CN110428420B - Method, apparatus and medium for determining flow information of coronary artery based on coronary artery CT angiography image of patient - Google Patents

Method, apparatus and medium for determining flow information of coronary artery based on coronary artery CT angiography image of patient Download PDF

Info

Publication number
CN110428420B
CN110428420B CN201910837899.5A CN201910837899A CN110428420B CN 110428420 B CN110428420 B CN 110428420B CN 201910837899 A CN201910837899 A CN 201910837899A CN 110428420 B CN110428420 B CN 110428420B
Authority
CN
China
Prior art keywords
image
flow
vessel
perfusion
interest
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.)
Active
Application number
CN201910837899.5A
Other languages
Chinese (zh)
Other versions
CN110428420A (en
Inventor
武丹
李育威
李俊环
张洪凯
白军杰
尹游兵
曹坤琳
宋麒
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Keya Medical Technology Corp
Original Assignee
Shenzhen Keya Medical Technology Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Keya Medical Technology Corp filed Critical Shenzhen Keya Medical Technology Corp
Publication of CN110428420A publication Critical patent/CN110428420A/en
Application granted granted Critical
Publication of CN110428420B publication Critical patent/CN110428420B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • 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
    • G06T2207/20081—Training; Learning
    • 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
    • G06T2207/20084—Artificial neural networks [ANN]
    • 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
    • G06T2207/30104—Vascular flow; Blood flow; Perfusion

Abstract

The present disclosure relates to a method, apparatus, and medium for determining flow information of coronary arteries based on a coronary CT angiographic image of a patient. The method comprises the following steps: acquiring a CCTA image of the patient; extracting arterial features of the vessel of interest based on the acquired CCTA image; extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images; predicting flow information of the coronary artery using a trained prediction model based on at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest. The method can obtain more accurate and patient-specific coronary flow information from the CCTA image, further obtain more reliable virtual FFR prediction based on the flow information, and provide quantitative patient-specific flow physiological information for doctors through conventional CCTA examination.

