Automatic stent detection

• US 9,307,926 B2
• Filed: 10/02/2013
• Issued: 04/12/2016
• Est. Priority Date: 10/05/2012
• Status: Active Grant

• Current Assignee: Volcano Corporation (Koninklijke Philips N.V.)
• Original Assignee: Volcano Corporation (Koninklijke Philips N.V.)
• Inventors: Begin, Elizabeth, Kemp, Nathaniel J., Sproul, Jason, Elmaanaoui, Badr
• Primary Examiner(s): Lu, Tom Y
• Application Number:
  US14/044,441
• Publication Number:
  US 20140100449A1
• Time in Patent Office: 923 Days
• US Class Current: 1/1
• CPC Class Codes: A61B 34/20 Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
• CPC Class Codes: A61B 5/0066 Optical coherence imaging
• CPC Class Codes: A61B 5/0073 by tomography, i.e. reconstruction of 3D images from 2D projections A61B5/0066 takes precedence
• CPC Class Codes: A61B 5/061 Determining position of a probe within the body employing means separate from the probe, e.g. sensing internal probe position employing impedance electrodes on the surface of the body A61B1/00158 takes precedence
• CPC Class Codes: A61B 5/6852 Catheters
• CPC Class Codes: A61B 8/0841 for locating instruments
• CPC Class Codes: A61B 8/12 in body cavities or body tracts, e.g. by using catheters catheters per se A61M25/00
• CPC Class Codes: A61B 8/14 Echo-tomography
• CPC Class Codes: A61B 8/4416 related to combined acquisition of different diagnostic modalities, e.g. combination of ultrasound and X-ray acquisitions
• CPC Class Codes: A61B 8/5223 for extracting a diagnostic or physiological parameter from medical diagnostic data for algorithms to analyse biomedical images G06T7/0012
• CPC Class Codes: A61B 8/5261 combining images from different diagnostic modalities, e.g. ultrasound and X-ray
• CPC Class Codes: G06F 18/2135 based on approximation criteria, e.g. principal component analysis
• CPC Class Codes: G06F 18/24143 Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/10101 Optical tomography; Optical coherence tomography [OCT]
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/10136 3D ultrasound image
• CPC Class Codes: G06T 2207/20048 Transform domain processing
• CPC Class Codes: G06T 2207/20076 : Probabilistic image processing
• CPC Class Codes: G06T 2207/20081 : Training; Learning
• CPC Class Codes: G06T 2207/30021 : Catheter; Guide wire
• CPC Class Codes: G06T 2207/30101 : Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/0012 : Biomedical image inspection
• CPC Class Codes: G06T 7/0014 : using an image reference approach
• CPC Class Codes: G06T 7/70 : Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06T 7/74 : involving reference images or patches
• CPC Class Codes: G06V 10/7715 : Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
• CPC Class Codes: G16H 50/30 : for calculating health indices; for individual health risk assessment

Abstract

This invention relates generally to the detection of objects, such as stents, within intraluminal images using principal component analysis and/or regional covariance descriptors. In certain aspects, a training set of pre-defined intraluminal images known to contain an object is generated. The principal components of the training set can be calculated in order to form an object space. An unknown input intraluminal image can be obtained and projected onto the object space. From the projection, the object can be detected within the input intraluminal image. In another embodiment, a covariance matrix is formed for each pre-defined intraluminal image known to contain an object. An unknown input intraluminal image is obtained and a covariance matrix is computed for the input intraluminal image. The covariances of the input image and each image of the training set are compared in order to detect the presence of the object within the input intraluminal image.

Claims

1. A computer-readable, non-transitory medium storing software code representing instructions that when executed by a computing system cause the computing system to perform a method of detecting an object within an intraluminal image, the method comprising

generating a set of pre-defined intraluminal images known to display an object;

computing principal components for the set to create an object space for the object;

projecting an input intraluminal image onto the object space; and

detecting the object in the input intraluminal image,

wherein the object is selected from the group comprising a stent strut or a guidewire.

2. The computer-readable, non-transitory medium of claim 1, wherein the step of detecting further comprises

calculating an error between the input intraluminal image and the object space.

3. The computer-readable, non-transitory medium of claim 2, wherein a small error constitutes a positive detection of the object in the input intraluminal image.

4. The computer-readable, non-transitory medium of claim 1, wherein the pre-defined intraluminal images and input intraluminal image are one-dimensional, two-dimensional, or three-dimensional.

5. The computer-readable, non-transitory medium of claim 1, further comprising

post-processing the input intraluminal image.

6. The computer-readable, non-transitory medium of claim 5, wherein the step of post-processing comprises removing false object detections and highlighting the object detections.

