A high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images

By using high-precision labeling method and manual CT insertion algorithm in the vascular 3D reconstruction technology, the problem of poor accuracy of the vascular 3D model in the existing technology is solved, and high-precision vascular 3D reconstruction is achieved, which improves the accuracy of diagnosis and medical treatment. Understanding of the relationship between problems.

CN114882175BActive Publication Date: 2024-11-22ZHEJIANG SHUREN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210492476.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-28
Publication Date
2024-11-22
Estimated Expiration
2042-04-28

AI Technical Summary

Technical Problem

The models generated by the existing vascular 3D reconstruction technology are poor in accuracy and fail to effectively reflect the patient's actual vascular conditions, resulting in low diagnostic reference value.

Method used

The high-precision semi-automatic vascular 3D reconstruction algorithm based on CT images is adopted, and the high-precision labeling method and manual CT insertion algorithm are used to perform vascular inner wall labeling and hollow insertion processing to improve the accuracy of the vascular 3D model.

Benefits of technology

It significantly improves the accuracy and accuracy of vascular 3D reconstruction, enhances the medical staff's ability to judge the lesion site, assists in diagnosis and the intuitiveness of the patient's family in understanding the patient's condition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114882175B_ABST
    Figure CN114882175B_ABST
Patent Text Reader

Abstract

The present invention discloses a high-precision semi-automatic 3D reconstruction algorithm of blood vessels based on CT images, comprising the following steps: S1, creating a vascular CT image sequence; S2, using a high-precision annotation method to annotate a layer of a single or multiple blood vessels according to the created CT image sequence; S3, calculating the centroid distance xyDist, the offset xyXaxis of the x-axis, and the offset xyYaxis of the y-axis between each layer of CT slices in a selected directory; S4, sorting xyDist, xyXaxis, and xyYaxis from large to small, and performing slice insertion processing between CT slices with the largest deviation value. The present invention reconstructs blood vessels in 3D by using a high-precision annotation method and a manual CT slice insertion algorithm, significantly improving the accuracy of 3D reconstruction of blood vessels, greatly improving the accuracy of medical staff in judging the patient's lesion site based on the 3D reconstruction of blood vessels, and can assist in the diagnosis of the patient's symptoms. At the same time, it can allow the patient's family members to intuitively understand the patient's physical condition, which improves the doctor-patient relationship.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of biomedical image processing, and in particular relates to a high-precision semi-automatic blood vessel 3D reconstruction algorithm based on CT images. Background Art

[0002] Three-dimensional reconstruction of medical images is an interdisciplinary research field. It uses computer technology and graphic imaging technology to display human tissue represented by two-dimensional slices in a three-dimensional manner based on the two-dimensional slice sequence obtained clinically. Doctors can observe the patient's tissue from multiple angles and levels based on the reconstructed model, and even get an accurate and intuitive understanding of the size, location, shape, etc. of the lesion, thereby making an accurate diagnosis. Among them, vascular 3D reconstruction is the process of establishing a three-dimensional model of vascular tissue based on a two-dimensional slice sequence.

[0003] Currently, common vascular 3D reconstruction mainly adopts automated modeling, which automatically generates a vascular 3D model by transmitting the corresponding vascular two-dimensional slice sequence into the modeling system. This method does not take into account the patient's actual vascular condition, and the generated vascular 3D model has poor accuracy, which makes the actual reference value of the vascular 3D model low.

[0004] Therefore, in response to the above technical problems, it is necessary to provide a high-precision semi-automatic vascular 3D reconstruction algorithm based on CT images. Summary of the invention

[0005] The purpose of the present invention is to provide a high-precision semi-automatic blood vessel 3D reconstruction algorithm based on CT images to solve the above-mentioned problems.

[0006] In order to achieve the above purpose, the technical solution provided by an embodiment of the present invention is as follows:

[0007] A high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images, comprising the following steps:

[0008] S1. Create a vascular CT image sequence;

[0009] S2, using a high-precision annotation method to annotate a layer of a single or multiple blood vessels according to the created CT image sequence;

[0010] S3, calculating the centroid distance xyDist, the x-axis offset xyXaxis, and the y-axis offset xyYaxis between each CT slice in the selected directory;

[0011] S4, sorting xyDist, xyXaxis, and xyYaxis from large to small, and performing slice insertion between CT slices with the largest deviation value;

[0012] S5, performing 3D reconstruction according to the relevant data sets of each layer of CT slices in the current directory;

[0013] S6. The reconstructed 3D vascular model is judged. If there is a hole, the CT slice with the largest deviation value in S4 is inserted by a manual CT insertion algorithm. After the insertion is completed, the vascular 3D reconstruction is performed until the reconstructed 3D vascular model has no holes.