Description

Method, apparatus and medium for determining flow information of coronary artery based on coronary artery CT angiography image of patient
Cross Reference to Related Applications
This application claims priority from U.S. provisional application No. 62/726,989, filed on 5.9.2018, the entire contents of which are incorporated herein by reference.
Technical Field
The present disclosure relates generally to medical image processing and analysis. More particularly, the present disclosure relates to methods, apparatus, and media for determining flow information of coronary arteries based on medical images of coronary arteries of a patient.
Background
Generally, the primary diagnostic method for coronary artery disease relies on invasive measurement of Fractional Flow Reserve (FFR). More recently, non-invasive methods such as computational fluid dynamics and depth learning techniques have been applied to Coronary Computed Tomography Angiography (CCTA) images to derive FFR (also referred to as virtual FFR). Flow information of the coronary arteries of the patient, such as inlet flow rate and flow ratio distribution, can provide important intermediate information for these new methods of predicting virtual FFR, if not directly available for diagnosis, such as patient-specific boundary conditions or flow characteristics. In particular, the prediction method of virtual FFR typically relies on these intermediate information to make the prediction. However, these intermediate information are typically estimated using empirical relationships derived from simple theoretical equations or statistical correlations, resulting in estimated flow information for the coronary arteries of the patient that is not patient-specific (i.e., does not exactly coincide with the patient's true condition), upon which the virtual FFR is derived, significantly impacting the prediction accuracy of the virtual FFR.
The present disclosure is provided to overcome the above technical drawbacks in the prior art.
Disclosure of Invention
The present disclosure is directed to a method, apparatus and medium for determining flow information of coronary arteries based on a CCTA image of a patient, which can obtain more accurate and patient-specific flow information of coronary arteries from the CCTA image, further can obtain more reliable virtual FFR prediction therefrom, and can also provide quantitative patient-specific flow physiological information to a doctor by a conventional CCTA examination.
In one aspect, the present disclosure provides a method of determining flow information of coronary arteries based on a Coronary Computed Tomography Angiography (CCTA) image of a patient, the method comprising: acquiring a CCTA image of the patient; extracting arterial features of the vessel of interest based on the acquired CCTA image; extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images; predicting flow information of the coronary artery using a trained prediction model based on at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest.
In another aspect, the present disclosure provides a method of determining flow information of coronary arteries based on perfusion Computed Tomography (CT) images of a patient, the method comprising: acquiring a sequence of perfusion CT images of the patient, the perfusion CT images including coronary arteries and a heart; determining a Volumetric Perfusion CT (VPCT) image based on the sequence of perfusion CT images of the patient, the VPCT image showing myocardial blood flow for each voxel therein; acquiring a Coronary Computed Tomography Angiography (CCTA) image of the patient and a segmentation result of a perfusion region of a left ventricle in the CCTA image; obtaining a partition result of a corresponding perfusion area of a left ventricle in the VPCT image by registering the VPCT image and the CCTA image based on the partition result of the perfusion area of the left ventricle in the CCTA image; and determining flow information of the coronary arteries based on a result of the division of the corresponding perfusion region of the left ventricle in the VPCT image.
In yet another aspect, the present disclosure provides an apparatus for determining flow information of coronary arteries based on a Coronary Computed Tomography Angiography (CCTA) image of a patient, the apparatus comprising: an interface configured to acquire CCTA images of the patient; and a processor configured to: extracting arterial features of the vessel of interest based on the acquired CCTA image; extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images; estimating flow information of the coronary arteries using a trained predictive model based on at least a subset of the extracted arterial and myocardial features of the vessel of interest.
In yet another aspect, the present disclosure provides a non-transitory computer readable medium having stored thereon instructions that, when executed by a processor, implement a method of determining flow information of coronary arteries based on Coronary Computed Tomography Angiography (CCTA) images of a patient according to various embodiments of the present disclosure, and/or implement a method of determining flow information of coronary arteries based on perfusion Computed Tomography (CT) images of a patient according to various embodiments of the present disclosure.
With the method, apparatus and medium for determining flow information of coronary arteries based on CCTA images of patients according to various embodiments of the present disclosure, more accurate and patient-specific flow information of coronary arteries can be obtained from CCTA images, virtual FFR prediction benefits from which prediction results can be more reliable and accurate, and quantitative patient-specific flow physiological information can be provided to a doctor by a conventional CCTA examination.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
In the drawings, which are not necessarily drawn to scale, like reference numerals may depict like parts in different views. Like numbers with letter suffixes or different letter suffixes may represent different instances of similar components. The drawings illustrate various embodiments, generally by way of example and not by way of limitation, and together with the description and claims, serve to explain the disclosed embodiments. Such embodiments are illustrative and not intended to be exhaustive or exclusive embodiments of the present method, apparatus, system, or non-transitory computer-readable medium having stored thereon instructions for carrying out the method.
Fig. 1 shows a flow diagram of a method of determining flow information of coronary arteries based on CCTA images of a patient according to an embodiment of the present disclosure;
fig. 2 illustrates a sub-flow of extracting arterial features of a vessel of interest based on a CCTA image of a patient according to an embodiment of the present disclosure;
fig. 3 illustrates a sub-flow of extracting corresponding myocardial features of a vessel of interest based on a CCTA image of a patient according to an embodiment of the present disclosure;
fig. 4(a) and 4(b) show diagrams of 3D models of coronary artery trees reconstructed based on CCTA images of patients, where the partitioning of the perfusion region on the left ventricle is shown and the terminal branches are identified, according to embodiments of the present disclosure;
FIG. 5 illustrates a flow chart of a method of determining flow information of coronary arteries based on perfusion Computed Tomography (CT) images of a patient according to an embodiment of the present disclosure;
6(a) -6 (c) show an axial, sagittal and coronal view, respectively, of a Volumetric Perfusion CT (VPCT) image determined from a sequence of perfusion CT images of a patient;
fig. 7 illustrates a configuration of an apparatus for determining flow information of coronary arteries based on CCTA images of a patient according to an embodiment of the present disclosure;
fig. 8 shows a block diagram of an apparatus for determining flow information of coronary arteries based on CCTA images of a patient according to an embodiment of the present disclosure;
fig. 9 shows a configuration of an apparatus for determining flow information of coronary arteries based on a perfusion CT image of a patient according to an embodiment of the present disclosure.
Detailed Description
