CN110428420A - The method, apparatus and medium of flowing information coronarius are determined based on the coronary artery CT angiographic image of patient - Google Patents
The method, apparatus and medium of flowing information coronarius are determined based on the coronary artery CT angiographic image of patient Download PDFInfo
- Publication number
- CN110428420A CN110428420A CN201910837899.5A CN201910837899A CN110428420A CN 110428420 A CN110428420 A CN 110428420A CN 201910837899 A CN201910837899 A CN 201910837899A CN 110428420 A CN110428420 A CN 110428420A
- Authority
- CN
- China
- Prior art keywords
- image
- blood vessel
- coronarius
- myocardial
- perfusion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
This disclosure relates to which a kind of coronary artery CT angiographic image based on patient determines the method, apparatus and medium of flowing information coronarius.The described method includes: obtaining the CCTA image of the patient；Artery characteristics based on acquired CCTA image zooming-out concern blood vessel；Based on the correspondence myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out；At least subset in artery characteristics and corresponding myocardial features based on extracted concern blood vessel, using trained prediction model, to predict the flowing information coronarius.This method can obtain more acurrate and patient-specific flowing information coronarius from CCTA image, more reliable virtual FFR prediction can be further obtained accordingly, quantitative patient-specific flowing physiological information can also be provided for doctor by conventional CCTA inspection.
Description
Cross reference to related applications
This application claims U.S. Provisional Application No. 62/726,989 priority that September in 2018 is submitted on the 5th, complete
Portion's content is incorporated herein by reference.
Technical field
The present disclosure relates generally to Medical Image Processings and analysis.More specifically, this disclosure relates to being used for the hat based on patient
The medical image of shape artery determines the method, apparatus and medium of flowing information coronarius.
Background technique
In general, the Main Diagnosis method of coronary artery disease depends on the invasive measurement of blood flow reserve score (FFR).Most
Closely, such as computational fluid dynamics and depth are applied to coronary artery computer tomography angiography (CCTA) image
The noninvasive method of learning art is spent, to obtain FFR (also referred to as virtual FFR).The flowing information coronarius of patient, example
It also can be the prediction side of these new virtual FFR even if cannot be directly used to diagnose if entrance flow rate and flow-rate ratio are distributed
Method provides important average information, such as patient-specific boundary condition or flow characteristics.Particularly, virtual FFR
Prediction technique often rely on these average informations to be predicted.However, these average informations are managed usually using from simple
Estimate by the empirical relation that equation or statistic correlation obtain, leads to the flowing information of coronary artery artery of estimated patient simultaneously
(namely with the real conditions of patient and be inaccurately consistent) of non-patient-specific, obtains virtual FFR on this basis, then shows
Write the precision of prediction for affecting virtual FFR.
The disclosure is provided to overcome above-mentioned technological deficiency in the prior art.
Summary of the invention
The disclosure is intended to provide a kind of CCTA image based on patient to determine method, the dress of flowing information coronarius
It sets and medium, more acurrate and patient-specific flowing information coronarius can be obtained from CCTA image, further may be used
Obtain more reliable virtual FFR prediction accordingly, can also be checked by conventional CCTA provided for doctor it is quantitative patient-specific
Flow physiological information.
In one aspect, the disclosure provides a kind of coronary artery computer tomography angiography based on patient
(CCTA) image is come the method that determines flowing information coronarius, which comprises obtains the CCTA image of the patient；
Artery characteristics based on acquired CCTA image zooming-out concern blood vessel；Based on paying close attention to blood described in acquired CCTA image zooming-out
The correspondence myocardial features of pipe；At least subset in artery characteristics and corresponding myocardial features based on extracted concern blood vessel, benefit
With trained prediction model, to predict the flowing information coronarius.
On the other hand, the disclosure provides a kind of perfusion computer tomography (CT) image based on patient to determine
The method of flowing information coronarius, which comprises obtain the sequence of the Perfusion CT image of the patient, the perfusion
CT image includes coronary artery and heart；Volume Perfusion CT (VPCT) is determined based on the sequence of the Perfusion CT image of the patient
Image, the VPCT image show the myocardial blood flow of wherein each voxel；Obtain the coronary artery computerized tomography of the patient
The division result of the perfusion area of photography angiography (CCTA) image and the left ventricle in the CCTA image；Based on described
The division result of the perfusion area of left ventricle in CCTA image, by matching to the VPCT image and the CCTA image
Standard, to obtain the division result of the corresponding perfusion area of left ventricle in the VPCT image；And based on left in the VPCT image
The division result of the corresponding perfusion area of ventricle, to determine the flowing information coronarius.
It yet still another aspect, the disclosure provides a kind of coronary artery computer tomography angiography based on patient
(CCTA) image determines the device of flowing information coronarius, and described device includes: interface, is configured to obtain the trouble
The CCTA image of person；And processor, it is configured that the artery characteristics based on acquired CCTA image zooming-out concern blood vessel；
Based on the correspondence myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out；Artery based on extracted concern blood vessel
At least subset in feature and myocardial features, using trained prediction model, to estimate the flowing information coronarius.
It yet still another aspect, the disclosure provides a kind of non-transitory computer-readable medium, it is stored thereon with instruction, the finger
It enables when executed by the processor, realizes the coronary artery computer tomography based on patient according to each embodiment of the disclosure
Angiography (CCTA) image is come the method that determines flowing information coronarius, and/or realizes according to each implementation of the disclosure
The method that flowing information coronarius is determined based on perfusion computer tomography (CT) image of patient of example.
Flowing information coronarius is determined based on the CCTA image of patient using according to each embodiment of the disclosure
Method, apparatus and medium, more acurrate and patient-specific flowing information coronarius can be obtained from CCTA image, it is empty
Quasi- FFR prediction, which benefits from this, can make prediction result more reliable accurate, can also be provided by conventional CCTA inspection for doctor quantitative
Patient-specific flowing physiological information.
It should be appreciated that foregoing general description and the following detailed description be only it is exemplary and illustrative, be not
Limitation to claimed invention.
