阅读说明：本技术 一种对cta冠状动脉图像进行数据处理预测ffr的方法 (Method for predicting FFR (fringe field response) by processing data of CTA (computed tomography angiography) coronary artery image ) 是由 刘有军 刘金城 毛伯* 毛伯䶮 冯月 于 2019-09-04 设计创作，主要内容包括：一种对CTA冠状动脉图像进行数据处理预测FFR的方法,属于图像处理和建立模型领域,是一种基于CTA图像无创预测FFR的一种新方法。获得冠心病患者的CTA图像之后,进行冠脉三维模型重建；测得冠状动脉各分支血管的体积(V)；通过冠状动脉的血管长度确定各血管的灌注面积(M)；根据CTA图像获取的数据来计算出单支血管上的V/M值；冠状动脉单支血管的体积质量比可以无创的预测与FFR之间的关系,单支血管的V/M阈值16.92等同于FFR为0.8,当单支血管V/M大于该阈值时,FFR大于0.8,小于该阈值时FFR小于0.8。该方法可以建立一种准确的预测模型来反映与FFR之间的关系。(A method for predicting FFR by data processing on a CTA coronary artery image belongs to the field of image processing and model building, and is a novel method for non-invasively predicting FFR based on the CTA image. After a CTA image of a patient with coronary heart disease is obtained, reconstructing a coronary artery three-dimensional model; measuring the volume (V) of each branch vessel of the coronary artery; determining the perfusion area (M) of each vessel by the vessel length of the coronary artery; calculating V/M values on single branch blood vessels according to data acquired by the CTA image; the relation between the volume-mass ratio of the coronary artery single branch vessel and the FFR can be non-invasively predicted, the V/M threshold value 16.92 of the single branch vessel is equivalent to the FFR of 0.8, when the V/M of the single branch vessel is larger than the threshold value, the FFR is larger than 0.8, and when the V/M of the single branch vessel is smaller than the threshold value, the FFR is smaller than 0.8. The method can establish an accurate prediction model to reflect the relationship with the FFR.)

1. A method of data processing a CTA coronary image to predict FFR, comprising the steps of:

1.1 obtaining a CTA coronary artery image of a coronary heart disease patient, performing three-dimensional reconstruction, obtaining a three-dimensional model of the coronary artery of the patient, and particularly obtaining a three-dimensional model in an STL format;

1.2 measurement of coronary vessel branch volume (V);

1.3 determination of perfusion area (M) of single branch vessel of coronary artery;

1.4 finding the V/M value of the coronary artery single branch vessel, and the V/M value can be used for predicting FFR.

2. A method for predicting FFR from data processing of a CTA coronary image as recited in claim 1, wherein said step 1.1 comprises:

2.1 CTA images of coronary heart disease patients were obtained from the hospital.

2.2, introducing a DICOM (computed tomography angiography) format of a CTA image obtained from a hospital into reconstruction software;

2.3 after the image is imported into the software Mimics, selecting a proper image threshold value for selecting an interested area;

2.4 after the threshold is selected, selecting dynamic region growth, concentrating pixels with similar properties to form a region, and separating coronary artery and other soft tissue structures;

2.5 after threshold selection and dynamic region increase, starting to reconstruct the coronary artery structure of the region of interest and obtaining the STL format of the three-dimensional model of the patient.

3. The method of claim 1 in which step 1.2 comprises:

3.1, introducing the STL format of the coronary artery structure obtained in the step 2.5 into a model processing software Geomagic studio 12 version, and measuring the volume of the coronary artery three-dimensional model;

3.2 the coronary artery structure is divided into three parts by the segmentation function in the model segmentation software: right Coronary Artery (RCA), Left Anterior Descending (LAD), Left Circumflex (LCX);

3.3 the volume of the three branches of the coronary artery was measured.

4. A method for data processing a CTA coronary image to predict FFR as recited in claim 1, wherein said step 1.3 comprises:

4.1, reintroducing the STL in the step 2.5 into reconstruction software Mimics;

4.2 fitting the vessel center lines of three branches including the coronary artery RCA, the coronary artery LAD and the coronary artery LCX by using the function of fitting the center lines by using the Mimics;

4.3 measuring the length Lrca, Llad, Llcx and total coronary artery length of each branch through the fitted coronary artery central line;

and 4.4, according to a myocardial mass distribution formula M-Mtot L0/L, wherein M is the myocardial mass corresponding to the single branch vessel, Mtot is the total myocardial mass of the left ventricle, L0 is the length of the corresponding single branch vessel, and L is the total length of the coronary artery, so that the myocardial mass corresponding to the single branch vessel can be obtained.

