KR20160005256A - Seed point detection method for coronary artery extraction from CCTA - Google Patents
Seed point detection method for coronary artery extraction from CCTA Download PDFInfo
- Publication number
- KR20160005256A KR20160005256A KR1020140083789A KR20140083789A KR20160005256A KR 20160005256 A KR20160005256 A KR 20160005256A KR 1020140083789 A KR1020140083789 A KR 1020140083789A KR 20140083789 A KR20140083789 A KR 20140083789A KR 20160005256 A KR20160005256 A KR 20160005256A
- Authority
- KR
- South Korea
- Prior art keywords
- coronary artery
- heart
- tracking
- patient
- vessel
- Prior art date
Links
- 210000004351 coronary vessel Anatomy 0.000 title claims abstract description 41
- 238000001514 detection method Methods 0.000 title abstract description 11
- 238000000605 extraction Methods 0.000 title description 3
- 101000666730 Homo sapiens T-complex protein 1 subunit alpha Proteins 0.000 title 1
- 102100038410 T-complex protein 1 subunit alpha Human genes 0.000 title 1
- 238000000034 method Methods 0.000 claims description 20
- 238000010968 computed tomography angiography Methods 0.000 abstract description 3
- 210000004204 blood vessel Anatomy 0.000 description 8
- 210000000709 aorta Anatomy 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 239000002245 particle Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000002583 angiography Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 241000272517 Anseriformes Species 0.000 description 1
- 241001300078 Vitrea Species 0.000 description 1
- 230000001746 atrial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000002966 varnish Substances 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/46—Arrangements for interfacing with the operator or the patient
- A61B6/467—Arrangements for interfacing with the operator or the patient characterised by special input means
- A61B6/469—Arrangements for interfacing with the operator or the patient characterised by special input means for selecting a region of interest [ROI]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
- A61B6/503—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of the heart
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
- A61B6/504—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of blood vessels, e.g. by angiography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- High Energy & Nuclear Physics (AREA)
- Physics & Mathematics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Optics & Photonics (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Vascular Medicine (AREA)
- Human Computer Interaction (AREA)
- Pulmonology (AREA)
- Theoretical Computer Science (AREA)
- Cardiology (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Description
The present invention relates to a seed point detection method for tracking a coronary artery, and more particularly, to a seed point detection method for three-dimensional tracking of a CT coronary artery.
Techniques for tracking blood vessels among many commonly used methods for reconstructing coronary arteries from coronary computed tomographic angiography (CCTA) for various purposes (such as FFR-ct) have. Such a vessel tracing technique requires a seed point for tracing the blood vessel.
Conventional methods of finding the starting point for tracking blood vessels are entirely dependent on the location and shape of the aorta.
It is an object of the present invention to provide a new method for finding a starting point for tracing blood vessels in applying a blood vessel tracking technique for three-dimensional reconstruction of a coronary artery.
In other words, the present invention aims to provide a method that is different from methods for finding a conventional starting point that depends entirely on the position and shape of the aorta, reduces the amount of calculation, and has a high probability of finding a starting point.
A seed point detection method for three-dimensional tracking of a CT coronary artery according to the present invention includes: obtaining a CT angiogram image of a coronary artery of a patient using an X-ray CT scanner; Selecting an axial slice image from the center of the heart and including the coronary arteries of the heart of the patient in the obtained CT angiographic image; Extracting a region of interest (ROI) from the selected axial slice image; Detecting seed points based on a Hessian-based vesselness filter with integrated local geometric features in the extracted region of interest; And tracking the coronary artery of the patient's heart using the detected seed points.
According to the present invention, the detection of a seed point for coronary artery extraction is an indispensable factor. Commercial software QAngioCT, Xellis, Vitrea, etc. for analyzing the CCTA is aorta-based software It is possible to extract the starting point of the coronary artery.
In addition, if the coronary artery analysis software based on the method proposed in the present invention is made, it will be in a comparative advantage. You will be able to grow into software that competes with the above-mentioned commercial software.
1 is a diagram schematically illustrating a process of tracking a coronary artery in a CCTA according to a seed point detection method for three-dimensional tracking of a CT coronary artery according to the present invention.
FIG. 2 is a diagram for explaining a process of selecting a region of interest (ROI).
Figure 3 is a view showing some examples of tracked contours of vessel and non-vessel structures.
FIG. 4 is a conceptual diagram for determining a vessel direction by calculating an eigenvector corresponding to a minimum eigenvalue of a hennesian matrix in a voxel.
5 is a diagram showing an example of the profiles of the maximum value and the minimum value at the vessel and non-vessel points.