Detailed description of the invention
In the attached drawing being not drawn necessarily to scale, identical appended drawing reference can describe the similar department in different views
Part.Same numbers with letter suffix or different letter suffix can indicate the different instances of like.Attached drawing usually passes through
Example rather than various embodiments are shown by way of limitation, and be used for together with specification and claims to institute
Disclosed embodiment is illustrated.Such embodiment is illustrative, and be not intended as this method, device, system or
It is stored thereon with the exhaustive or exclusive embodiment of the non-transitory computer-readable medium of the instruction for realizing this method.
Fig. 1, which is shown, determines flowing information coronarius based on the CCTA image of patient according to the embodiment of the present disclosure
The flow chart of method；
Fig. 2 shows the sons that the CCTA image zooming-out based on patient according to the embodiment of the present disclosure pays close attention to the artery characteristics of blood vessel
Process；
Fig. 3 shows the correspondence myocardial features of the CCTA image zooming-out concern blood vessel based on patient according to the embodiment of the present disclosure
Sub-process；
Fig. 4 (a) and Fig. 4 (b) shows the coronary artery of the CCTA image reconstruction based on patient according to the embodiment of the present disclosure
The diagram of the 3D model of tree, wherein show the division result of the perfusion area on left ventricle and identify end branch；
Fig. 5 shows determining based on perfusion computer tomography (CT) image of patient according to the embodiment of the present disclosure
The flow chart of the method for flowing information coronarius；
Volume Perfusion CT (VPCT) figure determined by the sequence of the Perfusion CT image of patient is shown respectively in Fig. 6 (a)-Fig. 6 (c)
Axial view, sagittal figure and the anterior view of picture；
Fig. 7, which is shown, determines flowing information coronarius based on the CCTA image of patient according to the embodiment of the present disclosure
Device configuration；
Fig. 8, which is shown, determines flowing information coronarius based on the CCTA image of patient according to the embodiment of the present disclosure
Device block diagram；
Fig. 9, which is shown, determines flowing letter coronarius based on the Perfusion CT image of patient according to the embodiment of the present disclosure
The configuration of the device of breath.
Specific embodiment
Hereinafter, technical term " blood flow " (such as " blood flow " in " myocardial blood flow ", " coronary blood flow ")
Indicate the volumetric blood that the unit time is flowed through in corresponding organ tissue.Technical term " flow rate " also illustrates that the stream flowed through the unit time
Body volume.Technical term " CCTA image " herein can be indicated by coronary artery computer tomography angiographic imaging
The device section CCTA collected (such as axial, sagittal, coronal section etc.) image, can also indicate by these cross-sectional images into
The 3D rendering that row post-processing is for example rebuild.Similarly, technical term herein " Perfusion CT image " can also indicate by
Computed tomography imaging device Perfusion CT section collected (such as axial, sagittal, coronal section etc.) image is perfused,
It can indicate to be post-processed the 3D rendering for example rebuild by these cross-sectional images.Technical term " VPCT image " can be with
It indicates the 3D Perfusion CT image that sequence based on Perfusion CT image calculates blood flow to obtain, and wherein each voxel is shown
Myocardial blood flow.
In addition, it is so-called different and different to " training " the operation view " prediction model " of " prediction model " herein, for example, needle
To the prediction model being made of neural network, " training " can be indicated using training data for example, by using stochastic gradient descent method tune
The parameter of whole neural network is so as to the maximized training process of objective function；And it is directed to the prediction model being made of explicit formula,
" training " can then indicate to determine the process of the parameter (such as constant) in formula using training data.Technical term " end
Branch " indicates the vessel branch of not sub-branch.
Fig. 1 shows the CCTA image according to an embodiment of the present disclosure based on patient to determine flowing letter coronarius
The flow chart of the method 100 of breath.As shown in Figure 1, obtaining the CCTA image of the patient in step 101.It, can be in step 102
Artery characteristics based on acquired CCTA image zooming-out concern blood vessel.And it, can be based on acquired in step 103
The correspondence myocardial features of blood vessel are paid close attention to described in CCTA image zooming-out.Artery characteristics can obtain from concern blood vessel itself and describe to close
In the characteristic of the concern blood vessel；The correspondence myocardial features for paying close attention to blood vessel can be obtained from the cardiac muscle of the perfusion area of concern blood vessel, and
Characteristic about the cardiac muscle is described.Although the two can be with one note that step 102 and step 103 successively execute in Fig. 1
It rises and executes or be performed in a different order, as long as being executed before step 104.Then, in step 104, institute can be based on
At least subset in the artery characteristics and corresponding myocardial features of the concern blood vessel of extraction is come pre- using trained prediction model
Survey the flowing information coronarius.The raw information of this method 100 is originated from the CCTA image of sufferers themselves, it is accurate and with
Higher resolution ratio has fed back patient-specific image information coronarius；It is mentioned by focusing on the feature of concern blood vessel
It takes, compares the accuracy rate that work load is reduced to all vessel extraction characteristic remarkables and has taken into account prediction；In addition, it can be with base
It is predicted in both the myocardial features for the blood vessel feature and function feature for reflecting design feature coronarius, so that prediction base
In characteristic information more comprehensively, prediction result is more accurate, more meets the distinctive structure and function coronarius of patient.
In some embodiments, the concern of artery characteristics and corresponding myocardial features is extracted for it in step 102 and 103
Blood vessel can depend on the flowing information and prediction model coronarius to be determined.
For example, blood vessel can will be paid close attention to if to determine the flowing information of left anterior descending branch (LAD) and its all sub-branches
Accordingly it is defined as left anterior descending branch (LAD) and its all sub-branches；If determining arteria coroaria sinistra (LCA) and its all sub-branches
Flowing information, then can will concern blood vessel accordingly be defined as LCA and its all sub-branches, can be related in this way with Reduction Computation
The range of blood vessel is paid close attention to, and reduces the workload for extracting artery characteristics and corresponding myocardial features.For example, if to determine left front
Descending branch (LAD) and its flowing information of all sub-branches can also divide concern blood vessel from left anterior descending branch (LAD) and its all sons
Branch is extended to comprising other branches coronarius and/or sub-branch, the range for the concern blood vessel being so related in Reduction Computation
And while reducing the workload for extracting artery characteristics and corresponding myocardial features, with reference to other branches and/or sub-branch pair
The interaction for determining the branch and/or sub-branch of its flowing information, so that prediction result is more acurrate.In another example true
In the case where the flow distribution ratio for determining each blood vessel in coronary arterial tree, concern blood vessel can be defined as all each blood vessels.