5. The method of claim 1 in which step 1.4 comprises:

5.1 calculating the ratio of the volume (V) of the single coronary artery blood vessel and the myocardial mass (M) to obtain the V/M value of the single coronary artery blood vessel.

6. A method of data processing a CTA coronary image to predict FFR as claimed in claim 1 wherein V/M is equal to FFR 0.8 bounded by a median of 16.92, with V/M greater than 16.92 being equal to FFR; a smaller V/M value at a V/M value of less than 16.92 corresponds to a smaller FFR.

The technical field is as follows:

the invention belongs to the field of image processing and model building, and relates to a method for predicting FFR (fringe field regression) by processing data of a CTA (computed tomography angiography) coronary artery image.

Background art:

clinical FFR (fractional flow reserve) has great limitations, and this method is invasive and expensive to judge myocardial ischemia. Research shows that the overall coronary artery volume (V) and the left ventricular myocardium mass (M) are related to the FFR value, and the relationship with the FFR value is predicted by calculating the overall V/M, and reflects the supply and demand relationship between the overall coronary artery volume and the left ventricular myocardium. However, since the FFR value is associated with each blood vessel and the measurement is invasive, a more accurate method of data processing the CTA coronary images should be established to non-invasively reflect the relationship to the FFR value.

The invention content is as follows:

compared with the prior method, the method has the advantages that the supply-demand relation on the coronary artery single branch vessel is deeply considered, and the relation between the FFR value and the blood vessel volume and the corresponding myocardial mass is reflected or predicted only by processing the data of the CTA coronary artery image.

The invention discloses a method for predicting FFR (fractional Fourier transform) by data processing of a CTA (computed tomography angiography) coronary artery image, which is characterized by comprising the following steps of:

1.1 obtaining a CTA coronary artery image of a coronary heart disease patient, performing three-dimensional reconstruction, obtaining a three-dimensional model of the coronary artery of the patient, and particularly obtaining a three-dimensional model in an STL format;

1.2 measurement of coronary vessel branch volume (V);

1.3 determination of perfusion area (M) of single branch vessel of coronary artery;

1.4 finding the V/M value of the coronary artery single branch vessel, and the V/M value can be used for predicting FFR.

Further step 1.1 comprises:

2.1 CTA images of coronary heart disease patients were obtained from the hospital.

2.2, introducing a DICOM (computed tomography angiography) format of a CTA image obtained from a hospital into reconstruction software;

2.3 after the image is imported into the software Mimics, selecting a proper image threshold value for selecting an interested area;

2.4 after the threshold is selected, selecting dynamic region growth, concentrating pixels with similar properties to form a region, and separating coronary artery and other soft tissue structures;

2.5 after threshold selection and dynamic region increase, starting to reconstruct the coronary artery structure of the region of interest and obtaining the STL format of the three-dimensional model of the patient.

Further said step 1.2 comprises:

3.1, introducing the STL format of the coronary artery structure obtained in the step 2.5 into a model processing software Geomagic studio 12 version, and measuring the volume of the coronary artery three-dimensional model;

3.2 the coronary artery structure is divided into three parts by the segmentation function in the model segmentation software: right Coronary Artery (RCA), Left Anterior Descending (LAD), Left Circumflex (LCX);

3.3 the volume of the three branches of the coronary artery was measured.

Further step 1.3 includes:

4.1, reintroducing the STL in the step 2.5 into reconstruction software Mimics;

4.2 fitting the vessel center lines of three branches including the coronary artery RCA, the coronary artery LAD and the coronary artery LCX by using the function of fitting the center lines by using the Mimics;

4.3 measuring the length Lrca, Llad, Llcx and total coronary artery length of each branch through the fitted coronary artery central line;

4.4 according to a myocardial mass distribution formula M ═ Mtot × L0/L, wherein M is the myocardial mass corresponding to the single branch vessel to be obtained, Mtot is the total myocardial mass of the left ventricle, L0 is the length of the corresponding single branch vessel, and L is the total length of the coronary artery, so that the myocardial mass corresponding to the single branch vessel can be obtained;

further preferably, said step 1.4 comprises:

5.1 calculating the ratio of the volume (V) of the single coronary artery blood vessel and the myocardial mass (M) to obtain the V/M value of the single coronary artery blood vessel.

The invention uses a median 16.92 as a boundary, namely V/M is equal to FFR is equal to 0.8, and when V/M is larger than 16.92, the larger V/M is equal to the larger FFR; a smaller V/M value at a V/M value of less than 16.92 corresponds to a smaller FFR.

The method of the present invention can further determine whether to invasively detect the FFR by predicting the approximate direction of the FFR in advance.

Description of the drawings:

FIG. 1: the structure and flow diagram of the method of the present invention;

Detailed Description

The present invention will be further illustrated and verified by the following examples, but the present invention is not limited to the following examples.