In the following, the technical term "blood flow" (e.g. "blood flow" in "myocardial blood flow", "coronary blood flow") denotes the volume of blood flowing per unit time in the respective organ tissue. The technical term "flow rate" also denotes the volume of fluid flowing through per unit of time. The technical term "CCTA image" herein may refer to images of CCTA cross-sections (e.g., axial, sagittal, coronal, etc.) acquired by a coronary computed tomography angiography imaging apparatus, as well as to 3D images resulting from post-processing, e.g., reconstruction, of such cross-section images. Similarly, the term "perfusion CT image" in the present context may also refer to perfusion CT cross-sectional (e.g. axial, sagittal, coronal, etc.) images acquired by a perfusion computed tomography imaging apparatus, as well as to 3D images resulting from post-processing, e.g. reconstruction, of these cross-sectional images. The technical term "VPCT image" may denote a 3D perfusion CT image obtained by calculating a blood flow based on a sequence of perfusion CT images and showing a myocardial blood flow per voxel therein.
In addition, the operation of "training" on the "prediction model" in this document differs depending on the "prediction model", for example, for a prediction model composed of a neural network, "training" may refer to a training process of adjusting parameters of the neural network so that an objective function is maximized using training data, for example, using a stochastic gradient descent method; for a predictive model consisting of an explicit formula, "training" may refer to the process of using training data to determine parameters (e.g., constants) in the formula. The technical term "terminal branch" denotes a branch of a blood vessel without a sub-branch.
Fig. 1 shows a flowchart of a method 100 of determining flow information of coronary arteries based on CCTA images of a patient according to an embodiment of the present disclosure. As shown in fig. 1, at step 101, a CCTA image of the patient is acquired. At step 102, arterial features of the vessel of interest may be extracted based on the acquired CCTA images. And at step 103, corresponding myocardial features of the vessel of interest may be extracted based on the acquired CCTA images. Arterial features may be obtained from the vessel of interest itself and describe characteristics about the vessel of interest; corresponding myocardial features of the vessel of interest may be obtained from the myocardium of the perfused region of the vessel of interest and described with respect to characteristics of the myocardium. Note that while step 102 and step 103 are performed sequentially in fig. 1, they may be performed together or in a different order, so long as they are performed prior to step 104. Then, in step 104, the flow information of the coronary artery may be predicted using a trained prediction model based on at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest. The raw information of the method 100 is derived from the patient's own CCTA image, which feeds back the patient-specific coronary image information accurately and with a higher resolution; by focusing on the feature extraction of the concerned blood vessel, compared with the feature extraction of all blood vessels, the method remarkably reduces the workload and considers the accuracy rate of prediction; in addition, the method can predict the coronary artery based on the blood vessel characteristics reflecting the structural characteristics of the coronary artery and the myocardial characteristics reflecting the functional characteristics of the coronary artery, so that the feature information based on the prediction is more comprehensive, the prediction result is more accurate, and the method is more suitable for the structure and the function of the coronary artery special for a patient.
In some embodiments, the vessel of interest for which the arterial features and corresponding myocardial features are extracted in steps 102 and 103 may depend on the flow information and predictive model of the coronary arteries to be determined.
For example, if flow information is to be determined for the left anterior descending branch (LAD) and all its sub-branches, the vessel of interest may be defined accordingly as the left anterior descending branch (LAD) and all its sub-branches; if flow information of the Left Coronary Artery (LCA) and all its sub-branches is to be determined, the vessel of interest can be defined as LCA and all its sub-branches accordingly, which can reduce the range of the vessel of interest involved in the calculation and reduce the workload of extracting arterial features and corresponding myocardial features. For example, if the flow information of the left anterior descending branch (LAD) and all its subbranches is to be determined, the blood vessel of interest may also be expanded from the left anterior descending branch (LAD) and all its subbranches to include other branches and/or subbranches of the coronary artery, so that the interaction of the other branches and/or subbranches with the branch and/or subbranch whose flow information is to be determined is referred to while reducing the scope of the blood vessel of interest involved in the calculation and reducing the workload of extracting arterial features and corresponding myocardial features, so that the prediction result is more accurate. For another example, in a case where the flow distribution ratio of each blood vessel in the coronary artery tree is to be determined, the blood vessel of interest may be defined as all the respective blood vessels. For another example, the blood vessel of interest may be defined to include each terminal branch, and thus, the flexibility of the myocardial feature extraction may be improved. Specifically, regardless of the flow information of which branch artery of the coronary artery is to be determined, the corresponding myocardial characteristics of all the terminal branches (such as, but not limited to, the myocardial characteristics of the perfusion region of each terminal branch, the myocardial characteristics of the extension region of each terminal branch including the perfusion region thereof, etc.) may be screened and integrated, for example, the corresponding myocardial characteristics of each terminal branch belonging to the branch artery may be screened and integrated, so as to obtain the myocardial characteristics of the perfusion region of the branch artery as the corresponding myocardial characteristics of the branch artery.
In some embodiments, the flow information of the coronary arteries to be determined is output as a prediction model, with different prediction models having different inputs. If the total inlet flow rate of the LAD is to be determined, and the prediction model employed is expressed as:
Q＝Q0Mbequation (1)
Wherein Q represents the flow rate of the vessel of interest, M is the myocardial mass of the perfused region of the vessel of interest in the left ventricle, Q0And b represents the current state of the vessel of interest of the patientCan be defined as the LAD itself, or belong to each terminal branch of the LAD. For example, in the case where the blood vessel of interest is defined as the LAD itself, the myocardial blood flow Q of the perfused area of the LAD may be directly calculated as the total inlet flow rate of the LAD according to equation (1); in the case where the vessel of interest is defined as belonging to each distal branch of the LAD, the myocardial blood flow of the perfusion region of each distal branch may be calculated and summed according to equation (1) to obtain the myocardial blood flow of the perfusion region of the LAD as the total inlet flow rate of the LAD. The formula (1) is consistent with the biological anisotropic scaling law, and when the method is used as a prediction model of the flow rate of each blood vessel of the coronary artery, the calculation and training are simple, the workload is low, and the prediction accuracy is high.
In some embodiments, equation (1) may also be extended to Q ═ f (V), where V represents a subset of the feature vectors formed by the extracted arterial features of the vessel of interest and the corresponding myocardial features, and f (V) represents a function of V, which may be expressed as an explicit equation such as equation (1), or generalized to be understood as a learning network.
In some embodiments, the learning network, such as but not limited to a linear regression model, a multi-layer neural network, or the like, takes as input at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest. For example, the learning network may define which branch artery's arterial features and corresponding myocardial features are used as inputs, and the vessel of interest may be defined according to the input definition of the learning network.
In some embodiments, a feature vector may be constructed based on the extracted arterial features and corresponding myocardial features of the vessel of interest, but the feature parameters in the feature vector need not be all used for prediction at times, and in particular, a subset thereof may be selected based on the prediction model and the flow information to be predicted. For example, when the flow rate of a certain segment of branch is predicted by using the above formula (1) as a prediction model, the artery characteristics of the segment of branch may not be considered, and only the myocardial characteristics of the perfusion region of the segment of branch may be considered. For another example, when a learning network is used to construct the prediction model, the learning network may specify inputs such as which arterial feature parameters and which myocardial feature parameters of which branch and its sub-branches are to be predicted, and a corresponding subset of the feature parameters in the feature vector may be considered in the prediction.