For another example concern blood vessel can also be defined as including each end branch, in this way, the flexible of myocardial features extraction can be improved
Degree.It particularly, can be to all each end branches no matter to determine the flowing information of which section bifurcated artery coronarius
Corresponding myocardial features (myocardial features of the perfusion area of such as, but not limited to each end branch, each end branch includes
Myocardial features etc. in the expansion area of its perfusion area) it is screened and is integrated, such as filter out and belong to each of this section of bifurcated artery
The corresponding myocardial features of a end branch are simultaneously integrated, to obtain the myocardial features of the perfusion area of this section of bifurcated artery, are made
For the correspondence myocardial features of this section of bifurcated artery.
In some embodiments, output of the flowing information coronarius to be determined as prediction model, difference prediction
Model has different inputs.If the prediction model for determining the main entrance flow rate of LAD, and using indicates are as follows:
Q=Q0Mb, formula (1)
Wherein, Q indicates the flow rate of the concern blood vessel, and M is the myocardium matter that the perfusion area of blood vessel is paid close attention to described in left ventricle
Amount, Q0The attribute constant under the concern blood vessel current state of the patient is indicated with b, then can be defined as concern blood vessel
LAD itself, or belong to each end branch of LAD.For example, in the case where concern blood vessel is defined as LAD itself, Ke Yigen
Main entrance flow rate of the myocardial blood flow Q of the perfusion area of LAD as LAD is directly calculated according to formula (1)；And in concern blood vessel definition
In the case where each end branch to belong to LAD, the perfusion area of each end branch can be first calculated according to formula (1)
Myocardial blood flow is simultaneously summed, the main entrance flow rate to obtain the myocardial blood flow of the perfusion area of LAD, as LAD.Above-mentioned formula
(1) it is consistent with the friction speed scaling law of biology (allometric scaling law), as each blood vessel of coronary artery
When the prediction model of flow rate, calculates and training is simple, workload is low, and precision of prediction is higher.
In some embodiments, it is also possible to which formula (1) is extended to Q=f (V), wherein V indicates extracted concern blood
The subset for the feature vector that the artery characteristics of pipe and corresponding myocardial features are constituted, f (V) indicate that the function of V, the function can indicate
For the explicit formula of such as formula (1), extensive learning network can also be interpreted as.
In some embodiments, the learning network is such as, but not limited to linear regression model (LRM), multilayer neural network etc., institute
Input of at least subset as the learning network in the artery characteristics and corresponding myocardial features of the concern blood vessel of extraction.Example
Such as, learning network, which can define, uses the artery characteristics of which section bifurcated artery and corresponding myocardial features as input, can be according to
The input for practising network defines to limit concern blood vessel.
In some embodiments, spy can be constituted with myocardial features are corresponded to based on the artery characteristics of extracted concern blood vessel
Vector is levied, but the characteristic parameter in this feature vector need not be completely used for predicting sometimes, it particularly, can be according to prediction model
Its subset is selected with the flowing information to be predicted.For example, predicting certain section as prediction model using formula (1) above
When the flow rate of branch, the artery characteristics of this section of branch can not be considered, and only consider that the cardiac muscle of the perfusion area of this section of branch is special
Sign.For another example learning network prespecified can input, such as which section branch when constituting prediction model using learning network
And its sub-branch which kind artery characteristics parameter and which corresponding kind myocardial features parameter, when prediction then can mutually be considered as
The respective subset of characteristic parameter in feature vector.
The extraction sub-process of the artery characteristics of concern blood vessel is specifically described below.
As shown in Fig. 2, the extraction sub-process 200 of the artery characteristics of concern blood vessel starts from step 201, to acquired
CCTA image executes artery segmentation, and extracts the center line of coronary arterial tree.Then, in step 202, from the CCTA image,
Based on extracted center line, the artery characteristics of concern blood vessel are extracted.The artery characteristics may include the concern blood vessel
In radius relevant information, length relevant information, volume relevant information, strength related information and lumen in attenuation gradient relevant information
At least one.
In some embodiments, the radius relevant information of the concern blood vessel may include such as mean lumen areas, retouch
The radial features of concern blood vessel are stated.The averaging of " mean lumen areas " can be and concern blood vessel itself is executed, or
Concern blood vessel and its upstream blood vessel are executed, or concern blood vessel and its downstream blood vessel are executed.In some embodiments,
The absolute value or normalized value of these parameters can be used.
In some embodiments, the length relevant information of the concern blood vessel can be the length of concern blood vessel itself, or
Pay close attention to the cumulative length of blood vessel and its upstream blood vessel, or the cumulative length of concern blood vessel and its downstream blood vessel.In some implementations
In example, the absolute value or normalized value of these parameters can be used.
In some embodiments, the volume relevant information of the concern blood vessel can be the volume of concern blood vessel itself, or
Person pays close attention to the cumulative volume of blood vessel and its upstream blood vessel, or the cumulative volume of concern blood vessel and its downstream blood vessel.In some realities
It applies in example, the absolute value or normalized value of these parameters can be used.
In some embodiments, the strength related information of the concern blood vessel may include such as average cavity intensity, retouch
The strength characteristic of concern blood vessel is stated.The averaging of " average cavity intensity " can be and concern blood vessel itself is executed, or
Concern blood vessel and its upstream blood vessel are executed, or concern blood vessel and its downstream blood vessel are executed.In some embodiments,
The absolute value or normalized value of these parameters can be used.
In some embodiments, attenuation gradient (TAG) in the lumen of all blood vessels, i.e. position in the lumen of blood vessel can be calculated
The radiation decay depreciation set is between the fore-and-aft distance (at a distance from center line) since the entrance of the blood vessel to the position
Linear regression coeffficient (can be the variable quantity of the CT value away from vascular entrance per unit longitudinal length, that is, the variation of HU
Amount).The TAG of all blood vessels reflects each blood vessel in coronary arterial tree along the size distributed intelligence of center line, it is possible thereby to
The flow-rate ratio distribution in entire coronary arterial tree is determined based on the TAG of all blood vessels, as flowing information coronarius.In
It, can also be entire coronal dynamic to determine based on other morphology laws of all blood vessels other than TAG in some embodiments
Flow-rate ratio distribution in arteries and veins tree is used as flowing information coronarius.