Figure 6 is an illustration of examples of starting points of coronary arteries extracted using the method according to the present invention.
Hereinafter, a method of detecting a seed point for three-dimensional tracking of a CT coronary artery according to an embodiment of the present invention will be described in detail.
Techniques for tracking blood vessels among many commonly used methods for reconstructing coronary arteries from coronary computed tomographic angiography (CCTA) for various purposes (such as FFR-ct) have. Such a vessel tracing technique requires a seed point for tracing the blood vessel. In the present invention, a new method for finding such a starting point is proposed.
First, to find the seed point,
1. The existing Hessen matrix-based filtering technique is fused with new local geometric features. The new local geometric feature uses the property that the shapes of the blood vessels are the same as the cylinders so that the shapes of successive sections in the axial direction are similar.
2. To reduce the computational complexity, an axial slice image showing all three coronary arteries of the heart is selected, and a region of interest (ROI) is extracted from the slice. And uses only the three-dimensional data contained in the ROI to find the above-mentioned local geometric feature.
3. This new seed point does not depend on the aorta compared with the conventional method, so it becomes relatively useful when the aorta is difficult to see or when it is difficult to find a starting point by using aorta.
4. This starting point is located in the center of the heart.
Figure 1 illustrates the following key steps: (1) selecting an ROI at an axial slice at the center of the heart; (2) detecting seed points based on a Hessian-based vesselness filter incorporating new local geometric features; And (3) coronary computed tomography angiography (CCTA), which consists of tracing the coronary arteries.
(1) ROI selection in axial slice
In the CCTA, the three main coronary arteries run almost perpendicular to the axial slices and appear as small, lighter circles in the middle of the heart. We first choose an axial slice that contains the parts of the three main coronary arteries. The selected slice is located at the approximate center of the scan range, and the index P of the selected slice is calculated according to the following equation (1).
Where N is the number of total slices in a given CCTA scan and c r is a constant. c r may be selected flexibly in some ranges, or some slices may be used at the same time.
Figure 2 shows the ROI selection procedure. Considering that the coronary arteries are usually within the heart zone, heart zone detection techniques [21] are applied to obtain only the heart zone in the selected slice. All potential edges in the detected heart zone are extracted using a Sobel edge detector, and then all short edges are removed by an opening motion. Contour tracing is then used to extract single line strokes for each border component.
The neighborhood of the next tracked pixel is searched from the starting pixel of the current contour and the direction of movement from the previous tracked pixel to the current tracked pixel is stored in the variable dir [i], where i is the current tracked pixel . Then, the absolute curvature (Kc) of each traced contour is defined and calculated as follows.
Here, the length L represents the number of pixels belonging to the traced contour line. For example, a circle with a radius of R has a curvature of 1 / R, and its arc length is 2πR. Their multiplication is 2π, which is invariant to the radius of the circle. When the shapes are complicated, Kc and L increase depending on their complexity and size. 2π is the minimum value of Kc that any closed curve can have, and any number less than 2π is discarded because it is an open contour. Therefore, non-vessel structures can be removed by thresholding of length and curvature (Kc). FIG. 3 shows some examples of tracked contours of vessel and non-vessel structures, and Table 1 shows the curvature and length values of the traced contours in FIG. The large structure (Fig. 3 (f)) has a large length and the straight lines (Fig. 3 (e), (g), and (h)) have a small curvature. The thresholds (T L , T C ) for the removal of these non-vessel structures were set at 80 and 10, respectively.
After this shape analysis, for each candidate of the vessel structure, a bounding box is obtained, which is referred to as the ROI for such vessel candidate. Based baseline filter for detecting seed points for three main coronary arteries only for pixels belonging to ROIs. Since the number of pixels in the ROIs is less than 1% of the total number of pixels in the slice image and one or more slices are used among hundreds of axial slices (to ensure there are no missing coronary arteries) The processing time of the points detection step is significantly reduced.
(2) detection of seed points for coronary arteries
In order to divide the vessel regions in the medical images, a Hessian-based vesselness filter has been widely used. Such a filter measures the similarity to the tubular structure and finds the direction of the vessel relative to each voxel. The goodness F (x) is obtained based on all eigenvalues of the Hessian matrix. However, such filters often experience a step-edge response at the boundary of a non-veil structure. This step-edge response becomes more evident in the CCTA because the atrial chambers have similar CT density values to coronary arteries, which can lead to false extraction results. Therefore, it is difficult to determine a fixed threshold value of the Vessel Nisses response in order to clearly distinguish the boundary between the Vessel Zone and the non-Vessel Structure.
To overcome these drawbacks, the ducks propose a new local geometric feature, GF (x), by performing shape analysis of three sections perpendicular to the vessel direction. As shown in FIG. 4, considering the voxel in A, the direction of Bessel is the minimum eigenvalue of the Hessian matrix in voxel A
Along each ray, a border point is detected using a radial gradient. The length of each line represents the distance between the center point and the border point. For each center point, the lines are sorted according to their lengths, and then the three longest lines and the three shortest lines are removed. Thus, each cross section has ten lines and the length of each line is denoted as d [i] [j], where i and j are the plane index [-1,0,1] and the line index [1, ..., 10]. The geometry of the cross sections is characterized by thin curves of gray. Each line index of j = [1, ..., 10 ], 3 -wire length, which are d [-1,0,1], which are the minimum value among the [j] B min [j] and the maximum value B max, which are about the [j] is calculated. Finally, the proposed local geometric feature is defined as: < RTI ID = 0.0 >
Here, k is a constant constant. FIG. 5 shows an example of the profiles of the maximum value and the minimum value at the vessel and non-vessel points. At Bessel point A, since the three cross-sectional borders are similar to each other, the minimum values are closer to the maximum values than the non-Bessel points, resulting in a larger geometric feature value. On the other hand, at the non-Bessel point (E), the arcs between the maximum and minimum values are much larger than those at the Bessel points, resulting in minor geometric feature values. Physically, Equation (4) measures the uniformity of the shapes of the cross sections along the center axis of the vessel, i.e., through the plane indices of [-1,0,1].
The new local geometric feature GF (x) in the present invention is combined with the Hessian-based Vessel Ness F (x) according to the following equation (5).
Where T F and T GF are the threshold values for the Hessian based versus varnish and the new local geometric feature, respectively. Candidate seed points consisting of pixels belonging to ROIs that have been detected in the previous ROI selection step are selected such that their Hessian based vece ness and their local geometric features are larger than T F and T GF , The final bevelling value of only 1 is obtained. Therefore, at the candidate seed point, the proposed geometric feature GF (x) needs to be calculated only when the Hessian-based goodness of the pixel, F (x), is less than the threshold T F.
(3) tracing the coronary artery
For tracing the coronary arteries, a particle filtering based tracking algorithm is used with seed points, which are the centroids of the vessel regions detected in the previous step and their corresponding vessel directions. The tracking algorithm considers the Bessel segment as an ellipse in 3D space and divides it by tracking the ellipses in the spatial system. With two initial parameters of position and normal vector, particle filtering sequentially tracks the center points of the ellipses. Therefore, by sequentially applying our seed detection method and particle filtering-based tracking algorithm, three main coronary arteries are automatically segmented.
While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. I will understand. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.
Claims (1)
Obtaining a CT angiogram image for a coronary artery of a patient using an X-ray CT scanner;
Selecting an axial slice image from the center of the heart and including the coronary arteries of the heart of the patient in the obtained CT angiographic image;
Extracting a region of interest (ROI) from the selected axial slice image;
Detecting seed points based on a Hessian-based vesselness filter with integrated local geometric features in the extracted region of interest; And
And tracking the coronary artery of the patient's heart using the detected seed points. ≪ Desc / Clms Page number 18 > 14. A method for detecting a seed point for three-dimensional tracking of a CT coronary artery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020140083789A KR20160005256A (en) | 2014-07-04 | 2014-07-04 | Seed point detection method for coronary artery extraction from CCTA |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020140083789A KR20160005256A (en) | 2014-07-04 | 2014-07-04 | Seed point detection method for coronary artery extraction from CCTA |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| KR20160005256A true KR20160005256A (en) | 2016-01-14 |
Family
ID=55173030
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020140083789A KR20160005256A (en) | 2014-07-04 | 2014-07-04 | Seed point detection method for coronary artery extraction from CCTA |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR20160005256A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709930A (en) * | 2016-12-29 | 2017-05-24 | 上海联影医疗科技有限公司 | Method and device for cutting volume of interest of three-dimensional medical image |
| WO2018074661A1 (en) * | 2016-10-19 | 2018-04-26 | 순천향대학교 산학협력단 | Coronary artery blood vessel subtraction device and method using vessel correspondence optimization |
| KR20180097037A (en) * | 2017-02-22 | 2018-08-30 | 연세대학교 산학협력단 | A method for automatically extracting a starting point of coronary arteries, and an apparatus thereof |
| WO2019004624A1 (en) * | 2017-06-30 | 2019-01-03 | 연세대학교 산학협력단 | Method for detecting left atrial appendage by using geometric information of heart |
| EP3726460A1 (en) * | 2019-04-06 | 2020-10-21 | Kardiolytics Inc. | Autonomous segmentation of contrast filled coronary artery vessels on computed tomography images |
-
2014
- 2014-07-04 KR KR1020140083789A patent/KR20160005256A/en not_active Application Discontinuation
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018074661A1 (en) * | 2016-10-19 | 2018-04-26 | 순천향대학교 산학협력단 | Coronary artery blood vessel subtraction device and method using vessel correspondence optimization |
| KR101852689B1 (en) * | 2016-10-19 | 2018-06-11 | 순천향대학교 산학협력단 | Coronary Vessel extraction apparatus using vessel correspondence optimization and method thereof |
| CN106709930A (en) * | 2016-12-29 | 2017-05-24 | 上海联影医疗科技有限公司 | Method and device for cutting volume of interest of three-dimensional medical image |
| US10417767B2 (en) | 2016-12-29 | 2019-09-17 | Shenzhen United Imaging Healthcare Co., Ltd. | Systems and methods for image segmentation |
| CN106709930B (en) * | 2016-12-29 | 2020-03-31 | 上海联影医疗科技有限公司 | Method and device for segmenting interested volume of three-dimensional medical image |
| KR20180097037A (en) * | 2017-02-22 | 2018-08-30 | 연세대학교 산학협력단 | A method for automatically extracting a starting point of coronary arteries, and an apparatus thereof |
| WO2019004624A1 (en) * | 2017-06-30 | 2019-01-03 | 연세대학교 산학협력단 | Method for detecting left atrial appendage by using geometric information of heart |
| EP3726460A1 (en) * | 2019-04-06 | 2020-10-21 | Kardiolytics Inc. | Autonomous segmentation of contrast filled coronary artery vessels on computed tomography images |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bouma et al. | Automatic detection of pulmonary embolism in CTA images | |
| KR102050649B1 (en) | Method for extracting vascular structure in 2d x-ray angiogram, computer readable medium and apparatus for performing the method | |
| KR101004342B1 (en) | Method for detecting liver region and hepatoma in computer tomography images | |
| EP3497669B1 (en) | Method for automatically detecting systemic arteries in arbitrary field-of-view computed tomography angiography (cta). | |
| Mouton et al. | Materials-based 3D segmentation of unknown objects from dual-energy computed tomography imagery in baggage security screening | |
| US9888896B2 (en) | Determining a three-dimensional model dataset of a blood vessel system with at least one vessel segment | |
| KR20160005256A (en) | Seed point detection method for coronary artery extraction from CCTA | |
| Xiao et al. | Pulmonary fissure detection in CT images using a derivative of stick filter | |
| Szymczak et al. | Coronary vessel trees from 3d imagery: a topological approach | |
| EP2580737B1 (en) | Tissue classification | |
| Cavalcanti et al. | Lung nodule segmentation in chest computed tomography using a novel background estimation method | |
| Han et al. | A fast seed detection using local geometrical feature for automatic tracking of coronary arteries in CTA | |
| US8311268B2 (en) | Image object separation | |
| US8774496B2 (en) | Compound object separation | |
| Cui et al. | Coronary artery segmentation via hessian filter and curve-skeleton extraction | |
| KR101494975B1 (en) | Nipple automatic detection system and the method in 3D automated breast ultrasound images | |
| Lee et al. | Intensity-vesselness Gaussian mixture model (IVGMM) for 2D+ t segmentation of coronary arteries for X-ray angiography image sequences | |
| KR102000615B1 (en) | A method for automatically extracting a starting point of coronary arteries, and an apparatus thereof | |
| Bukenya et al. | 3D segmentation of the whole heart vasculature using improved multi-threshold Otsu and white top-hat scale space hessian based vessel filter | |
| Beck et al. | Validation and detection of vessel landmarks by using anatomical knowledge | |
| Rahman et al. | An algorithm for extracting centerline of the aorta from CT/MR 3D images | |
| Dehkordi | Extraction of the best frames in coronary angiograms for diagnosis and analysis | |
| Yaguchi et al. | Semi-automated segmentation of solid and GGO nodules in lung CT images using vessel-likelihood derived from local foreground structure | |
| Skoura et al. | Integrating edge detection and fuzzy connectedness for automated segmentation of anatomical branching structures | |
| Raj et al. | An efficient lung segmentation approach for interstitial lung disease |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WITN | Withdrawal due to no request for examination |