[0014] Furthermore, the high-precision labeling method in S2 mainly includes two parts: defining line segments, angles, and generating labeling points. By processing the two parts of defining line segments, angles, and generating labeling points, the labeling points on the inner wall of the blood vessel are generated, thereby improving the accuracy of subsequent 3D reconstruction of the blood vessel.

[0015] Furthermore, the manual CT slice insertion algorithm in S6 includes determining the change in blood vessel size, calculating the centroid coordinates of the slice insertion, and extracting the centroid coordinates. The manual CT slice insertion algorithm is used to insert the CT slice with a cavity, thereby improving the accuracy of the subsequently established 3D blood vessel model.

[0016] Furthermore, the manual CT slice insertion algorithm determines the trend of the blood vessel becoming thinner or thicker by judging the size of the slice area.

[0017] Furthermore, when the blood vessel becomes thinner, the insert is corroded, and when the blood vessel becomes thicker, the insert is expanded.

[0018] Furthermore, the centroid coordinates of the insert are calculated as follows:

[0019]

[0020] Compared with the prior art, the present invention has the following advantages:

[0021] The present invention reconstructs blood vessels in 3D by adopting a high-precision labeling method and a manual CT insert algorithm, thereby significantly improving the accuracy of 3D reconstruction of blood vessels, greatly improving the accuracy of medical staff's judgment of the location of patient lesions based on 3D reconstruction of blood vessels, and can assist in the diagnosis of patient symptoms. At the same time, it allows the patient's family members to intuitively understand the patient's physical condition, thereby improving the doctor-patient relationship. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 This is a flow chart of a high-precision semi-automatic blood vessel 3D reconstruction algorithm based on CT images in one embodiment of the present invention;

[0024] Figure 2 The original lung CT film in one embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of marking the inner diameter and outer diameter of a simulated blood vessel in one embodiment of the present invention;

[0026] Figure 4 It is a sequence diagram after simulating the inner diameter and outer diameter of a blood vessel in one embodiment of the present invention;

[0027] Figure 5 A marking schematic diagram of a high-precision marking method in one embodiment of the present invention;

[0028] Figure 6 Schematic diagram of a manual CT slice insertion algorithm in one embodiment of the present invention. DETAILED DESCRIPTION

[0029] The present invention will be described in detail below in conjunction with the various embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological or functional changes made by a person skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0030] The present invention discloses a high-precision semi-automatic 3D reconstruction algorithm of blood vessels based on CT images. Figure 1-Figure 6 As shown, the following steps are included:

[0031] S1. Create a vascular CT image sequence;

[0032] S2. using a high-precision annotation method to annotate a layer of a single or multiple blood vessels according to the created CT image sequence, wherein the high-precision annotation method mainly includes defining line segments, angles, and generating annotation points;

[0033] S3, calculating the centroid distance xyDist, the x-axis offset xyXaxis, and the y-axis offset xyYaxis between each CT slice in the selected directory;

[0034] S4, sorting xyDist, xyXaxis, and xyYaxis from large to small, and performing slice insertion between CT slices with the largest deviation value;

[0035] S5, performing 3D reconstruction according to the relevant data sets of each layer of CT slices in the current directory;

[0036] S6. The reconstructed 3D vascular model is judged. If there is a hole, the CT slice with the largest deviation value in S4 is inserted by a manual CT insertion algorithm. After the insertion is completed, the vascular 3D reconstruction is performed until the reconstructed 3D vascular model has no holes.

[0037] refer to Figure 5 As shown, the high-precision labeling method in S2 mainly includes two parts: defining line segments, angles, and generating labeling points. By processing the two parts of defining line segments, angles, and generating labeling points, the labeling points of the inner wall of the blood vessel are generated, thereby improving the accuracy of subsequent 3D reconstruction of the blood vessel.

[0038] Specifically, the high-precision labeling method includes the following steps:

[0039] The first step is to define the line segments and angles. Figure 5 It can be seen that the centroid coordinates of the blood vessel are set to O, and points A and B on the inner wall of the blood vessel are manually marked. The straight line connecting A and B generates the marked point D. In this way, the accuracy of the marked point D is low. The radii of OA and OB are set to rA and rB respectively, and the inclination angles are α+β respectively. According to the other adjacent marked points A1 and B1 of the blood vessel, a curve can be interpolated;

[0040] The second step is to generate the annotation point C. The curve generated in the first step passes through the vascular annotation points A and B, and the interpolation point is C. At this time, the curve intersects with the ∠AOB bisector at point C, the inclination angle = (α+β) / 2, and the radius = (2*rA / 3+rB / 3). This point C is the newly generated annotation point.

[0041] Specifically, the generated annotation point C should satisfy the following equation:

[0042] By generating new annotation points, the accuracy of annotating the inner and outer diameters of blood vessels on CT scans is improved, and the accuracy of 3D reconstruction of blood vessels is improved.

[0043] f(x)=a i +b i (xx i )+c i (xx i ) 2 +d i (xx i ) 3

[0044] Among them, i refers to a total of 4 point sequences: A1, A, B, and B1.

[0045] refer to Figure 6As shown, the manual CT slice insertion algorithm in S6 includes determining the change in blood vessel size, calculating the centroid coordinates of the slice insertion, and extracting the centroid coordinates. The manual CT slice insertion algorithm is used to insert the CT slice with a cavity, thereby improving the accuracy of the subsequently established 3D blood vessel model.

[0046] The manual CT slice insertion algorithm determines the trend of blood vessel thinning or thickening by judging the size of the slice area.

[0047] Specifically, when the blood vessel becomes thinner, the insert is corroded, and when the blood vessel becomes thicker, the insert is expanded.

[0048] The coordinates of the centroid of the insert are calculated as follows:

[0049]

[0050] Specifically, the manual CT slice insertion algorithm includes the following steps:

[0051] In the first step, assume that a new CT slice is to be inserted between CT slice 1 and CT slice 2, the centroid coordinates of CT1 and CT2 are (x1, y1) and (x2, y2), and the area of ​​the CT slice is calculated;

[0052] The second step is to determine whether the blood vessels are getting thinner or thicker based on the trend of area size changes;

[0053] Step 3, a. When the blood vessels become thinner: If a CT slice needs to be inserted between CT1 and CT2, it is named CT1mid, and CT1 should be eroded, that is, CT1mid = CT1erode (erosion treatment)

[0054] b. When the blood vessels become thicker: If a CT slice needs to be inserted between CT1 and CT2, it is named CT1mid. Otherwise, the expansion operation of CT1 becomes CT1mid=CT1dilate;

[0055] The fourth step is to use the formula Calculate the centroid coordinates of CT1mid;

[0056] The fifth step is to copy the centroid coordinates (xmid, ymid) of the extracted CT1mid to the centroid and manually insert the vascular CT slice.

[0057] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0058] In addition, it should be understood that although the present specification is described according to embodiments, not every embodiment contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation methods that those skilled in the art can understand.

Claims

1. A high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images, characterized in that: The following steps are involved: S1. Create a vascular CT image sequence; S2, using a high-precision annotation method to annotate a layer of a single or multiple blood vessels according to the created CT image sequence; S3, calculating the centroid distance xyDist, the x-axis offset xyXaxis, and the y-axis offset xyYaxis between each CT slice in the selected directory; S4, sorting xyDist, xyXaxis, and xyYaxis from large to small, and performing slice insertion between CT slices with the largest deviation value; S5, performing 3D reconstruction according to the relevant data sets of each layer of CT slices in the current directory; S6. The reconstructed 3D vascular model is judged. If there is a hole, the CT slice with the largest deviation value in S4 is inserted by a manual CT insertion algorithm. After the insertion is completed, the vascular 3D reconstruction is performed until the reconstructed 3D vascular model has no holes.

2. The high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images according to claim 1, characterized in that: The high-precision marking method in S2 mainly includes two parts: defining line segments and angles and generating marking points.

3. The high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images according to claim 1, characterized in that: The manual CT slice insertion algorithm in S6 includes determining the change in blood vessel size, calculating the slice centroid coordinates, and extracting the centroid coordinates.

4. The high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images according to claim 3, characterized in that: The manual CT slice insertion algorithm determines the trend of blood vessel thinning or thickening by judging the size of the slice area.

5. The high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images according to claim 4, characterized in that: When the blood vessel becomes thinner, the insert is corroded, and when the blood vessel becomes thicker, the insert is expanded.

6. The high-precision semi-automatic 3D blood vessel reconstruction algorithm based on CT images according to claim 5, characterized in that: The centroid coordinates of the insert are calculated as follows: