CN111445473A - Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction - Google Patents
Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction Download PDFInfo
- Publication number
- CN111445473A CN111445473A CN202010243718.9A CN202010243718A CN111445473A CN 111445473 A CN111445473 A CN 111445473A CN 202010243718 A CN202010243718 A CN 202010243718A CN 111445473 A CN111445473 A CN 111445473A
- Authority
- CN
- China
- Prior art keywords
- intravascular ultrasound
- image
- cross
- longitudinal axis
- segmentation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20068—Projection on vertical or horizontal image axis
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
A vessel membrane accurate segmentation method and system based on multi-angle reconstruction of intravascular ultrasound image sequences are characterized in that corresponding intravascular ultrasound longitudinal axis images are obtained by reconstructing intravascular ultrasound cross-section image sequences, then an improved clustering method is adopted to perform preliminary segmentation on a vessel membrane in the intravascular ultrasound longitudinal axis images, and then a vessel membrane accurate segmentation result under the cross-section images is restored according to the preliminary segmentation result. The invention can simultaneously obtain the vascular membrane segmentation result of two image modes of intravascular ultrasound (namely a cross-section image mode and a longitudinal axis image mode), and overcomes the influence of bifurcation and bypass blood vessels to a certain extent. The blood vessel condition is comprehensively analyzed from the cross section visual angle and the longitudinal axis visual angle, and quantitative parameters are calculated, so that the blood vessel condition is more comprehensively evaluated, reference is provided for further diagnosis of doctors, and the method has clinical practical value.
Description
Technical Field
The invention relates to a technique in the field of image processing, in particular to a vascular membrane accurate segmentation method based on multi-angle reconstruction of intravascular ultrasound image sequences.
Background
Intravascular ultrasound, as shown in fig. 1, sends an ultrasound probe into a vascular lumen for imaging through a catheter technique, can observe the shape and the wall structure of the lumen through a cross-sectional image of the blood vessel, has the advantages of intuition, accuracy and the like, is considered as a 'gold standard' for coronary heart disease diagnosis, and plays a very important role in improving the understanding of coronary artery lesions and guiding interventional therapy. By accurately dividing the boundaries of vascular membranes (middle-adventitia and intima), determining a reference lumen and an actual lumen, quantitative parameters (lumen cross-sectional area, lumen area stenosis and the like) can be calculated for quantitatively analyzing the vascular condition, and the severity of lesion can be judged.
The existing manual segmentation method is characterized in that the blood vessel membranes (the middle-adventitia and the inner membrane) are manually drawn, the effect of the manual segmentation method is different from person to person, the process is time-consuming, and therefore the automatic segmentation algorithm of the blood vessel membranes is worthy of being researched and realized. Because the cross-section image is directly acquired by intravascular ultrasound, most of the existing blood vessel membrane automatic segmentation algorithms directly process the cross-section image. However, one drawback of intravascular ultrasound is that the acquired cross-sectional images reflect only the condition of a certain cross-section of the blood vessel. The cross-section images are directly processed, and the continuity information of the blood vessel structure contained in the whole intravascular ultrasound image sequence is ignored. In addition, when the cross section image is processed, only the cross section and the vascular structure information close to the cross sections can be utilized, so that the blood vessel is easily influenced by blood vessel bifurcation and accompanied by bypass blood vessels, and the accuracy of blood vessel membrane segmentation is reduced.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a vascular membrane accurate segmentation method and a vascular membrane accurate segmentation system based on intravascular ultrasound image sequence multi-angle reconstruction, which designs a vascular membrane segmentation result of a cross section image sequence after performing vascular membrane segmentation on a longitudinal axis image from longitudinal axis mode image reconstruction by introducing vascular structure continuity, can simultaneously obtain the vascular membrane segmentation result of two intravascular ultrasound image modes (namely a cross section image mode and a longitudinal axis image mode), comprehensively analyzes the vascular condition from a cross section visual angle and a longitudinal axis visual angle, calculates quantitative parameters, performs more comprehensive assessment on the vascular condition, provides reference for further diagnosis of doctors, and has clinical practical value.
