre-initialization image analysis of image data using an image analysis algorithm to generate a modified image and presenting the modified image on a display, wherein the image data represents a plurality of image elements;
performing an initialization of the modified image generated by the pre-initialization image analysis, wherein performing the initialization includes receiving a user input on the modified image on the display; and
segmenting the modified image using a segmentation algorithm that evaluates the received user input of the performed initialization and displaying a segmented image on the display,
wherein the modified image includes a proposed control contour and the proposed control contour on the modified image on the display comprises a first visual indicator along a first portion of the proposed control contour and a second visual indicator along a second portion of the proposed control contour, wherein the first and second visual indicators are different and one of the first and second visual indicators designates that the respective portion of the proposed control contour is below a threshold level of information determined by the pre-initialization image analysis, and
wherein the threshold level of information determined by the pre-initialization image analysis comprises a signal-to-noise ratio of the image data, and wherein one of the first and second visual indicators designates that the respective portion of the proposed control contour is below the threshold level signal-to-noise ratio of the image data.

18. The method of claim 17 , wherein the one of the first and second visual indicators that designates that the respective portion of the proposed control contour is below the threshold level signal-to-noise ratio of the image data comprises one of a different color and a line pattern.

19. The method of claim 17 , wherein the image analysis algorithm increase contrast between neighboring image elements in the image data, wherein the modified image is presented on the display with the increased contrast between neighboring image elements, wherein the initialization is pe

Assignees

• Acist Medical Sys Inc
• Val Chum Lp

Inventors

• Cardinal Marie-Hélène Roy
• Destrempes François
• Jamello Joseph A
• Cloutier Guy
• Waters Kendall R
• Moore Thomas C

Classifications

• G06T2207/10101

  Optical tomography; Optical coherence tomography [OCT] · CPC title

• G06T2207/30101

  Blood vessel; Artery; Vein; Vascular · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• A61B8/12

  in body cavities or body tracts, e.g. by using catheters · CPC title

• G06T2207/10068

  Endoscopic image · CPC title