The following describes a sub-flow of extracting arterial features of a blood vessel in detail.
As shown in fig. 2, the sub-flow 200 of extraction of arterial features of a vessel of interest begins with step 201, performing arterial segmentation on the acquired CCTA image and extracting the centerline of the coronary artery tree. Then, at step 202, arterial features of a vessel of interest are extracted from the CCTA image based on the extracted centerline. The arterial characteristic may comprise at least one of radius related information, length related information, volume related information, intensity related information and intraluminal attenuation gradient related information of the vessel of interest.
In some embodiments, the information related to the radius of the vessel of interest may comprise, for example, an average lumen area, which describes a radial characteristic of the vessel of interest. The averaging of the "average lumen area" may be performed for the vessel of interest itself, or for the vessel of interest and its upstream vessels, or for the vessel of interest and its downstream vessels. In some embodiments, absolute or normalized values of these parameters may be used.
In some embodiments, the length-related information of the vessel of interest may be the length of the vessel of interest itself, or the cumulative length of the vessel of interest and its upstream vessels, or the cumulative length of the vessel of interest and its downstream vessels. In some embodiments, absolute or normalized values of these parameters may be used.
In some embodiments, the volume related information of the blood vessel of interest may be a volume of the blood vessel of interest itself, or a cumulative volume of the blood vessel of interest and its upstream blood vessels, or a cumulative volume of the blood vessel of interest and its downstream blood vessels. In some embodiments, absolute or normalized values of these parameters may be used.
In some embodiments, the intensity-related information of the vessel of interest may comprise, for example, a mean lumen intensity, which describes an intensity characteristic of the vessel of interest. The averaging of the "mean lumen intensity" may be performed for the vessel of interest itself, or for the vessel of interest and its upstream vessels, or for the vessel of interest and its downstream vessels. In some embodiments, absolute or normalized values of these parameters may be used.
In some embodiments, the intraluminal decay gradient (TAG) of all vessels, i.e. the linear regression coefficient (which may be the amount of change in CT value per unit longitudinal length from the vessel entrance, i.e. the amount of change in HU) between the radioactive decay value at a location within the lumen of a vessel and the longitudinal distance (distance along the centerline) from the entrance of that vessel to that location, may be calculated. The TAG of all the vessels reflects the size distribution information of the individual vessels along the centerline in the coronary artery tree, whereby the flow ratio distribution in the entire coronary artery tree can be determined based on the TAG of all the vessels as the flow information of the coronary artery. In some embodiments, the flow ratio distribution within the entire coronary artery tree may also be determined as flow information of the coronary arteries based on other morphological laws of all vessels except TAG.
The following describes a sub-flow of extracting corresponding myocardial features of a blood vessel of interest.
As shown in fig. 3, the sub-flow 300 of extraction of corresponding myocardial features of a vessel of interest begins at step 301 with performing left ventricular segmentation on the acquired CCTA image. In step 302, perfusion regions of respective vessels of interest are demarcated on the left ventricle, as shown in fig. 4(a) and 4(b), and the perfusion region PT n, n of the end branch EB n is any natural number from 1 to 7. In some embodiments, the division of the perfusion region PT n may be achieved in various ways. For example, the perfusion regions can be assigned to neighboring vessels as perfusion regions for the vessels based on the closest geodetic distance to the vessel from the respective partition of the left ventricle. Alternatively, the perfusion region may also be assigned based on the law of the dissimilarity scale between the cumulative coronary length and the myocardial mass using a stem-crown model of the coronary tree.
Next, in step 303, a myocardial feature is determined for the perfusion region PT n of each vessel of interest as a corresponding myocardial feature of the vessel of interest. For example, the myocardium characteristics may include at least one of myocardium mass and intensity related information. Myocardial mass can be obtained by integrating the voxel volumes within the perfusion region to obtain the total volume, then multiplying by myocardial density. And the intensity-related information describes an intensity characteristic of the perfusion region, which may include at least one of an average intensity, a maximum intensity, and a minimum intensity within the perfusion region; in some embodiments, absolute or normalized values of these parameters may be used.
Note that the perfusion regions PT n may correspond one-to-one to the end branches EB n, and the left ventricle may divide the perfusion regions PT n based on the respective end branches EB n, as shown in fig. 4(a) and 4(b), but is not limited thereto, and for example, the perfusion regions may be divided based on other levels of branches, such as the perfusion regions of the respective Left Coronary Artery (LCA) and Right Coronary Artery (RCA), or the perfusion regions of the respective LAD, Left Circumflex (LCX), and RCA.
Fig. 5 shows a flowchart of a method 500 of determining flow information of coronary arteries based on a perfusion CT image of a patient according to an embodiment of the present disclosure. In step 501, a sequence of perfusion CT images of the patient is acquired, the perfusion CT images including coronary arteries and a heart. At step 502, a 3D Volumetric Perfusion CT (VPCT) image is determined based on the sequence of perfusion CT images of the patient, e.g. by a maximum slope method or a deconvolution method or the like, with an axial, sagittal and coronal view as shown in fig. 6(a), 6(b) and 6(c), respectively, the VPCT image may show myocardial blood flow for each voxel therein. Then, in step 503, a CCTA image of the patient is acquired, along with the segmentation of the perfusion area of the left ventricle in the CCTA image (e.g., PT n-tag). Note that, although in fig. 5, the step 503 is executed after the steps 501 and 502, it is not limited to such an order as long as the steps 501 and 503 are completed before the step 504.
In step 504, the VPCT image and the CCTA image may be registered to compensate for their shifts and deviations due to acquisition at different phases of the cardiac cycle, and the segmentation results of the perfusion regions of the left ventricle in the CCTA image are transformed to obtain the segmentation results of the corresponding perfusion regions of the left ventricle in the VPCT image. Typically, CCTA images have a higher resolution but do not provide flow information for the coronary arteries, while VPCT, although containing information on myocardial blood flow, is limited by the resolution of perfusion CT, which is also lower. By registration and transformation, it is possible to benefit from an accurate segmentation of the perfusion regions in the CCTA image, accurately mapping the respective individual perfusion regions in the VPCT image. Further, various flow information of the coronary artery may be determined in step 505 by fully utilizing the voxel-level myocardial blood flow included in the VPCT image based on the precise partition result of the corresponding perfusion region of the left ventricle in the VPCT image. In particular, myocardial blood flow in a respective perfusion region divided in the left ventricle in the VPCT image may be determined as the flow rate of a vessel of interest supplying blood to the perfusion region.
The flow information of the coronary arteries determined based on the perfusion CT images of the patient according to various embodiments of the present disclosure above may be used as ground truth to train the predictive model according to various embodiments of the present disclosure. For example, at least a subset of arterial features and corresponding myocardial features of the vessel of interest may be extracted based on the acquired CCTA images; and training a predictive model by constructing training data based on at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest and the determined flow information of the coronary artery as ground truth.
In some embodiments, in case the flow information of the coronary arteries to be determined is the flow rate of the respective end branch, the flow information of the coronary arteries may be determined based on the VPCT image by any of the following ways. For example, the myocardial blood flow of each voxel may be integrated in the perfusion region of each end branch, and the integrated result may be the flow rate of the corresponding end branch, which shows the nearly equal relationship between the myocardial blood flow of the perfusion region and the flow rate of the corresponding end branch supplying blood, and conforms to the physiological structural mechanism of the coronary artery. For example, the myocardial blood flow of all voxels in the left ventricle may be integrated to obtain the total myocardial blood flow of the left ventricle, and the flow rate of each distal branch may be determined by using the volume ratio of the perfusion region of each distal branch based on the total myocardial blood flow. Specifically, similar to the above description of the relationship between the myocardial blood flow of the perfusion areas and the corresponding end branches supplying blood, the total myocardial blood flow of the left ventricle can be regarded as being provided by the end branches corresponding to the respective perfusion areas together, i.e. equal to the sum of the flow rates of the end branches, the ratio of the flow rates of the end branches also corresponds to the volume ratio of the perfusion areas supplying blood, and the flow rate of each end branch can be obtained by multiplying the total myocardial blood flow of the left ventricle by the volume ratio of the perfusion areas of the end branches.
In some embodiments, in the case where the flow information of the coronary artery to be determined is the total inlet flow rate of each of the left and right coronary arteries and the flow ratio distribution of each blood vessel, the flow information of the coronary artery may be determined as follows. The myocardial blood flow of each voxel in the left ventricle may be integrated based on the VPCT image to obtain a total myocardial blood flow of the left ventricle, and the total inlet flow rates of the left and right coronary arteries may be determined based on the total myocardial blood flow using any one of a volume ratio of a perfusion region of the left coronary artery to a perfusion region of the right coronary artery, a myocardial blood flow ratio of the perfusion region of the left coronary artery to a perfusion region of the right coronary artery, and a statistically empirical ratio of the total inlet flow rates of the left and right coronary arteries. And, the flow ratio distribution of individual vessels within the coronary artery may be determined based on intraluminal attenuation gradient information or other morphological laws of the individual vessels. For example, based on the total inlet flow rate of the left coronary artery, in combination with intraluminal decay gradient information or other morphological laws (reflecting the size distribution of individual vessels therein) of individual vessels within the left coronary artery, the flow rate of individual vessels within the left coronary artery may be determined; the calculation method is also suitable for branch arteries such as the right coronary artery.
In some embodiments, in the case where the flow information of the coronary artery to be determined is the total inlet flow rate of the coronary artery and the flow ratio distribution of the individual vessels, the flow information of the coronary artery may be determined as follows.
Left ventricular myocardium mass may be calculated based on perfusion CT images of the coronary arteries of the patient, and then the total inlet flow rate of the coronary arteries is determined based on the left ventricular myocardium mass using equation (2) as follows:
Qlv＝Qlv0Mlv bequation (2)
Wherein Q islvThe total inlet flow rate representing the coronary artery is also the total myocardial blood flow to the left ventricle, MlvIs the myocardial mass of the left ventricle, Qlv0And b represents the property constant of the patient's left ventricle in its current state. The formula (2) is consistent with the biological anisotropic scaling law, and when the method is used as a prediction model of the total myocardial blood flow of the left ventricle of the coronary artery, the calculation and training are simple, the workload is low, and the prediction accuracy is high.
And, the myocardial blood flow of each voxel may be integrated within the perfusion region of the respective end branch based on the VPCT image, the integration result being the flow rate of the respective end branch. A flow ratio distribution for each terminal branch may be determined based on the flow rate of each terminal branch and the total inlet flow rate of the coronary artery.
In some embodiments, where the flow information of the coronary artery to be determined is the flow rate of any branch artery (e.g., without limitation, the left anterior descending branch) and the flow rate of the distal branch therein, the flow information of the coronary artery may be determined as follows. The flow rate of the branch artery may be determined by integrating the myocardial blood flow of each voxel in the perfusion region of the branch artery based on the VPCT image, or by integrating the myocardial blood flow of each voxel in the left ventricle based on the VPCT image and multiplying by the volumetric proportion of the perfusion region of the branch artery. And, the flow ratio distribution of each vessel within the coronary artery, including the flow ratio distribution of each distal branch in the left anterior descending branch, may be determined based on intraluminal attenuation gradient information or other morphological laws for each vessel. The flow rate of each of the terminal branches may then be determined based on the determined flow rates of the branched arteries and the flow ratio distribution of each of the terminal branches.
Fig. 7 shows a configuration of an apparatus 700 for determining flow information of coronary arteries based on CCTA images of a patient according to an embodiment of the present disclosure. As shown in fig. 7, the apparatus 700 may include an acquisition unit 701, an extraction unit 702, and a prediction unit 705. The acquisition unit 701 may be configured to receive a CCTA image of a patient, e.g., a CCTA image acquired by a CCTA imaging apparatus or a 3D CCTA image reconstructed by post-processing thereof, and transmit it to the extraction unit 702. The extraction unit 702 may include an artery feature extraction unit 703 and a myocardium feature extraction unit 704, wherein the artery feature unit 703 may be configured to extract artery features of a vessel of interest based on the received CCTA image, and the myocardium feature extraction unit 704 may be configured to extract corresponding myocardium features of the vessel of interest based on the acquired CCTA image. Of the extracted arterial features and corresponding myocardial features of the vessel of interest, at least a subset of these features, which are needed for prediction, may be transmitted to the prediction unit 705, in order to predict the flow information of the coronary arteries by it using the received trained prediction model.
Fig. 8 shows a block diagram of an apparatus 800 for determining flow information of coronary arteries based on CCTA images of a patient according to an embodiment of the present disclosure. As shown in fig. 8, the apparatus 800 may include a communication interface 803 configured to acquire CCTA images of the patient; and a processor 801 configured to: extracting arterial features of the vessel of interest based on the acquired CCTA image; extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images; estimating flow information of the coronary arteries using a trained predictive model based on at least a subset of the extracted arterial and myocardial features of the vessel of interest.
In some embodiments, the communication interface 803 may be further configured to: acquiring a sequence of perfusion CT images of the coronary arteries of the patient; and acquiring a division result of the perfusion area of the left ventricle in the CCTA image of the patient so as to process and obtain ground truth values and training data on the basis. Accordingly, the processor 801 may be further configured to: determining a Volumetric Perfusion CT (VPCT) image based on a sequence of perfusion CT images of the coronary arteries together with the heart of the patient, the VPCT image showing myocardial blood flow for each voxel therein; obtaining a partition result of a corresponding perfusion region of a left ventricle in the VPCT image by registering the VPCT image and the CCTA image based on the partition result of the perfusion region of the left ventricle in the CCTA image; determining myocardial blood flow in a respective perfusion region divided in the left ventricle in the VPCT image as the flow rate of a vessel of interest supplying blood to the perfusion region (ground truth); extracting at least a subset of arterial features and corresponding myocardial features of the vessel of interest based on the acquired CCTA images; training the predictive model based on training data comprising at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest and the determined flow rate of the vessel of interest.
The processor 801 executes functions or methods implemented at least by code or instructions included in a program (e.g., image processing program 808) stored in the memory 805, thereby implementing methods of determining flow information of coronary arteries based on CCTA images of a patient according to various embodiments of the present disclosure, and/or determining flow information of coronary arteries based on perfusion CT images of a patient according to various embodiments of the present disclosure, and/or preparing training data and training a predictive model based on flow information of coronary arteries determined based on perfusion CT images of a patient according to various embodiments of the present disclosure.
In some embodiments, the image processing program 808 includes at least the extraction unit 702 and the prediction unit 705 shown in fig. 7. In some embodiments, the image processing program 808 may further include a VPCT image generation unit 902, a registration unit 903, a VPCT image division unit 904, and a flow information determination unit 905 (see the detailed description in conjunction with fig. 9 below) shown in fig. 9, and a training unit (not shown) that trains the prediction model.
Examples of the processor 801 include a Central Processing Unit (CPU), a Micro Processing Unit (MPU), a GPU, a microprocessor, a processor core, a multiprocessor, an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), and the like.
The memory 804 temporarily stores programs loaded from the storage 805 and provides a work area to the processor 801. Various data generated when the program is executed by processor 801, such as, but not limited to, medical image data 806, trained predictive models, etc., may also be temporarily stored in memory 804. The memory 804 includes, for example, a Random Access Memory (RAM) and a Read Only Memory (ROM).
The storage 805 stores programs executed by the processor 801, for example. The storage 805 includes, for example, a Hard Disk Drive (HDD), a Solid State Drive (SSD), and a flash memory.
The input/output interface 802 may include an input device through which the device 800 inputs various operations and an output device through which various processing results are output.
The communication interface 803 performs transmission and reception of various data via a network. The communication may be performed by cable or wirelessly, and any communication protocol may be used as long as it can communicate with each other.
Various components in the apparatus 800 may communicate information with each other via a bus 807. The storage medium may store the program in a "non-transitory tangible medium". Further, the program includes, for example, a software program or a computer program.
Further, at least some processing in apparatus 800 may be implemented by cloud computing configured by one or more computers. In some embodiments, at least some processing in apparatus 800 may be performed by another apparatus. In this case, at least some of the processing of each functional unit realized by the processor 801 may be performed by alternative means.
Fig. 9 shows a configuration of an apparatus 900 for determining flow information of coronary arteries based on a perfusion CT image of a patient according to an embodiment of the present disclosure. As shown in fig. 9, the apparatus 900 includes an acquisition unit 901, a VPCT image generation unit 902, a registration unit 903, a VPCT image division unit 904, and a flow information determination unit 905.
The acquisition unit 901 may be configured to: acquiring a sequence of perfusion CT images of the patient including coronary arteries and heart, CCTA images of the patient and a segmentation of a perfusion region of a left ventricle therein.
The VPCT image generation unit 902 may be configured to: a Volumetric Perfusion CT (VPCT) image showing myocardial blood flow for each voxel therein is determined based on a sequence of perfusion CT images of the coronary arteries of the patient together with the heart.
The registration unit 903 may be configured to: registering the VPCT image with the CCTA image.
The VPCT image dividing unit 904 may be configured to: the segmentation result of the corresponding perfusion region of the left ventricle in the VPCT image is obtained based on the segmentation result of the perfusion region of the left ventricle in the CCTA image and the registration result of the VPCT image and the CCTA image by the registration unit 903.
The flow information determination unit 905 may be configured to: determining flow information of the coronary arteries based on a result of a division of a corresponding perfusion region of a left ventricle in the VPCT image. For example, the flow information determination unit 905 may be further configured to: myocardial blood flow in respective perfusion regions partitioned in the left ventricle in the VPCT image is determined as the flow rate of the vessel of interest supplying blood to the perfusion region.
Various operations or functions are described herein that may be implemented as or defined as software code or instructions. Such content may be source code or differential code ("delta" or "patch" code) that may be executed directly ("object" or "executable" form). The software code or instructions may be stored in a computer-readable storage medium and, when executed, may cause a machine to perform the functions or operations described, and includes any mechanism for storing information in a form accessible by a machine (e.g., a computing device, an electronic system, etc.), such as recordable or non-recordable media (e.g., Read Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, etc.).
The exemplary methods described herein may be machine or computer-implemented, at least in part. Some examples may include a non-transitory computer-readable medium or machine-readable medium encoded with instructions operable to configure an electronic device to perform a method as described in the above examples. An implementation of such a method may include software code, such as microcode, assembly language code, higher level language code, or the like. Various programs or program modules may be created using various software programming techniques. For example, program segments or program modules may be designed using Java, Python, C + +, assembly language, or any known programming language. One or more of such software portions or modules may be integrated into a computer system and/or computer-readable medium. Such software code may include computer readable instructions for performing various methods. The software code may form part of a computer program product or a computer program module. Further, in one example, the software code can be tangibly stored on one or more volatile, non-transitory, or non-volatile tangible computer-readable media, such as during execution or at other times. Examples of such tangible computer-readable media may include, but are not limited to, hard disks, removable magnetic disks, removable optical disks (e.g., compact disks and digital video disks), magnetic cassettes, memory cards or sticks, Random Access Memories (RAMs), Read Only Memories (ROMs), and the like.
Moreover, although illustrative embodiments have been described herein, the scope includes any and all embodiments having equivalent elements, modifications, omissions, combinations (e.g., of aspects across various embodiments), adaptations or alterations based on the present disclosure. The elements in the claims are to be interpreted broadly based on the language employed in the claims and not limited to examples described in the specification or during the life of the application. Further, the steps of the disclosed methods may be modified in any manner, including by reordering steps or inserting or deleting steps. It is intended, therefore, that the description be regarded as examples only, with a true scope being indicated by the following claims and their full scope of equivalents.