The invention is realized by the following technical scheme:
the invention relates to a vascular membrane accurate segmentation method based on intravascular ultrasound image sequence multi-angle reconstruction, which comprises the steps of reconstructing an intravascular ultrasound cross-section image sequence to obtain a corresponding intravascular ultrasound longitudinal axis image, then performing primary segmentation on a vascular membrane in the intravascular ultrasound longitudinal axis image by adopting an improved clustering method, and recovering a vascular membrane accurate segmentation result under the cross-section image according to the primary segmentation result.
The intravascular ultrasound cross-sectional image sequence is a complete intravascular ultrasound cross-sectional image sequence acquired.
The reconstruction is carried out by introducing the continuity of the vascular structure contained in the whole cross-section image sequence and reflecting the continuity of the vascular structure on a corresponding longitudinal axis image through sampling and interpolation technologies; the specific operation comprises the following steps: resampling the intravascular ultrasound cross-section image sequence frame by frame, namely sampling the intravascular ultrasound cross-section image sequence by pixel points in the center of the image angle by angle and diameter, interpolating sampling values of all cross-section images in the sequence at the same angle together, and reconstructing a longitudinal axis mode image corresponding to the angle.
In the preliminary segmentation, a rough segmentation result of a vascular membrane is obtained by sequentially using fuzzy clustering on the reconstructed intravascular ultrasound longitudinal axis image; then optimizing through a morphological filter to obtain a vascular membrane preliminary segmentation result of the longitudinal axis intravascular ultrasound image as an initial contour; and then the longitudinal axis image is used for calculating an external force field, and the intravascular ultrasound longitudinal axis image vascular membrane preliminary segmentation result is obtained through active contour model optimization of a plurality of iterations.
The recovery means that: obtaining a vascular membrane segmentation result of the cross-section image through the inverse process of reconstructing the longitudinal axis image; then uniformly down-sampling on the cross section image to reduce the interference of bifurcation and accompanying bypass blood vessels; and finally, fitting a downsampling result circle to be smooth and optimizing the active contour model to obtain a vascular membrane accurate segmentation result of the intravascular ultrasound cross-section image.
The invention relates to a system for realizing the method, which comprises the following steps: the device comprises a reconstruction unit, a primary segmentation unit, a recovery unit and an optimized segmentation unit, wherein: the reconstruction unit obtains a corresponding intravascular ultrasound longitudinal axis image sequence through frame-by-frame resampling and interpolation processing according to the complete intravascular ultrasound cross-section image sequence; the preliminary segmentation unit adopts an improved clustering method to perform preliminary segmentation on the intravascular ultrasound longitudinal axis image and obtain a segmentation result; the recovery unit extracts a blood vessel membrane preliminary segmentation result under the cross-section image from the preliminary segmentation result of the longitudinal axis image through an inverse reconstruction process; and the optimization segmentation unit optimizes the preliminary segmentation result of the vascular membrane of the cross-section image through downsampling, fitting and smoothing and the active contour model so as to obtain an accurate segmentation result.
Technical effects
The invention integrally solves the problems that the cross section image cannot introduce the continuity of the vascular structure and the interference exists in the longitudinal axis image in the prior art. The cross section image sequence is reconstructed to obtain a longitudinal axis image sequence, so that continuity information of a blood vessel structure is introduced, the influence of blood vessel bifurcation and collateral blood vessels is effectively weakened when the blood vessel bifurcation and the collateral blood vessels interfere with each other, and a cross section mode image, a longitudinal axis mode image and a more effective blood vessel membrane segmentation result are realized.
Drawings
FIG. 1 is a schematic diagram of intravascular ultrasound principles;
FIG. 2a is an intravascular ultrasound image with cross section in the embodiment, and FIG. 2b is an exemplary diagram of the result of reconstructing the intravascular ultrasound image with longitudinal axis;
fig. 3a is an exemplary diagram of the vascular interference accompanying the bypass in the vertical axis image, and fig. 3b is an exemplary diagram of the vascular membrane segmentation result of the reconstructed cross-sectional image after the vascular interference accompanying the bypass in the cross-sectional image and the down-sampling recovery;
FIG. 4 is a flow chart of the method of the present invention;
in the figure: (a) the method comprises the steps of (a) obtaining a cross-section intravascular ultrasound image sequence, (b) reconstructing a longitudinal-axis intravascular ultrasound image, (c) obtaining a vascular membrane preliminary segmentation result by fuzzy clustering, (d) obtaining a vascular membrane preliminary segmentation result by morphological filtering, (e) obtaining a longitudinal-axis intravascular ultrasound image vascular membrane segmentation result, (f) reducing sampling and recovering a cross-section intravascular ultrasound image vascular membrane segmentation result;
fig. 5.1 is an exemplary diagram of a reconstruction unit, and fig. 5.2 is an exemplary diagram of a recovery unit;
in the figure: the solid line is the sampling (interpolation) position, the arrow indicates the sampling (interpolation) direction;
FIG. 6 is an exemplary diagram of an optimized segmentation unit;
FIG. 7 is an exemplary longitudinal intravascular ultrasound image vascular membrane segmentation of a data set embodiment disclosed;
in the figure: the dotted line represents the automatic segmentation result, and the solid line represents the gold standard;
FIG. 8 is an exemplary illustration of a cross-sectional intravascular ultrasound image vascular membrane segmentation of a data set embodiment of the disclosure;
in the figure: the dotted line represents the automatic segmentation result, and the solid line represents the gold standard;
FIG. 9 is a longitudinal intravascular ultrasound image vascular membrane of an embodiment of a self-constructed data set;
in the figure: the dotted line represents the automatic segmentation result, and the solid line represents the gold standard;
FIG. 10 is a longitudinal intravascular ultrasound image vascular membrane of an embodiment of a self-constructed data set;
in the figure: the dashed line represents the automatic segmentation result and the solid line represents the gold standard.
Detailed Description
As shown in fig. 4, this embodiment relates to a vascular membrane accurate segmentation method based on multi-angle reconstruction of an intravascular ultrasound image sequence, which reconstructs an intravascular ultrasound cross-sectional image sequence to obtain a corresponding intravascular ultrasound longitudinal image, then performs preliminary segmentation on a vascular membrane in the intravascular ultrasound longitudinal image by using an improved clustering method, and recovers a vascular membrane accurate segmentation result in the cross-sectional image according to the preliminary segmentation result.
The intravascular ultrasound cross-sectional image sequence is a complete intravascular ultrasound cross-sectional image sequence acquired.
The reconstruction is carried out by introducing the continuity of the vascular structure contained in the whole cross-section image sequence and reflecting the continuity of the vascular structure on a corresponding longitudinal axis image through sampling and interpolation technologies; the specific operation comprises the following steps: resampling the intravascular ultrasound cross-section image sequence frame by frame, namely sampling the intravascular ultrasound cross-section image sequence by pixel points in the center of the image angle by angle and diameter, interpolating sampling values of all cross-section images in the sequence at the same angle together, and reconstructing a longitudinal axis mode image corresponding to the angle.
The image sequence is 384 × 384 × in size, which is sampled along the radial direction (the image center is 192 row and 192 column pixel points, the length is 384 pixels), angle by angle (every 1 ° sampling, 180 times in total), and converted into the intravascular ultrasound image sequence in the polar coordinate system in size 384 × × 250. then, the intravascular ultrasound image sequence in the polar coordinate system (for example, the intravascular ultrasound image sequence with 0 ° vertical axis formed by the first column of all images) is interpolated column by column to obtain the corresponding vertical axis intravascular ultrasound image sequence (size 384 × 250 × 180), as shown in fig. 4.
In the preliminary segmentation, a rough segmentation result of the vascular membrane is obtained by fuzzy clustering on the reconstructed intravascular ultrasound longitudinal axis image; then optimizing through a morphological filter to obtain a vascular membrane preliminary segmentation result of the longitudinal axis intravascular ultrasound image as an initial contour; and then using the longitudinal axis image to calculate an external force field, and obtaining an intravascular ultrasound longitudinal axis image vascular membrane preliminary segmentation result through active contour model optimization of a plurality of iterations, wherein the method specifically comprises the following steps:
① for longitudinal axis intravascular ultrasound images, fuzzy clustering is used first to get a rough vessel membrane segmentation result, as shown in FIG. 4 (c).
The fuzzy clustering, that is, the clustering algorithm based on partitioning, makes the similarity between objects partitioned into the same cluster maximum, and the similarity between different clusters minimum. The objective function is:
wherein: u ═ Uik]Is a membership matrix, uikIs the degree of membership of the kth sample to the ith class, dik 2＝||xk-vi||2Is a sample xkAnd the cluster center (mean) viThe euclidean distance of (c). The constraint condition is that the sum of the membership degrees of a certain sample to each cluster is 1, namely:
the clustering problem translates into:
The fuzzy clustering comprises the following specific algorithm steps:
i) setting the clustering number c and a parameter m;
ii) initializing a membership matrix U (0), the sum of the row elements of U (0) being 1;
iii) calculating a new cluster center Vj;
iv) calculating a new membership matrix U;
v) when the termination condition is satisfied, | | U (k +1) -U (k) and | l ≦ e, stopping iteration, otherwise, continuing.
② in order to reduce the small interference caused by speckle noise in FIG. 4(c), a circular morphological filter with a diameter of 9 pixels is used for on-operation noise reduction.
Because the longitudinal-axis intravascular ultrasound image is interfered, especially by bifurcation and bypass vessels, which often appears in a limited and continuous angle, when the vascular membrane segmentation result of the (cross-sectional) intravascular ultrasound image is restored, if the vascular membrane segmentation result of the longitudinal-axis intravascular ultrasound image at all angles is directly used for restoration to obtain the final cross-sectional image vascular membrane segmentation result, the situations shown in fig. 3(b) and fig. 6(a) are inevitably generated, and the segmentation fails. In order to effectively overcome such a situation, the recovery means of the present embodiment is: as shown in fig. 5.2, the vascular membrane segmentation result of the cross-sectional image is obtained by the reverse process of reconstructing the longitudinal axis image; as shown in fig. 6, down-sampling is performed on the cross-sectional image to reduce the interference of bifurcation and bypass blood vessels; and finally, performing circle fitting smoothing and active contour model optimization on the down-sampling result to obtain a vascular membrane accurate segmentation result of the intravascular ultrasound cross-section image, which specifically comprises the following steps: after the vascular membrane segmentation result of the (cross-section) intravascular ultrasound image is restored by using the longitudinal-axis intravascular ultrasound images at all angles, uniform down-sampling is carried out at intervals of 40 degrees in the radial direction, and then the down-sampling result is subjected to circle fitting to obtain the segmentation result. In order to better segment the lumen contour with irregular shape, an active contour model is introduced to carry out optimization, and a final cross-section image vascular membrane segmentation result is obtained. In the invention, a snake model is used, the iteration times are only 15 times, and the calculation burden is very small.
The data set used includes two parts, the first part is a public data set and 2175 frames of images for 10 patients are acquired, 250 frames are acquired for 7 patients, 150 frames are acquired for 2 patients, 125 frames are acquired for 1 patient, the golden standard corresponding to vascular membrane segmentation is labeled by four experienced clinicians (the golden standard of the data set is labeled by the physician every 5 frames, and 435 images contain vascular membrane labels). the remaining images are supplemented by the volunteer under the direction of the clinician before implementation according to the existing golden standard, the golden standard of the vertical axis image is generated by combining the golden standards of the original and the supplement), the images contain typical interferences such as MH bifurcation, bypass vessels, etc. the acquisition device used is Si5(Volcano corporation), a 20MHz Eagle Eye catheter is provided, the second part is a self-constructed data set acquired by the department of Cardin Kan of Shanghai university Hospit, a 2-patient collection frame is acquired, wherein another 1 st intraosseous acquisition is not performed simultaneously, the physician acquisition of the same time is performed on the same patient 351-3540 frames, and the patient is labeled by the physician(s) for two physician).