The extraction sub-process of the correspondence myocardial features of concern blood vessel is specifically described below.
As shown in figure 3, the extraction sub-process 300 of the correspondence myocardial features of concern blood vessel starts from step 301, to acquired
CCTA image execute left ventricle segmentation.In step 302, the perfusion area of each concern blood vessel is marked off on the left ventricle,
As shown in Fig. 4 (a) and Fig. 4 (b), the perfusion area PT n of end branch EB n, n for 1 to 7 any natural number.In some implementations
In example, the division of perfusion area PT n can be realized using various methods.For example, can based on each subregion from left ventricle to
Perfusion area, is dispatched to neighbouring blood vessel by the nearest geodesic distance of blood vessel, the perfusion area as the blood vessel.Alternatively, can also benefit
With dry-hat model of coronary arterial tree, divided based on the friction speed scaling law between accumulation coronary artery length and myocardial mass
Send perfusion area.
Then, in step 303, the perfusion area PT n for each concern blood vessel determines myocardial features as the concern blood vessel
Corresponding myocardial features.For example, the myocardial features may include at least one of myocardial mass and strength related information.Cardiac muscle
Then quality can be obtained by being integrated to the voxel volume in perfusion area to obtain total volume multiplied by cardiac texture.
And strength related information describes the strength characteristic of perfusion area, may include mean intensity in the perfusion area, maximum intensity and
At least one of minimum strength；In some embodiments, the absolute value or normalized value of these parameters can be used.
Note that perfusion area PT n can be corresponded with end branch EB n, left ventricle can be based on each end branch
EB n divides perfusion area PT n, as shown in Fig. 4 (a) and Fig. 4 (b), but not limited to this, for example, it is also possible to be based on other ranks
Branch divide perfusion area, such as arteria coroaria sinistra (LCA) and the respective perfusion area of arteria coronaria dextra (RCA) or LAD,
Left Circumflex branch (LCX) and the respective perfusion area RCA.
Fig. 5, which is shown, determines flowing information coronarius based on the Perfusion CT image of patient according to the embodiment of the present disclosure
Method 500 flow chart.In step 501, the sequence of the Perfusion CT image of the patient is obtained, the Perfusion CT image includes
Coronary artery and heart.In step 502, the sequence of the Perfusion CT image based on the patient, such as by maximum-slope method or
Person's deconvolution method etc., to determine volume Perfusion CT (VPCT) image of 3D, axial view, sagittal figure and anterior view are respectively as schemed
6 (a), shown in Fig. 6 (b) and Fig. 6 (c), VPCT image can show the myocardial blood flow of wherein each voxel.Then, in step
503, obtain division result (such as the PT of the CCTA image of the patient and the perfusion area of the left ventricle in the CCTA image
N label).Although being not limited to such order note that step 503 executes after step 501 and 502 in Fig. 5, only
Step 501-503 is wanted to complete before step 504.
In step 504, VPCT image can be registrated with CCTA image, to compensate them due in cardiac cycle
Different phase obtain caused by displacement and deviation, and the division result of the perfusion area to the left ventricle in the CCTA image
It is converted, to obtain the division result of the corresponding perfusion area of left ventricle in the VPCT image.In general, CCTA image point
Resolution is higher but does not provide flowing information coronarius, although and information of the VPCT comprising myocardial blood flow, be limited to
The resolution ratio of Perfusion CT, resolution ratio are also relatively low.By being registrated and converting, the perfusion area in CCTA image can benefit from
Accurate to divide, Precision Mapping goes out corresponding each perfusion area in VPCT image.It in turn, can be in step 505, based on described
The accurate division result of the corresponding perfusion area of left ventricle, makes full use of the voxel level for including in VPCT image in VPCT image
Myocardial blood flow, to determine the various flowing informations coronarius.Particularly, it can determine in the VPCT image
Left ventricle in myocardial blood flow in the corresponding perfusion area that divides, the flow rate as the concern blood vessel to the perfusion area blood supply.
Flowing coronarius determined by the Perfusion CT image based on patient above according to each embodiment of the disclosure
Information can be used as ground truth, be trained to according to the prediction model of each embodiment of the disclosure.For example, can be based on
At least subset in the artery characteristics and corresponding myocardial features of blood vessel is paid close attention to described in acquired CCTA image zooming-out；And it is based on
It is extracted it is described concern blood vessel artery characteristics and corresponding myocardial features at least subset and as ground truth
The identified flowing information coronarius, collectively forms training data, to be trained to prediction model.
In some embodiments, the case where the flowing information coronarius to be determined is the flow rate of each end branch
Under, the flowing information coronarius can be determined by following any method based on the VPCT image.Example
Such as, the myocardial blood flow of each voxel can be integrated in the perfusion area of each end branch, integral result can be made
For the flow rate of associated end branch, this is embodied between the myocardial blood flow of perfusion area and the flow rate of the associated end branch of blood supply
Intimate relation of equality, meet physiological structures mechanism coronarius.In another example can also be to voxels all in left ventricle
Myocardial blood flow is integrated to obtain total myocardial blood flow of left ventricle, utilizes each end based on total myocardial blood flow
The volume ratio of the perfusion area of branch, to determine the flow rate of each end branch.Particularly, similar to the heart of perfusion area above
Total myocardial blood flow of the elaboration of relationship between muscle flow and the associated end branch of blood supply, left ventricle can be considered by each
The corresponding end branch in perfusion area provides jointly, that is, is equal to the summation of the flow rate of each end branch, each end branch
Flow rate proportion be also consistent with the volume ratio of the perfusion area of its blood supply, by total myocardial blood flow of left ventricle multiplied by each
The volume ratio of the perfusion area of end branch, the flow rate of available each end branch.
It in some embodiments, is that arteria coroaria sinistra and arteria coronaria dextra are each in the flowing information coronarius to be determined
From main entrance flow rate and each blood vessel flow-rate ratio distribution in the case where, can determine in the following way described coronal dynamic
The flowing information of arteries and veins.Can be integrated based on myocardial blood flow of the VPCT image to voxel each in left ventricle come
To total myocardial blood flow of left ventricle, it is based on total myocardial blood flow, it is coronal dynamic using the perfusion area of arteria coroaria sinistra and the right side
The myocardial blood flow ratio of the perfusion area of the volume ratio of the perfusion area of arteries and veins, the perfusion area of arteria coroaria sinistra and arteria coronaria dextra and
Any one of arteria coroaria sinistra and the statistics empirical ratios of main entrance flow rate of arteria coronaria dextra, to determine arteria coroaria sinistra
With the respective main entrance flow rate of arteria coronaria dextra.And can in the lumen based on each blood vessel attenuation gradient information or other
Morphology law, to determine the flow-rate ratio distribution of each blood vessel in coronary artery.For example, the main entrance based on arteria coroaria sinistra is flowed
Rate, in conjunction with blood vessel each in arteria coroaria sinistra lumen in attenuation gradient information or other morphology laws (reflection is wherein each
The size distribution relation of blood vessel), it can determine the flow rate of each blood vessel in arteria coroaria sinistra；The calculation method is also applied for right hat
The bifurcated arteries such as shape artery.
It in some embodiments, is main entrance flow rate coronarius and each in the flowing information coronarius to be determined
In the case where the flow-rate ratio distribution of a blood vessel, the flowing information coronarius can be determined in the following way.
Myocardium of left ventricle quality can be calculated based on the Perfusion CT image coronarius of the patient, be then based on the left heart
Room myocardial mass determines the main entrance flow rate coronarius using following formula (2):
Qlv=Qlv0Mlv b, formula (2)
Wherein, QlvIndicate that the main entrance flow rate coronarius is also total myocardial blood flow to left ventricle, MlvIt is left
The myocardial mass of ventricle, Qlv0The attribute constant under the left ventricle current state of the patient is indicated with b.Formula (2) and biology
Friction speed scaling law (allometric scaling law) be consistent, total myocardial blood flow as left ventricle coronarius
When the prediction model of amount, calculates and training is simple, workload is low, and precision of prediction is higher.
And can based on the VPCT image in the perfusion area of each end branch to the myocardial blood flow of each voxel
Amount is integrated, flow rate of the integral result as associated end branch.It can flow rate based on each end branch and the hat
The main entrance flow rate of shape artery, to determine the flow-rate ratio distribution of each end branch.
It in some embodiments, is that either branch artery is (such as but unlimited in the flowing information coronarius to be determined
In left anterior descending branch) flow rate and end branch therein flow rate in the case where, can determine the hat in the following way
The flowing information of shape artery.It can be based on the VPCT image to the myocardium blood of each voxel in the perfusion area of the bifurcated artery
Flow is integrated, or is integrated and multiplied based on myocardial blood flow of the VPCT image to each voxel in left ventricle
With the volume ratio of the perfusion area of the bifurcated artery, to determine the flow rate of the bifurcated artery.And each blood can be based on
Attenuation gradient information or other morphology laws in the lumen of pipe, to determine the flow-rate ratio distribution of each blood vessel in coronary artery,
Flow-rate ratio distribution including each end branch in left anterior descending branch.It is then possible to based on the identified bifurcated artery
Flow rate and the distribution of the flow-rate ratio of each end branch therein, to determine the flow rate of each end branch.
Fig. 7, which is shown, determines flowing information coronarius based on the CCTA image of patient according to the embodiment of the present disclosure
Device 700 configuration.As shown in fig. 7, the device 700 may include acquiring unit 701, extraction unit 702 and predicting unit
705.Acquiring unit 701 is configurable to receive the CCTA image of patient, such as is schemed by CCTA imaging device CCTA collected
Obtained 3D CCTA image is rebuild in picture or thus post-processing, and is transmitted to extraction unit 702.The extraction unit
702 may include artery characteristics extraction unit 703 and myocardial features extraction unit 704, wherein artery characteristics unit 703 can be with
Be configured to received CCTA image zooming-out concern blood vessel artery characteristics, the myocardial features extraction unit 704 can be with
It is configured to pay close attention to the correspondence myocardial features of blood vessel described in acquired CCTA image zooming-out.Extracted concern blood vessel moves
In arteries and veins feature and corresponding myocardial features, the subset of these features needed is at least predicted, it can be for transmission to the predicting unit
705, to utilize the trained prediction model received by it, to predict flowing information coronarius.
Fig. 8, which is shown, determines flowing information coronarius based on the CCTA image of patient according to the embodiment of the present disclosure
Device 800 block diagram.As shown in figure 8, described device 800 may include communication interface 803, it is configured to obtain the patient
CCTA image；And processor 801, it is configured that the artery based on acquired CCTA image zooming-out concern blood vessel is special
Sign；Based on the correspondence myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out；Based on extracted concern blood vessel
At least subset in artery characteristics and myocardial features, using trained prediction model, to estimate the flowing coronarius
Information.
In some embodiments, the communication interface 803 is also configured as: obtaining the filling coronarius of the patient
Infuse the sequence of CT image；The division result of the perfusion area of the left ventricle in the CCTA image of the patient is obtained, with basic herein
Upper processing obtains ground truth and training data.Correspondingly, the processor 801 is also configured as: being based on the patient
Coronary artery determine that volume Perfusion CT (VPCT) image, the VPCT image show it together with the sequence of the Perfusion CT image of heart
In each voxel myocardial blood flow；The division result of perfusion area based on the left ventricle in the CCTA image, by institute
It states VPCT image to be registrated with the CCTA image, to obtain the corresponding perfusion area of the left ventricle in the VPCT image
Division result；Determine the myocardial blood flow in the corresponding perfusion area divided in the left ventricle in the VPCT image, as to this
The flow rate (ground truth) of the concern blood vessel of perfusion area blood supply；Based on the dynamic of concern blood vessel described in acquired CCTA image zooming-out
At least subset in arteries and veins feature and corresponding myocardial features；Artery characteristics and corresponding cardiac muscle based on the extracted concern blood vessel
The flow rate composing training data of at least subset and identified concern blood vessel in feature, to be instructed to the prediction model
Practice.
Processor 801 executes at least by including in the program (such as image processing program 808) that is stored in reservoir 805
Code or the instruction function or method realized, to realize according to the CCTA image based on patient of each embodiment of the disclosure
Method to determine flowing information coronarius, and/or the Perfusion CT figure based on patient according to each embodiment of the disclosure
As come the method that determines flowing information coronarius, and/or the Perfusion CT based on patient according to each embodiment of the disclosure
The method that the flowing information coronarius that image determines prepares training data and is trained to prediction model.
In some embodiments, image processing program 808 includes at least extraction unit 702 shown in fig. 7 and prediction is single
Member 705.In some embodiments, image processing program 808 may further include VPCT image generation unit shown in Fig. 9
902, registration unit 903, VPCT image division unit 904 and flowing information determination unit 905 (see below middle combination Fig. 9's
Detailed description) and training unit (not shown) that prediction model is trained.
The example of processor 801 includes central processing unit (CPU), microprocessing unit (MPU), GPU, microprocessor, place
Manage device core, multiprocessor, specific integrated circuit (ASIC) and field programmable gate array (FPGA) etc..
Memory 804 temporarily stores the program loaded from reservoir 805 and provides workspace to processor 801.It is handling
Device 801 executes the various data generated when program, such as, but not limited to medical image 806, trained prediction model etc.,
It can also be temporarily stored in memory 804.Memory 804 includes such as random access memory (RAM) and read-only memory
(ROM)。
Reservoir 805 stores the program for example executed by processor 801.Reservoir 805 includes such as hard disk drive
(HDD), solid state drive (SSD) and flash memory.
Input/output interface 802 may include input unit and the various processing of output that device 800 inputs various operations
As a result output device.
Communication interface 803 executes sending and receiving for various data via network.Communication can by cable or wirelessly
It executes, and any communication protocol can be used, as long as can communicate with one another.
Each component in device 800 can transmit information via bus 807 each other.Storage medium can store program
In " non-transitory tangible medium ".In addition, the program includes such as software program or computer program.
In addition, at least some of device 800 processing can by the cloud computing by one or more allocation of computer come
It realizes.In some embodiments, at least some of device 800 processing can be executed with another device.In this case, by
At least some of the processing of each functional unit that reason device 801 is realized processing can be executed by replacement device.
Fig. 9, which is shown, determines flowing letter coronarius based on the Perfusion CT image of patient according to the embodiment of the present disclosure
The configuration of the device 900 of breath.As shown in figure 9, the device 900 includes acquiring unit 901, VPCT image generation unit 902, registration
Unit 903, VPCT image division unit 904 and flowing information determination unit 905.
Acquiring unit 901 is configurable to: obtain the patient includes the Perfusion CT image of coronary artery and heart
Sequence, the division result of the perfusion area of the CCTA image and left ventricle therein of the patient.
VPCT image generation unit 902 is configurable to: Perfusion CT of the coronary artery based on the patient together with heart
The sequence of image determines that volume Perfusion CT (VPCT) image, the VPCT image show the myocardial blood flow of wherein each voxel.
Registration unit 903 is configurable to: being registrated to the VPCT image with the CCTA image.
VPCT image division unit 904 is configurable to: being drawn the perfusion area based on the left ventricle in the CCTA image
Point as a result, and registration unit 903 registration result of the VPCT image and the CCTA image is schemed to obtain the VPCT
The division result of the corresponding perfusion area of left ventricle as in.
Flowing information determination unit 905 is configurable to: based on the corresponding perfusion area of left ventricle in the VPCT image
Division result, to determine the flowing information coronarius.For example, flowing information determination unit 905 can be configured further
Are as follows: the myocardial blood flow in the corresponding perfusion area divided in the left ventricle in the VPCT image is determined, as to the perfusion area
The flow rate of the concern blood vessel of blood supply.
This document describes various operations or function, it can be implemented as software code or instruction or be defined as software code
Or instruction.Such content can be the source code or differential code that can directly execute (" object " or " executable " form)
(" delta " or " patch " code).Software code or instruction may be stored in a computer readable storage medium, and work as quilt
When execution, machine can be made to execute described function or operation, and including being used for machine (for example, computing device, electronics
System etc.) addressable form storage information any mechanism, such as recordable or non-recordable medium is (for example, read-only storage
Device (ROM), random access memory (RAM), magnetic disk storage medium, optical storage media, flash memory device etc.).
Illustrative methods described herein can be at least partly machine or computer implemented.Some examples can wrap
The non-transitory computer-readable medium or machine readable media with instruction encoding are included, described instruction can be operated to configure electronics dress
Set the method executed as noted in the example given above.The realization of this method may include software code, such as microcode, compilation language
Say code, more advanced language codes etc..Various software programming techniques can be used to create in various programs or program module.Example
Such as, Java, Python, C, C++, assembler language or any of programming language can be used to design program segment or program mould
Block.Software section as one or more or module can be integrated into computer system and/or computer-readable medium.
This software code may include the computer-readable instruction for executing various methods.Software code can form computer journey
A part of sequence product or computer program module.In addition, in one example, software code can such as during execution or
Other times are tangibly stored in one or more volatibility, non-transitory or non-volatile visible computer readable medium.
The example of these tangible computer-readable mediums can include but is not limited to hard disk, moveable magnetic disc, removable CD (example
Such as, CD and digital video disc), cassette, storage card or stick, random access memory (RAM), read-only memory (ROM) etc..
In addition, although there is described herein illustrative embodiments, range include with based on the disclosure equivalent elements,
Modification is omitted, any and all embodiments of combination (for example, combination of the scheme across various embodiments), adjustment or change.Power
Benefit require in element will be construed broadly as based on language used in claim, and be not limited in this specification or
The example of the duration description of the application.In addition, the step of disclosed method, can modify in any way, including
By rearrangement step or insertion or delete step.Therefore, it is intended that description is only considered as example, real range is by following
Claim and its whole equivalency ranges indicate.
Claims (19)
1. a kind of coronarius to determine based on coronary artery computer tomography angiography (CCTA) image of patient
The method of flowing information, which comprises
Obtain the CCTA image of the patient；
Artery characteristics based on acquired CCTA image zooming-out concern blood vessel；
Based on the correspondence myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out；
At least subset in artery characteristics and corresponding myocardial features based on extracted concern blood vessel, utilizes trained prediction
Model, to predict the flowing information coronarius.
2. the method according to claim 1, wherein based on the dynamic of acquired CCTA image zooming-out concern blood vessel
Arteries and veins feature includes:
Artery segmentation is executed to acquired CCTA image, and extracts the center line of coronary arterial tree；And
From the CCTA image, it is based on extracted center line, extracts the artery characteristics of concern blood vessel, the artery characteristics include
The interior decaying of radius relevant information, length relevant information, volume relevant information, strength related information and lumen of the concern blood vessel
At least one of gradient relevant information.
