A method for extracting arteriovenous points based on CTP images

Abstract

The present invention discloses a method for extracting arteriovenous points based on CTP images, comprising the following steps: S1, obtaining a blood vessel region image; S2, using the lower 1 / 4 region of the brain region to screen out the vein region; S3, extracting the brightest point in the vein region as the vein point, and calculating the vein curve; S4, setting a time threshold according to the venous phase, and screening out the candidate artery region according to the time threshold; S5, performing spatial position constraints on the registration of the candidate artery region to obtain the artery region; S6, extracting the brightest point in the artery region as the artery point, and calculating the artery curve. The present invention realizes the highly automatic and high-accuracy extraction of arteries and veins in CTP images by combining clinical experience and physiological and anatomical knowledge.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to an arteriovenous point extraction method based on a CTP image. Background Art

[0002] CTP cerebral perfusion imaging is a very important imaging tool in the process of stroke diagnosis and treatment. It is an effective means for early diagnosis of cerebral infarction, guidance of emergency thrombolytic therapy and prognosis judgment. Cerebral perfusion parameter diagrams can accurately reflect hemodynamic parameters such as cerebral blood flow, cerebral blood volume, mean contrast agent transit time and contrast agent peak time, thereby evaluating cerebral perfusion status. For the extraction of arteriovenous positions in CTP cerebral perfusion images, the existing methods mainly use the information carried in CTP cerebral perfusion images for extraction and analysis to identify the positions of arteries and veins. However, since the image acquisition process is easily affected by external factors such as the environment and machine performance, this method is only based on the arteriovenous extraction method of the image itself, which is prone to false positives and the accuracy is not high enough. Summary of the invention

[0003] In order to solve the above problems, the present invention provides an arteriovenous point extraction method based on CTP images to improve the accuracy of arteriovenous extraction in CTP images.

[0004] The present invention adopts the following technical solutions:

[0005] A method for extracting arteriovenous points based on CTP images comprises the following steps:

[0006] S1, obtaining a blood vessel region image;

[0007] S2, using the lower 1 / 4 of the brain to screen out the venous area;

[0008] S3, extracting the brightest point in the vein area as the vein point, and calculating the vein curve;

[0009] S4, setting a time threshold according to the venous phase, and screening out the candidate arterial region according to the time threshold;

[0010] S5, performing spatial position constraints on the registration of the candidate artery region to obtain the artery region;

[0011] S6. Extract the brightest point in the arterial region as the arterial point, and calculate the arterial curve.

[0012] Furthermore, step S2 is specifically implemented by the following steps:

[0013] S21, selecting a region in the blood vessel region image with a time series change rate of 20-40 and a CT value higher than 150-200Hu as a candidate vein region;

[0014] S22, obtaining the time series ID corresponding to the maximum CT value of the candidate vein region;

[0015] S23, screen out images whose ID is greater than or equal to the time series ID and contains the candidate vein region, and filter out the region corresponding to the lower 1 / 4 region of the brain region in the image, so as to obtain the vein region.

[0016] Furthermore, when calculating the venous curve, a weighted average calculation is performed on an area near the brightest point in the venous area, so as to obtain the venous curve.

[0017] Furthermore, the step S4 is specifically as follows: obtaining the peak time of the vein point as a time threshold, filtering out all regions at and after the time threshold, and obtaining an artery candidate region.

[0018] Furthermore, the time threshold adopts the first 2 seconds of the venous phase time.

[0019] Furthermore, step S4 is specifically implemented by the following steps:

[0020] S41, obtaining an artery registration map, wherein the artery registration map is a clinical standard brain vascular region map containing an artery registration region, wherein the artery registration region includes a middle artery registration region and an anterior artery registration region;

[0021] S42. By comparing the artery registration map, the middle artery and the anterior artery regions are selected from the artery candidate regions as the artery regions.

[0022] Furthermore, when calculating the arterial curve, a weighted average calculation is performed on an area near the brightest point in the arterial area to obtain the arterial curve.

[0023] After adopting the above technical solution, the present invention has the following advantages compared with the background technology:

[0024] The method of the present invention combines the maximum CT value time series ID and the clinical blood vessel distribution characteristics of the lower 1 / 4 area of ​​the brain, namely the occipital lobe area, to filter out some areas when extracting veins, thereby obtaining the vein area; when extracting arteries, the time threshold is set as the screening standard of the candidate artery area in combination with the clinical CTP image acquisition experience, and then the middle artery and anterior artery areas are further screened out according to the physiological and anatomical knowledge, thereby determining the arterial area; by combining the clinical experience and the physiological and anatomical knowledge, the highly automatic and high-accuracy extraction of arteries and veins in the CTP image is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0027] Example

[0028] like Figure 1 As shown, a method for extracting arteriovenous points based on CTP images includes the following steps:

[0029] S1. Obtain a vascular region image; a region where the time series change rate is greater than 0 indicates that blood is flowing through, and can be considered as a vascular region.

[0030] S2, using the lower 1 / 4 of the brain to screen out the venous area;

[0031] The step S2 is specifically implemented by the following steps:

[0032] S21, selecting a region in the blood vessel region image with a time series change rate of 20-40 and a CT value higher than 150-200Hu as a candidate vein region;

[0033] S22, obtaining the time series ID corresponding to the maximum CT value of the candidate vein region;

[0034] S23, screen out images whose ID is greater than or equal to the time series ID and contains the candidate vein region, and filter out the region corresponding to the lower 1 / 4 region of the brain region in the image, so as to obtain the vein region.

[0035] S3, extracting the brightest point in the vein area as the vein point, and calculating the vein curve;

[0036] When calculating the venous curve, a weighted average calculation is performed on an area near the brightest point in the venous area, thereby obtaining the venous curve.

[0037] S4, setting a time threshold according to the venous phase, and screening out the candidate arterial region according to the time threshold;

[0038] The step S4 specifically includes: obtaining the peak time of the vein point as a time threshold, filtering out all regions after the time threshold and obtaining an artery candidate region.

[0039] The time threshold adopts the first 2 seconds of the venous phase time.

[0040] Step S4 is specifically implemented by the following steps:

[0041] S41, obtaining an artery registration map, wherein the artery registration map is a clinical standard brain vascular region map containing an artery registration region, wherein the artery registration region includes a middle artery registration region and an anterior artery registration region;

[0042] S42. By comparing the artery registration map, the middle artery and the anterior artery regions are selected from the artery candidate regions as the artery regions.

[0043] S5, performing spatial position constraints on the registration of the candidate artery region to obtain the artery region;

[0044] S6. Extract the brightest point in the arterial region as the arterial point, and calculate the arterial curve.

[0045] When calculating the arterial curve, a weighted average calculation is performed on an area near the brightest point in the arterial area to obtain the arterial curve.

[0046] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for extracting arteriovenous points based on CTP images, characterized in that: The steps include: S1, obtaining a blood vessel region image; S2, using the lower 1 / 4 of the brain to screen out the venous area; The step S2 is specifically implemented by the following steps: S21, selecting a region in the blood vessel region image with a time series change rate of 20-40 and a CT value higher than 150-200Hu as a candidate vein region; S22, obtaining the time series ID corresponding to the maximum CT value of the candidate vein region; S3, extracting the brightest point in the vein area as the vein point, and calculating the vein curve. When calculating the vein curve, the area near the brightest point in the vein area is taken for weighted average calculation, so as to obtain the vein curve; S4, setting a time threshold according to the venous phase, and screening out the candidate arterial region according to the time threshold; The step S4 specifically includes: obtaining the peak time of the vein point as a time threshold, filtering out all regions after the time threshold and obtaining an artery candidate region; S5, performing spatial position constraints on the registration of the candidate artery region to obtain the artery region; S6. Extract the brightest point in the arterial region as the arterial point, and calculate the arterial curve.

2. The method for extracting arteriovenous points based on CTP images according to claim 1, characterized in that: The time threshold adopts the first 2 seconds of the venous phase time.

3. The method for extracting arteriovenous points based on CTP images according to claim 1, characterized in that: Step S4 is specifically implemented by the following steps: S41, obtaining an artery registration map, wherein the artery registration map is a clinical standard brain vascular region map containing an artery registration region, wherein the artery registration region includes a middle artery registration region and an anterior artery registration region; S42. By comparing the artery registration map, the middle artery and the anterior artery regions are selected from the artery candidate regions as the artery regions.

4. The method for extracting arteriovenous points based on CTP images according to claim 3, characterized in that: When calculating the arterial curve, a weighted average calculation is performed on an area near the brightest point in the arterial area to obtain the arterial curve.

Images