Three evaluation indexes are calculated based on the segmentation result and the corresponding gold standard: the degree of similarity between the segmentation results and gold standard is quantitatively evaluated by Housdorff Distance (HD), Jaccard Measure (JM), and Percent Area Difference (PAD).
HD is defined as the maximum distance of each point in the segmentation result profile to its nearest neighbor of the corresponding gold standard:
wherein: b isautoTo segment the resulting contours, BmanTo correspond to the contour of the gold standard, dist is the Euclidean distance between points p and q.
JM is defined as the intersection of the segmented result and the gold standard divided by the union of the two, and is used to measure the similarity between the segmented result and the gold standard:
wherein: zautoFor dividing the resulting area, ZmanArea is the Area corresponding to the gold standard.
PAD is defined as the percentage difference between the segmented result and the gold standard, and is used to measure the difference between the segmented result and the gold standard:
smaller HD and PAD, and larger JM mean that the segmentation result is closer to the gold standard.
In this embodiment, as shown in fig. 2(b), the lumen region in the longitudinal-axis intravascular ultrasound image reconstructed based on the cross-sectional intravascular ultrasound image sequence is clearly visible, and the condition of the blood vessel in the longitudinal-axis direction can be effectively displayed. As shown in fig. 3 and 6, the downsampling recovery strategy can effectively reduce the interference caused by bifurcations and by-pass vessels.
When the acquired blood vessel membrane ultrasonic cross-section image sequence is too long, the size of a reconstructed longitudinal axis image is increased, and an efficient segmentation algorithm is also required to be introduced from the viewpoint of time consumption of calculation. After obtaining the fuzzy clustering result of the vertical axis image, i.e. the result of the preliminary segmentation of the blood vessel membrane, there are some small interferences as shown in fig. 4(c), which need to be optimized. The small interference areas are small, and the areas formed by the small interference areas and the outer membrane structure can be effectively distinguished through the area size. Therefore, the small interference can be effectively removed by performing the on operation with the morphological filter of an appropriate size, as shown in fig. 4 (d).
As shown in fig. 7, 8 and 9, 10, to disclose the vascular membrane segmentation results and corresponding gold criteria of longitudinal and cross-sectional intravascular ultrasound images of the dataset and the self-constructed dataset, it can be observed that the automatic segmentation results are very close to the gold criteria.
From the evaluation index point of view, as shown in table 1, the evaluation index of the mid-adventitia and intima segmentation was comparable or superior to that of the other three published methods for the public data set. Particularly, in table 2, the evaluation indexes of the algorithm are better for intravascular ultrasound images with bifurcated vessels and bypass vessels than those of the other three methods. In addition, compared with other three methods, the algorithm realizes the vascular membrane segmentation of the cross-section image and the longitudinal axis image simultaneously, and has more obvious advantages compared with other three methods which only realize the vascular membrane segmentation of the cross-section image for evaluating the condition of the whole segment of the blood vessel. For the self-constructed data set, as shown in table 3, 3 intravascular ultrasound image sequences with different lengths of 2 patients are effectively segmented into the middle-adventitia, which indicates that the method is not easily affected by changes in the sequence lengths to a certain extent.
From the point of view of the computation time, the computation time of the present algorithm is advantageous compared to most other automatic or semi-automatic methods, as shown in table 4.
Table 1.
Table 2.
TABLE 3
TABLE 4
In summary, the embodiment is based on acquiring a complete (cross-sectional) intravascular ultrasound image sequence, and obtains a corresponding longitudinal-axis intravascular ultrasound image sequence through image processing and reconstruction. On the basis, the embodiment can realize the full-automatic segmentation of the vascular membrane of the intravascular ultrasound image in two modes of the longitudinal axis and the cross section, has higher automation, robustness and accuracy, and particularly realizes a better segmentation effect on the intravascular ultrasound image with bifurcated and bypass vessels.
The foregoing embodiments may be modified in many different ways by those skilled in the art without departing from the spirit and scope of the invention, which is defined by the appended claims and all changes that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims (9)
1. A vascular membrane accurate segmentation method based on multi-angle reconstruction of intravascular ultrasound image sequences is characterized in that corresponding intravascular ultrasound longitudinal axis images are obtained by reconstructing intravascular ultrasound cross-section image sequences, then the vascular membrane in the intravascular ultrasound longitudinal axis images is initially segmented by adopting an improved clustering method, and then the vascular membrane accurate segmentation result under the cross-section images is restored according to the initial segmentation result;
the intravascular ultrasound cross-sectional image sequence is a complete intravascular ultrasound cross-sectional image sequence acquired.
2. The method for precisely segmenting the vascular membrane according to claim 1, wherein the reconstruction is reflected on the corresponding longitudinal axis image by introducing the continuity of the vascular structure contained in the whole cross-sectional image sequence and by a sampling and interpolation technology.
3. The method for accurately segmenting the vascular membrane according to claim 1, wherein the preliminary segmentation is carried out by using a fuzzy clustering method on the reconstructed intravascular ultrasound longitudinal axis image to obtain a result of the rough segmentation of the vascular membrane; then optimizing through a morphological filter to obtain a vascular membrane preliminary segmentation result of the longitudinal axis intravascular ultrasound image as an initial contour; and then the longitudinal axis image is used for calculating an external force field, and the intravascular ultrasound longitudinal axis image vascular membrane preliminary segmentation result is obtained through active contour model optimization of a plurality of iterations.
4. The method for precisely segmenting the vascular membrane according to claim 1, wherein the restoration is: obtaining a vascular membrane segmentation result of the cross-section image through the inverse process of reconstructing the longitudinal axis image; then uniformly down-sampling on the cross section image to reduce the interference of bifurcations and bypass blood vessels; and finally, fitting a downsampling result circle to be smooth and optimizing the active contour model to obtain a vascular membrane accurate segmentation result of the intravascular ultrasound cross-section image.
5. The vascular membrane precise segmentation method according to claim 1 or 2, wherein the reconstruction specifically comprises: resampling the intravascular ultrasound cross-section image sequence frame by frame, namely sampling the intravascular ultrasound cross-section image sequence by pixel points in the center of the image angle by angle and diameter, interpolating sampling values of all cross-section images in the sequence at the same angle together, and reconstructing a longitudinal axis mode image corresponding to the angle.
6. The precise division method of the vascular membrane according to claim 2, wherein the preliminary division comprises the following steps:
① for longitudinal axis intravascular ultrasound images, firstly fuzzy clustering is used to obtain rough vascular membrane segmentation results;
②, performing open operation noise reduction optimization by using a circular morphological filter with the diameter of 9 pixels;
the fuzzy clustering, that is, the clustering algorithm based on partitioning, makes the similarity between objects partitioned into the same cluster maximum, and the similarity between different clusters minimum, and its objective function is:
wherein: u ═ Uik]Is a membership matrix, uikIs the degree of membership of the kth sample to the ith class, dik 2＝||xk-vi||2Is a sample xkAnd the cluster center (mean) viThe constraint condition is that the sum of the membership degrees of a certain sample to each cluster is 1, namely:
the clustering problem translates into:
7. The method for accurately segmenting the vascular membrane according to claim 6, wherein the fuzzy clustering comprises the following specific algorithm steps:
i) setting the clustering number c and a parameter m;
ii) initializing a membership matrix U (0), the sum of the row elements of U (0) being 1;
iii) calculating a new cluster center Vj;
iv) calculating a new membership matrix U;
v) when the termination condition is satisfied, | | U (k +1) -U (k) and | l ≦ e, stopping iteration, otherwise, continuing.
8. The method for precisely segmenting the vascular membrane according to claim 1 or 4, wherein the restoring specifically comprises: after restoring the vascular membrane segmentation result of the cross-section intravascular ultrasound image by using the longitudinal-axis intravascular ultrasound images at all angles, uniformly down-sampling at intervals of 40 degrees in the radial direction, and performing circle fitting on the down-sampling result to obtain a segmentation result; and finally, introducing a movable contour model for optimization to obtain a final cross-section image vascular membrane segmentation result.