3. method according to claim 1 or 2, which is characterized in that based on concern described in acquired CCTA image zooming-out
The correspondence myocardial features of blood vessel include:
Left ventricle segmentation is executed to acquired CCTA image；
The perfusion area of each concern blood vessel is marked off on the left ventricle；
Determine that correspondence myocardial features of the myocardial features as the concern blood vessel, the cardiac muscle are special for the perfusion area of each concern blood vessel
Sign includes at least one of myocardial mass and strength related information.
4. the method according to claim 1, wherein the concern blood vessel is coronarius depending on to be determined
Flowing information and prediction model.
5. the method according to claim 1, wherein the concern blood vessel includes each end branch.
6. according to the method described in claim 3, it is characterized in that, the flowing information coronarius includes entire coronal dynamic
The flow-rate ratio of each blood vessel in arteries and veins tree is distributed.
7. according to the method described in claim 3, it is characterized in that, the flowing information coronarius to be determined includes the pass
The flow rate of blood vessel is infused, the prediction model indicates are as follows:
Q=Q0Mb, formula (1)
Wherein, Q indicates the flow rate of the concern blood vessel, and M is the myocardial mass that the perfusion area of blood vessel is paid close attention to described in left ventricle, Q0
The attribute constant under the concern blood vessel current state of the patient is indicated with b.
8. according to the method described in claim 3, it is characterized in that, the prediction model is extracted based on learning network composition
Concern blood vessel artery characteristics and corresponding myocardial features in input of at least subset as the learning network.
9. according to the method described in claim 3, it is characterized in that, the prediction model is using coronal dynamic based on the patient
The flowing information coronarius that arteries and veins is obtained together with perfusion computer tomography (CT) image of heart is true as ground
Value, to be trained.
10. a kind of perfusion computer tomography (CT) image based on patient determines the side of flowing information coronarius
Method, which comprises
The sequence of the Perfusion CT image of the patient is obtained, the Perfusion CT image includes coronary artery and heart；
Determine that volume Perfusion CT (VPCT) image, the VPCT image are shown wherein based on the sequence of the Perfusion CT image of the patient
The myocardial blood flow of each voxel；
It obtains in coronary artery computer tomography angiography (CCTA) image and the CCTA image of the patient
The division result of the perfusion area of left ventricle；
The division result of perfusion area based on the left ventricle in the CCTA image, by the VPCT image and the CCTA
Image is registrated, to obtain the division result of the corresponding perfusion area of left ventricle in the VPCT image；And
Based on the division result of the corresponding perfusion area of left ventricle in the VPCT image, to determine the flowing letter coronarius
Breath.
11. according to the method described in claim 10, it is characterized in that, the corresponding perfusion based on left ventricle in the VPCT image
The division result in area, to determine that the flowing information coronarius includes:
The myocardial blood flow in the corresponding perfusion area divided in the left ventricle in the VPCT image is determined, as to the perfusion area
The flow rate of the concern blood vessel of blood supply.
12. according to the method described in claim 10, it is characterized in that, the method also includes:
Based at least sub in the artery characteristics and corresponding myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out
Collection；
Based on at least subset in the artery characteristics and corresponding myocardial features of the extracted concern blood vessel and determine
The flowing information composing training data coronarius.
13. according to the method described in claim 10, it is characterized in that, being each in the flowing information coronarius to be determined
In the case where the flow rate of end branch, the hat can be determined by following any method based on the VPCT image
The flowing information of shape artery:
In the perfusion area of each end branch, the myocardial blood flow of each voxel is integrated, integral result is as corresponding
The flow rate of end branch；
The myocardial blood flow of voxels all in left ventricle is integrated to obtain total myocardial blood flow of left ventricle, based on described
Total myocardial blood flow utilizes the volume ratio of the perfusion area of each end branch, to determine the flow rate of each end branch.
14. according to the method described in claim 10, it is characterized in that, being left hat in the flowing information coronarius to be determined
It, can be by as follows in the case that shape artery and the respective main entrance flow rate of arteria coronaria dextra and the flow-rate ratio of each blood vessel are distributed
Mode determines the flowing information coronarius:
It is integrated based on myocardial blood flow of the VPCT image to voxel each in left ventricle to obtain total heart of left ventricle
Muscle flow is based on total myocardial blood flow, utilizes the body of the perfusion area of the perfusion area and arteria coronaria dextra of arteria coroaria sinistra
The myocardial blood flow ratio and arteria coroaria sinistra of the perfusion area of product ratio, the perfusion area of arteria coroaria sinistra and arteria coronaria dextra and the right side
Any one of the statistics empirical ratios of main entrance flow rate coronarius, to determine that arteria coroaria sinistra and arteria coronaria dextra are each
From main entrance flow rate；
Attenuation gradient information or other morphology laws in lumen based on each blood vessel, to determine each blood vessel in coronary artery
Flow-rate ratio distribution.
15. according to the method described in claim 10, it is characterized in that, being coronal in the flowing information coronarius to be determined
In the case where the main entrance flow rate of artery and the flow-rate ratio distribution of each blood vessel, it can determine in the following way described coronal
The flowing information of artery:
Perfusion CT image coronarius based on the patient calculates myocardium of left ventricle quality, based on myocardium of left ventricle quality benefit
With following formula (2), to determine the main entrance flow rate coronarius:
Qlv=Qlv0Mlv b, formula (2)
Wherein, QlvIndicate that the main entrance flow rate coronarius is also total myocardial blood flow to left ventricle, MlvIt is left ventricle
Myocardial mass, Qlv0The attribute constant under the left ventricle current state of the patient is indicated with b；And
The myocardial blood flow of each voxel is integrated in the perfusion area of each end branch based on the VPCT image, product
Divide flow rate of the result as associated end branch, flow rate and the main entrance coronarius based on each end branch flow
Rate determines the flow-rate ratio distribution of each end branch.