Claims (16)

1. A method of determining flow information of coronary arteries based on a CCTA image of a patient, the method comprising:
acquiring a CCTA image of the patient, a segmentation of a perfusion region of a left ventricle in the CCTA image of the patient, and a sequence of perfusion CT images, wherein the perfusion CT images include coronary arteries and a heart;
extracting arterial features of the vessel of interest based on the acquired CCTA image;
extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images;
predicting flow information of the coronary artery by using a trained prediction model based on the extracted artery features and at least a subset of the corresponding myocardium features of the blood vessel of interest, wherein the prediction model is trained based on the artery features and at least a subset of the corresponding myocardium features of the blood vessel of interest and a ground truth value of the flow information of the coronary artery, and the ground truth value of the flow information of the coronary artery is obtained by the following steps:
determining a VPCT image based on a sequence of perfusion CT images of the patient, the VPCT image showing myocardial blood flow for each voxel therein;
obtaining a partition result of a corresponding perfusion area of a left ventricle in the VPCT image by registering the VPCT image and the CCTA image based on the partition result of the perfusion area of the left ventricle in the CCTA image;
determining a ground truth value for flow information of the coronary arteries based on a partitioning result of a corresponding perfusion region of a left ventricle in the VPCT image.
2. The method of claim 1, wherein extracting arterial features of a vessel of interest based on the acquired CCTA images comprises:
performing artery segmentation on the acquired CCTA image and extracting the center line of the coronary artery tree; and
extracting arterial features of a vessel of interest based on the extracted centerline from the CCTA image, the arterial features including at least one of radius-related information, length-related information, volume-related information, intensity-related information, and intraluminal attenuation gradient-related information of the vessel of interest.
3. The method of claim 1 or 2, wherein extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images comprises:
performing left ventricular segmentation on the acquired CCTA images;
dividing perfusion areas of various concerned blood vessels on the left ventricle;
determining a myocardial characteristic for the perfusion region of each vessel of interest as a corresponding myocardial characteristic of the vessel of interest, the myocardial characteristic comprising at least one of myocardial mass and intensity related information.
4. The method of claim 1, wherein the vessel of interest is dependent on flow information and a predictive model of the coronary artery to be determined.
5. The method of claim 1, wherein the vessel of interest includes respective terminal branches.
6. The method of claim 3, wherein the coronary flow information includes flow ratio distribution of individual vessels throughout the coronary artery tree.
7. The method according to claim 3, wherein the flow information of the coronary arteries to be determined comprises a flow rate of the vessel of interest, and the predictive model is represented as:
Q＝Q0Mbequation (1)
Wherein Q represents the flow rate of the vessel of interest, M is the myocardial mass of the perfused region of the vessel of interest in the left ventricle, Q0And b represents the property constant of the patient in the current state of the vessel of interest.
8. The method of claim 3, wherein the predictive model is constructed based on a learning network, at least a subset of the extracted arterial features and corresponding myocardial features of the vessel of interest being input to the learning network.
9. The method of claim 3, wherein the predictive model is trained using flow information of the coronary arteries of the patient, obtained based on perfusion CT images of the coronary arteries together with the heart, as ground truth.
10. The method according to claim 1, wherein determining flow information of the coronary arteries based on the partition of the corresponding perfusion region of the left ventricle in the VPCT image comprises in particular:
myocardial blood flow in a corresponding perfusion region divided in the left ventricle in the VPCT image is determined as the flow rate of a blood vessel of interest supplying blood to the perfusion region.
11. Method according to claim 1, characterized in that, in the case where the flow information of the coronary arteries to be determined is the flow rate of the respective terminal branch:
integrating myocardial blood flow of each voxel in a perfusion region of each end branch based on the VPCT image, wherein the integrated result is used as the flow rate of the corresponding end branch; and/or
Based on the VPCT image, myocardial blood flow of all voxels within the left ventricle is integrated to obtain a total myocardial blood flow of the left ventricle, based on which the flow rate of each distal branch is determined using the volume proportion of the perfusion area of each distal branch.
12. The method according to claim 1, characterized in that in case the flow information of the coronary arteries to be determined is the total inlet flow rate of each of the left and right coronary arteries and the flow ratio distribution of the individual vessels:
integrating myocardial blood flow of each voxel in the left ventricle based on the VPCT image to obtain total myocardial blood flow of the left ventricle, determining total inlet flow rates of the left and right coronary arteries, respectively, using any one of a volume ratio of perfusion regions of the left and right coronary arteries, a myocardial blood flow ratio of perfusion regions of the left and right coronary arteries, and a statistically empirical ratio of total inlet flow rates of the left and right coronary arteries, based on the total myocardial blood flow; and
and determining the flow ratio distribution of each blood vessel in the coronary artery based on the intraluminal attenuation gradient information of each blood vessel or the size distribution relation of each blood vessel.
13. The method according to claim 1, characterized in that in case the flow information of the coronary arteries to be determined is the total inlet flow rate of the coronary arteries and the flow ratio distribution of the individual vessels:
calculating a left ventricular myocardium mass based on the perfusion CT image of the coronary arteries of the patient, determining a total inlet flow rate of the coronary arteries using the following equation (2) based on the left ventricular myocardium mass:
Qlv＝Qlv0Mlv bequation (2)
Wherein Q islvThe total inlet flow rate representing the coronary artery is also the total myocardial blood flow to the left ventricle, MlvIs the myocardial mass of the left ventricle, Qlv0And b represents the property constant of the patient's left ventricle in its current state; and
integrating myocardial blood flow of each voxel in a perfusion region of each end branch based on the VPCT image, taking the integrated result as the flow rate of the corresponding end branch, and determining flow ratio distribution of each end branch based on the flow rate of each end branch and the total inlet flow rate of the coronary artery.
14. Method according to claim 1, characterized in that in case the flow information of the coronary artery to be determined is the flow rate of the branch artery and of the terminal branch therein:
determining a flow rate of the branch artery based on integrating myocardial blood flow of each voxel within a perfusion region of the branch artery based on the VPCT image or integrating myocardial blood flow of each voxel within a left ventricle based on the VPCT image and multiplying by a volume fraction of the perfusion region of the branch artery;
determining the flow ratio distribution of each blood vessel in the coronary artery based on the intraluminal attenuation gradient information of each blood vessel or the size distribution relation of each blood vessel;
determining flow rates of the respective end branches based on the determined flow rates of the branched arteries and a flow ratio distribution of the respective end branches therein.
15. An apparatus for determining flow information of coronary arteries based on a CCTA image of a patient, the apparatus comprising:
an interface configured to acquire a CCTA image of the patient, a segmentation of a perfusion region of a left ventricle in the CCTA image of the patient, and a sequence of perfusion CT images of a coronary artery of the patient, wherein the perfusion CT images include a coronary artery and a heart; and
a processor configured to:
extracting arterial features of the vessel of interest based on the acquired CCTA image;
extracting corresponding myocardial features of the vessel of interest based on the acquired CCTA images;
predicting the flow information of the coronary artery by using a trained prediction model based on the extracted at least subset of the artery features and the myocardium features of the blood vessel of interest, wherein the prediction model is trained based on the artery features and the at least subset of the corresponding myocardium features of the blood vessel of interest and the ground truth values of the flow information of the coronary artery, and the ground truth values of the flow information of the coronary artery are obtained by the following steps:
determining a VPCT image based on a sequence of perfusion CT images of the patient, the VPCT image showing myocardial blood flow for each voxel therein;
obtaining a partition result of a corresponding perfusion area of a left ventricle in the VPCT image by registering the VPCT image and the CCTA image based on the partition result of the perfusion area of the left ventricle in the CCTA image;
determining a ground truth value for flow information of the coronary arteries based on a partitioning result of a corresponding perfusion region of a left ventricle in the VPCT image.
16. A non-transitory computer readable medium having stored thereon instructions which, when executed by a processor, implement the method of determining flow information of coronary arteries based on CCTA images of a patient according to any one of claims 1-14.
CN201910837899.5A 2018-09-05 2019-09-05 Method, apparatus and medium for determining flow information of coronary artery based on coronary artery CT angiography image of patient Active CN110428420B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201862726989P 2018-09-05 2018-09-05
US62/726,989 2018-09-05