7. The computer-readable, non-transitory medium of claim 1, further comprising

performing the steps of generating, identifying, and projecting for at least one other object;

and

detecting the at least one other object in the input intraluminal image.

8. The computer-readable, non-transitory medium of claim 7, wherein the step of detecting the at least one other object further comprises

calculating an error between the input intraluminal image and the object space for the object and between the input intraluminal image and the object space for the at least one other object.

9. The computer-readable, non-transitory medium of claim 7, further wherein the smaller error constitutes a positive detection for the corresponding object.

10. The computer-readable, non-transitory medium of claim 7, wherein the pre-defined intraluminal images and input intraluminal image are one-dimensional, two-dimensional, or three-dimensional.

11. The computer-readable, non-transitory medium of claim 7, wherein the object is selected from the group consisting of a tissue, a stent strut, or a guidewire.

12. The computer-readable, non-transitory medium of claim 7, further comprising

post-processing the input intraluminal image.

13. The computer-readable, non-transitory medium of claim 12, wherein post-processing comprises removing false object detections and highlighting the object detections.

14. A system for automatically detecting an object within an intraluminal image, comprising:

a central processing unit (CPU); and

a storage device coupled to the CPU and having stored there information for configuring the CPU to;

generate a set of pre-defined intraluminal images known to display a object;

compute principal components for the set to create an object space for the object; and

project an input intraluminal data image onto the object space;

detect the object in the input intraluminal image,

wherein the object is selected from the group comprising a stent strut or a guidewire.

15. The system of claim 14, wherein detecting the object further comprises

calculating an error between the input intraluminal image and the object space.

16. The system of claim 15, wherein a small error as compared to the object space constitutes a positive detection of the object in the input intraluminal image.

17. The system of claim 14, wherein the pre-defined intraluminal images and input intraluminal image are one-dimensional, two-dimensional, or three-dimensional.

18. The system of claim 14, further comprising

post-processing the input intraluminal image.

19. The system of claim 18, wherein post-processing comprises removing false object detections and highlighting the object detections.

20. The system of claim 14, further comprising

performing the steps of generating, identifying, and projecting for at least one other object;

and

detecting the at least one other object in the input intraluminal image.

21. The system of claim 20, wherein the detecting the at least one other object further comprises

calculating an error between the input intraluminal image and the object space for the object and between the input intraluminal image and the object space for the at least one other object.

22. The system of claim 21, further wherein the smaller error constitutes a positive detection for the corresponding object.

23. The system of claim 20, wherein the pre-defined intraluminal images and input intraluminal image are one-dimensional, two-dimensional, or three-dimensional.

24. The system of claim 20, wherein the object is selected from the group consisting of a lumen border, a stent strut, or a guidewire.

25. The system of claim 20, further comprising

post-processing the input intraluminal image.

26. The system of claim 25, wherein post-processing comprises removing false object detections and highlighting the object detections.

27. A computer-readable, non-transitory medium storing software code representing instructions that when executed by a computing system cause the computing system to perform a method of detecting an object within a blood vessel, the method comprising:

generating a training set of pre-defined intraluminal images known to display an object, each intraluminal image comprising a feature;

computing a covariance for a feature within each intraluminal image of the training set; and

detecting the object within an input intraluminal image, the detecting step comprising;

computing a covariance for a feature within the intraluminal input image; and

comparing the covariance of the input intraluminal image to the covariances of the training set to detect the object in the input image,

wherein the object is selected from the group comprising a stent strut or a guidewire.

28. The computer-readable, non-transitory medium of claim 27, wherein a feature is selected from the group consisting of the Cartesian coordinates of a pixel, the intensity at each pixel, and the first and second order derivatives of the image in the x and y direction.

29. The computer-readable, non-transitory medium of claim 27, wherein the pre-defined intraluminal images and input intraluminal image are one-dimensional, two-dimensional, or three-dimensional.

30. A method for detecting an object in an intraluminal image, the method comprising the steps of:

generating a set of pre-defined intraluminal images known to display an object;

computing principal components for the set to create an object space for the object;

projecting an input intraluminal image onto the object space; and

detecting the object in the input intraluminal image,

wherein the object is selected from the group comprising a stent strut or a guidewire.

31. The method of claim 30, further comprising processing the input intraluminal image.

32. The method of claim 31, wherein said processing step processing comprises removing false object detections and highlighting the object detections.

33. The method of claim 30, wherein the step of detecting the at least one other object further comprises calculating an error between the input intraluminal image and the object space for the object and between the input intraluminal image and the object space for the at least one other object.