16. according to the method described in claim 10, it is characterized in that, being branch in the flowing information coronarius to be determined
In the case where the flow rate of artery and the flow rate of end branch therein, it can determine in the following way described coronarius
Flowing information:
It is integrated based on myocardial blood flow of the VPCT image to each voxel in the perfusion area of the bifurcated artery, or
Person is integrated based on myocardial blood flow of the VPCT image to each voxel in left ventricle and multiplied by the bifurcated artery
Perfusion area volume ratio, to determine the flow rate of the bifurcated artery；
Attenuation gradient information or other morphology laws in lumen based on each blood vessel, to determine each blood vessel in coronary artery
Flow-rate ratio distribution；
It is described to determine based on the flow rate of the identified bifurcated artery and the flow-rate ratio distribution of each end branch therein
The flow rate of each end branch.
17. a kind of coronarius to determine based on coronary artery computer tomography angiography (CCTA) image of patient
The device of flowing information, described device include:
Interface is configured to obtain the CCTA image of the patient；And
Processor is configured that
Artery characteristics based on acquired CCTA image zooming-out concern blood vessel；
Based on the correspondence myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out；
At least subset in artery characteristics and myocardial features based on extracted concern blood vessel, utilizes trained prediction mould
Type, to predict the flowing information coronarius.
18. device according to claim 17, which is characterized in that
The interface is additionally configured to:
Obtain the sequence of the Perfusion CT image coronarius of the patient；
Obtain the division result of the perfusion area of the left ventricle in the CCTA image of the patient；The processor is additionally configured to:
Volume Perfusion CT (VPCT) image is determined together with the sequence of the Perfusion CT image of heart based on the coronary artery of the patient,
The VPCT image shows the myocardial blood flow of wherein each voxel；
The division result of perfusion area based on the left ventricle in the CCTA image, by the VPCT image and the CCTA
Image is registrated, come obtain the left ventricle in the VPCT image corresponding perfusion area division result；
The myocardial blood flow in the corresponding perfusion area divided in the left ventricle in the VPCT image is determined, as to the perfusion area
The flow rate of the concern blood vessel of blood supply；
Based on at least subset in the artery characteristics and corresponding myocardial features for paying close attention to blood vessel described in acquired CCTA image zooming-out；
Based on the extracted concern blood vessel artery characteristics and corresponding myocardial features at least subset and identified
The flow rate composing training data of blood vessel are paid close attention to, to be trained to the prediction model.
19. a kind of non-transitory computer-readable medium, is stored thereon with instruction, described instruction is when executed by the processor, real
The now coronary artery computer tomography angiography described in any one of -9 based on patient according to claim 1
(CCTA) image is come the method that determines flowing information coronarius, and/or realizes any one of 0-16 according to claim 1
The method for determining flowing information coronarius based on perfusion computer tomography (CT) image of patient.
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 true CN110428420A (en) | 2019-11-08 |
| CN110428420B 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) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN112336365A (en) * | 2020-11-11 | 2021-02-09 | 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 |
| CN112967220A (en) * | 2019-11-28 | 2021-06-15 | 西门子医疗有限公司 | Computer-implemented method of evaluating a CT data set relating to perivascular tissue |
| 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 |
| CN116994067A (en) * | 2023-09-07 | 2023-11-03 | 佛山科学技术学院 | Method and system for predicting fractional flow reserve based on coronary artery calcification |
Citations (4)
| 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 |
-
2019
- 2019-09-05 CN CN201910837899.5A patent/CN110428420B/en active Active
Patent Citations (4)
| 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 |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112967220A (en) * | 2019-11-28 | 2021-06-15 | 西门子医疗有限公司 | Computer-implemented method of evaluating a CT data set relating to perivascular tissue |
| US11803966B2 (en) | 2019-11-28 | 2023-10-31 | Siemens Healthcare Gmbh | Computer-implemented method for evaluating a CT data set regarding perivascular tissue, evaluation device, computer program and electronically readable storage medium |
| CN112967220B (en) * | 2019-11-28 | 2024-05-14 | 西门子医疗有限公司 | Computer-implemented method of evaluating CT data sets relating to perivascular tissue |
| 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 |
| CN112336365A (en) * | 2020-11-11 | 2021-02-09 | 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 |
| CN113450322B (en) * | 2021-06-22 | 2024-05-24 | 东软医疗系统股份有限公司 | Method and device for judging provenance abnormality of coronary artery |
| CN116994067A (en) * | 2023-09-07 | 2023-11-03 | 佛山科学技术学院 | Method and system for predicting fractional flow reserve based on coronary artery calcification |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110428420B (en) | 2022-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20230139102A1 (en) | Method and system for quantifying limitations in coronary artery blood flow during physical activity in patients with coronary artery disease | |
| US20230218347A1 (en) | Systems and methods for processing electronic images to simulate flow | |
| CN110428420A (en) | The method, apparatus and medium of flowing information coronarius are determined based on the coronary artery CT angiographic image of patient | |
| US20220406470A1 (en) | Systems and methods for processing electronic images to determine flow using flow ratio | |
| US11389130B2 (en) | System and methods for fast computation of computed tomography based fractional flow reserve | |
| JP6215469B2 (en) | Method and system for modeling blood flow using boundary conditions for optimized diagnostic capabilities | |
| US10134129B2 (en) | Method and system for hemodynamic computation in coronary arteries | |
| US20170039340A1 (en) | Methods and systems for predicting sensitivity of blood flow calculations to changes in anatomical geometry | |
| EP4002269A1 (en) | Systems and methods for image-based object modeling using multiple image acquisitions or reconstructions | |
| CN110944586B (en) | Inflammation Estimation from X-ray Image Data | |
| EP3695784A1 (en) | Prediction of coronary microvascular dysfunction from coronary computed tomography | |
| CN110444275B (en) | System and method for rapid calculation of fractional flow reserve | |
| CN110506293A (en) | Spectrum imaging FFR | |
| EP3949831A2 (en) | Learning method for generating multiphase collateral image and multiphase collateral image generating method using machine learning | |
| US11810290B2 (en) | Method and computer system for generating a combined tissue-vessel representation | |
| US20210219850A1 (en) | Providing a blood flow parameter set for a vascular malformation | |
| US20230030618A1 (en) | Making measurements in images | |
| JP7443197B2 (en) | Medical image processing device, system and method | |
| Ogiela et al. | Cognitive methods for semantic image analysis in medical imaging applications | |
| CN111815590A (en) | Method and system for acquiring heart gravity center and spine gravity center based on CT sequence image |
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 |