Publications (2)

Publication Number Publication Date
CN110428420A CN110428420A (en) 2019-11-08
CN110428420B true CN110428420B (en) 2022-05-17

Family

ID=68418746

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910837899.5A Active CN110428420B (en) 2018-09-05 2019-09-05 Method, apparatus and medium for determining flow information of coronary artery based on coronary artery CT angiography image of patient

Country Status (1)

Country Link
CN (1) CN110428420B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4276751A3 (en) 2019-11-28 2024-01-10 Siemens Healthcare GmbH Computer-implemented method for evaluating a ct data set regarding perivascular tissue, evaluation device, computer program and electronically readable storage medium
CN111738982A (en) * 2020-05-28 2020-10-02 数坤（北京）网络科技有限公司 Vascular cavity concentration gradient extraction method and device and readable storage medium
CN112336365B (en) * 2020-11-11 2021-09-28 上海市第六人民医院 Myocardial blood flow distribution image acquisition method, myocardial blood flow distribution image acquisition system, myocardial blood flow distribution image acquisition medium and electronic equipment
CN113450322A (en) * 2021-06-22 2021-09-28 沈阳先进医疗设备技术孵化中心有限公司 Method and device for judging origin abnormality of coronary artery
CN116994067B (en) * 2023-09-07 2024-05-07 佛山科学技术学院 Method and system for predicting fractional flow reserve based on coronary artery calcification

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105848577A (en) * 2013-10-01 2016-08-10 皇家飞利浦有限公司 System and method for myocardial perfusion pathology characterization
CN106023202A (en) * 2016-05-20 2016-10-12 苏州润心医疗科技有限公司 Coronary artery fractional flow reserve calculation method based on heart CT image
CN106456078A (en) * 2013-10-17 2017-02-22 西门子保健有限责任公司 Method and system for machine learning based assessment of fractional flow reserve
CN107123112A (en) * 2017-01-23 2017-09-01 上海联影医疗科技有限公司 blood flow state analysis system and method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105848577A (en) * 2013-10-01 2016-08-10 皇家飞利浦有限公司 System and method for myocardial perfusion pathology characterization
CN106456078A (en) * 2013-10-17 2017-02-22 西门子保健有限责任公司 Method and system for machine learning based assessment of fractional flow reserve
CN106023202A (en) * 2016-05-20 2016-10-12 苏州润心医疗科技有限公司 Coronary artery fractional flow reserve calculation method based on heart CT image
CN107123112A (en) * 2017-01-23 2017-09-01 上海联影医疗科技有限公司 blood flow state analysis system and method

Also Published As

Publication number Publication date
CN110428420A (en) 2019-11-08

Similar Documents

Publication Publication Date Title
CN110428420B (en) Method, apparatus and medium for determining flow information of coronary artery based on coronary artery CT angiography image of patient
US11728039B2 (en) Methods and systems for predicting sensitivity of blood flow calculations to changes in anatomical geometry
US20210275124A1 (en) Method for calculating coronary artery fractional flow reserve on basis of myocardial blood flow and ct images
US20190304592A1 (en) System and method for calculating vessel flow parameters based on angiography
US11399779B2 (en) System-independent quantitative perfusion imaging
US9965891B2 (en) Systems and methods for image-based object modeling using multiple image acquisitions or reconstructions
US9867584B2 (en) Method of determining the blood flow through coronary arteries
CN110944586B (en) Inflammation Estimation from X-ray Image Data
US11389130B2 (en) System and methods for fast computation of computed tomography based fractional flow reserve
US20180315505A1 (en) Optimization of clinical decision making
US11328423B2 (en) Characterization of plaque
CN111429502B (en) Method and system for generating a centerline of an object and computer readable medium
CN109688908B (en) Apparatus and method for determining fractional flow reserve
US11369278B2 (en) Methods and systems for determining lumen volume and coronary blood flow
EP3695784A1 (en) Prediction of coronary microvascular dysfunction from coronary computed tomography
EP3564963A1 (en) System and methods for fast computation of computed tomography based fractional flow reserve
CN109949300B (en) Method, system and computer readable medium for anatomical tree structure analysis
CN114677492A (en) Coronary artery microcirculation hemodynamics simulation analysis method and device
JP2018509971A (en) Method and apparatus for quantitative flow analysis
CN110070534B (en) Method for automatically acquiring feature sequence based on blood vessel image and device for predicting fractional flow reserve
CN110506293A (en) Spectrum imaging FFR
EP3588378B1 (en) Method for determining at least one enhanced object feature of an object of interest
CN113129297A (en) Automatic diameter measurement method and system based on multi-phase tumor images
CN112690814A (en) Low-error coronary artery blood flow reserve fraction measuring method
CN111340934A (en) Method and computer system for generating a combined tissue-vessel representation

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