9. A system for implementing the method of any preceding claim, comprising: the device comprises a reconstruction unit, a primary segmentation unit, a recovery unit and an optimized segmentation unit, wherein: the reconstruction unit obtains a corresponding intravascular ultrasound longitudinal axis image through frame-by-frame resampling processing according to the complete intravascular ultrasound cross-section image sequence, and the preliminary segmentation unit performs preliminary segmentation on the intravascular ultrasound longitudinal axis image by adopting an improved clustering method and obtains a segmentation result; the recovery unit extracts a blood vessel membrane preliminary segmentation result under the cross-section image from the preliminary segmentation result of the longitudinal axis image through an inverse reconstruction process; and the optimization segmentation unit optimizes the preliminary segmentation result of the vascular membrane of the cross-section image through downsampling, fitting and smoothing and the active contour model so as to obtain an accurate segmentation result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010243718.9A CN111445473B (en) | 2020-03-31 | 2020-03-31 | Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010243718.9A CN111445473B (en) | 2020-03-31 | 2020-03-31 | Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111445473A true CN111445473A (en) | 2020-07-24 |
| CN111445473B CN111445473B (en) | 2023-04-28 |
Family
ID=71652595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010243718.9A Active CN111445473B (en) | 2020-03-31 | 2020-03-31 | Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111445473B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112927212A (en) * | 2021-03-11 | 2021-06-08 | 上海移视网络科技有限公司 | OCT cardiovascular plaque automatic identification and analysis method based on deep learning |
| CN114881948A (en) * | 2022-04-26 | 2022-08-09 | 青岛埃米博创医疗科技有限公司 | Exploration box automatic vessel generation teaching aid method based on Monte Carlo algorithm |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2449080A1 (en) * | 2003-11-13 | 2005-05-13 | Centre Hospitalier De L'universite De Montreal - Chum | Apparatus and method for intravascular ultrasound image segmentation: a fast-marching method |
| CN104376549A (en) * | 2014-11-20 | 2015-02-25 | 华北电力大学（保定） | Intravascular ultrasound image and intravascular-OCT image fusing method |
| CN108335304A (en) * | 2018-02-07 | 2018-07-27 | 华侨大学 | A kind of aortic aneurysm dividing method of abdominal CT scan sequence image |
| CN109166083A (en) * | 2018-09-03 | 2019-01-08 | 哈尔滨工业大学 | A method of for removing underwater picture bubble noise |
| CN109345538A (en) * | 2018-08-30 | 2019-02-15 | 华南理工大学 | A kind of Segmentation Method of Retinal Blood Vessels based on convolutional neural networks |
| CN110223271A (en) * | 2019-04-30 | 2019-09-10 | 深圳市阅影科技有限公司 | The automatic horizontal collection dividing method and device of blood-vessel image |
| CN110889846A (en) * | 2019-12-03 | 2020-03-17 | 哈尔滨理工大学 | Diabetes retina image optic disk segmentation method based on FCM |
-
2020
- 2020-03-31 CN CN202010243718.9A patent/CN111445473B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2449080A1 (en) * | 2003-11-13 | 2005-05-13 | Centre Hospitalier De L'universite De Montreal - Chum | Apparatus and method for intravascular ultrasound image segmentation: a fast-marching method |
| CN104376549A (en) * | 2014-11-20 | 2015-02-25 | 华北电力大学（保定） | Intravascular ultrasound image and intravascular-OCT image fusing method |
| CN108335304A (en) * | 2018-02-07 | 2018-07-27 | 华侨大学 | A kind of aortic aneurysm dividing method of abdominal CT scan sequence image |
| CN109345538A (en) * | 2018-08-30 | 2019-02-15 | 华南理工大学 | A kind of Segmentation Method of Retinal Blood Vessels based on convolutional neural networks |
| CN109166083A (en) * | 2018-09-03 | 2019-01-08 | 哈尔滨工业大学 | A method of for removing underwater picture bubble noise |
| CN110223271A (en) * | 2019-04-30 | 2019-09-10 | 深圳市阅影科技有限公司 | The automatic horizontal collection dividing method and device of blood-vessel image |
| CN110889846A (en) * | 2019-12-03 | 2020-03-17 | 哈尔滨理工大学 | Diabetes retina image optic disk segmentation method based on FCM |
Non-Patent Citations (2)
| Title |
|---|
| 任子晖等: "基于Curvelet变换与遗传神经网络的血管分割算法" * |
| 方玉宏等: "肾脏结构与功能的光学监测研究进展" * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112927212A (en) * | 2021-03-11 | 2021-06-08 | 上海移视网络科技有限公司 | OCT cardiovascular plaque automatic identification and analysis method based on deep learning |
| CN112927212B (en) * | 2021-03-11 | 2023-10-27 | 上海移视网络科技有限公司 | OCT cardiovascular plaque automatic identification and analysis method based on deep learning |
| CN114881948A (en) * | 2022-04-26 | 2022-08-09 | 青岛埃米博创医疗科技有限公司 | Exploration box automatic vessel generation teaching aid method based on Monte Carlo algorithm |
| CN114881948B (en) * | 2022-04-26 | 2023-05-12 | 青岛埃米博创医疗科技有限公司 | Monte Carlo algorithm-based method for automatically generating vascular teaching aid by exploring box |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111445473B (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhou et al. | The detection and quantification of retinopathy using digital angiograms | |
| Pellot et al. | A 3D reconstruction of vascular structures from two X-ray angiograms using an adapted simulated annealing algorithm | |
| Wu et al. | Automatic detection of coronary artery stenosis by convolutional neural network with temporal constraint | |
| US7567696B2 (en) | System and method for detecting the aortic valve using a model-based segmentation technique | |
| CN108348206B (en) | Collateral flow modeling for non-invasive Fractional Flow Reserve (FFR) | |
| CN110448319B (en) | Blood flow velocity calculation method based on contrast image and coronary artery | |
| US11557072B2 (en) | Clustering algorithm-based multi-parameter cumulative calculation method for lower limb vascular calcification indexes | |
| CN113420826B (en) | Liver focus image processing system and image processing method | |
| CN112837306B (en) | Coronary artery disease lesion functional quantitative method based on deep learning and mesopic theory | |
| Xu et al. | An improved algorithm for vessel centerline tracking in coronary angiograms | |
| WO2012126070A1 (en) | Automatic volumetric analysis and 3d registration of cross sectional oct images of a stent in a body vessel | |
| CN113470137B (en) | IVOCT image guide wire artifact removing method based on gray-scale weighting | |
| US8934690B2 (en) | Method for processing vascular structure images | |
| CN111445473A (en) | Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction | |
| CN114693710A (en) | Blood vessel lumen intimal contour extraction method and device, ultrasonic equipment and storage medium | |
| Chandrinos et al. | Image processing techniques for the quantification of atherosclerotic changes | |
| CN113470060B (en) | Coronary artery multi-angle curved surface reconstruction visualization method based on CT image | |
| Sonka et al. | Lumen centerline detection in complex coronary angiograms | |
| Acosta-Mesa et al. | Cervical cancer detection using colposcopic images: a temporal approach | |
| US20230394654A1 (en) | Method and system for assessing functionally significant vessel obstruction based on machine learning | |
| CN112651984A (en) | Blood vessel lumen intimal contour extraction method and device, ultrasonic equipment and storage medium | |
| Moshfegh et al. | Development of an innovative technology to segment luminal borders of intravascular ultrasound image sequences in a fully automated manner | |
| CN116309346A (en) | Medical image detection method, device, equipment, storage medium and program product | |
| CN113592802B (en) | Mitral valve annular displacement automatic detection system based on ultrasonic image | |
| CN113223704B (en) | Auxiliary diagnosis method for computed tomography aortic aneurysm based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |