WO2022160973A1

WO2022160973A1 - Coronary fractional flow reserve obtaining system and method, and medium

Info

Publication number: WO2022160973A1
Application number: PCT/CN2021/137375
Authority: WO
Inventors: 房劬; 赵夕
Original assignee: 上海杏脉信息科技有限公司
Priority date: 2021-01-26
Filing date: 2021-12-13
Publication date: 2022-08-04
Also published as: CN112950537A

Abstract

A coronary fractional flow reserve obtaining system and method, and a medium. The coronary fractional flow reserve obtaining system (1) comprises: a medical image obtaining module (11) used for obtaining a medical image of a patient, the medical image comprising a heart region of a patient; a coronary artery model obtaining module (12) connected to the medical image obtaining module (11) and used for obtaining a first coronary artery model and a second coronary artery model of the patient according to the medical image, wherein the first coronary artery model refers to an actual coronary artery model of the patient, and the second coronary artery model refers to a coronary artery model obtained after repairing at least one lesion of a coronary vessel of the patient; and a fractional flow reserve obtaining module (13) connected to the coronary artery model obtaining module (12) and used for obtaining the coronary fractional flow reserve of the patient according to the first coronary artery model and the second coronary artery model. Compared with the prior art, the coronary fractional flow reserve obtained by the coronary fractional flow reserve obtaining system (1) has higher accuracy.

Description

A system, method and medium for obtaining fractional coronary blood flow reserve

technical field

The invention belongs to the field of medical image processing, and relates to a system for obtaining coronary blood flow reserve fraction, in particular to a method for processing medical images of coronary parts of the heart by using computer image processing technology to obtain coronary blood flow dynamic indexes system, method and medium.

Background technique

Myocardial ischemia is usually caused by coronary artery (coronary artery) stenosis, which can be life-threatening in severe cases. Fractional Flow Reserve (FFR) is an effective indicator to measure the coronary blood supply in humans. Fractional coronary flow reserve refers to the ratio of the maximum blood flow Q _S that can be obtained in the myocardial area supplied by the blood vessel under the condition of coronary stenosis and the maximum blood flow Q _N that can theoretically be obtained in the same area under normal conditions. ,which is:

The methods of obtaining FFR in the prior art can be divided into invasive methods and non-invasive methods.

The method of obtaining FFR by an invasive method refers to invasively inserting a pressure guide wire into a coronary artery under the guidance of medical images, measuring the pressure at the distal end of the stenosis and the pressure at the proximal end of the stenosis, and calculating the FFR index. Invasive measurement methods cause great damage to patients, which limits the clinical application of FFR indicators.

The method of obtaining FFR by non-invasive method is to obtain the FFR index according to the medical image of the patient. In practical applications, both Q _S and Q _N are indicators that are difficult to obtain.

Therefore, in practice, it is often used

replace

To calculate FFR, at this time, the formula to obtain FFR is

That is: the mean coronary pressure distal to the stenosis in the state of maximal myocardial hyperemia

Coronary Oral Aortic Mean Pressure

ratio. However, the inventors found in practical applications,

and

There are often differences between the

replace

to calculate the FFR introduces additional error that affects the accuracy of the FFR. In particular, when the stenosis position of the coronary artery is far from the aorta, the error is also larger, and the accuracy of the obtained FFR is lower at this time.

SUMMARY OF THE INVENTION

In view of the above shortcomings of the prior art, the purpose of the present invention is to provide a system, method and medium for obtaining fractional coronary blood flow reserve, which are used to solve the problem of low accuracy of the obtained FFR in the prior art.

In order to achieve the above object and other related objects, a first aspect of the present invention provides a system for obtaining fractional coronary blood flow reserve, the system for obtaining fractional coronary blood flow reserve includes: a medical image obtaining module for obtaining the medical image of a patient. an image, the medical image includes the heart region of the patient; a coronary artery model acquisition module, connected to the medical image acquisition module, is used to acquire the first coronary artery model and the second coronary artery model of the patient according to the medical image; wherein , the first coronary artery model refers to the actual coronary artery model of the patient, and the second coronary artery model refers to the coronary artery model obtained after repairing at least one lesion of the coronary artery of the patient; the blood flow reserve fraction is obtained The module is connected to the coronary artery model acquisition module, and is used for acquiring the coronary blood flow reserve fraction of the patient according to the first coronary artery model and the second coronary artery model.

In an embodiment of the first aspect, the coronary artery model acquisition module includes a first coronary artery model acquisition sub-module and a second coronary artery model acquisition sub-module; the first coronary artery model acquisition sub-module and the The medical image acquisition module is connected to the medical image to obtain the first coronary artery model by segmenting the medical image; the second coronary artery model acquisition sub-module is connected to the first coronary artery model acquisition sub-module, using performing simulated repair on at least one lesion of the patient's coronary vessel in the first coronary model to obtain the second coronary model; or, the second coronary model obtaining sub-module and the medical image The acquisition module is connected to the medical image for simulating repair of at least one lesion of the coronary blood vessel of the patient, and segmenting the medical image after the simulating repair to obtain the second coronary artery model.

In an embodiment of the first aspect, the second coronary artery model acquisition sub-module includes: a first lesion parameter acquisition unit connected to the first coronary artery model acquisition sub-module for acquiring the first coronary artery model acquisition sub-module. a lesion parameter related to coronary artery stenosis in a coronary artery model; a first simulated repair unit, connected to the first lesion parameter acquisition unit, and configured to perform an analysis on the stenosis of the first coronary artery model according to the lesion parameter Modify to obtain the second coronary model.

In an embodiment of the first aspect, the second coronary artery model acquisition sub-module includes: a second lesion parameter acquisition unit, which is connected to the image acquisition module and is used to acquire the coronary artery in the medical image. stenosis-related lesion parameters; a second simulated repair unit, connected to the second lesion parameter acquisition unit, configured to modify the stenosis of the patient's coronary blood vessels in the medical image according to the lesion parameters, so as to obtain the The medical image after simulating restoration; the image segmentation unit is connected to the second simulating restoration unit, and is used for segmenting the medical image after simulating restoration, so as to obtain the second coronary artery model.

In an embodiment of the first aspect, the coronary artery model acquisition module includes a third coronary artery model acquisition sub-module and a fourth coronary artery model acquisition sub-module; the third coronary artery model acquisition sub-module and the The medical image acquisition module is connected to the medical image to obtain the first coronary artery model by segmenting the medical image; the fourth coronary artery model acquisition sub-module is connected to the third coronary artery model acquisition sub-module, using performing virtual treatment on at least one lesion of the coronary vessel of the patient in the first coronary model to obtain the second coronary model; or the fourth coronary model obtaining sub-module and the medical image obtaining The modules are connected, and are used for performing virtual treatment on at least one lesion of the coronary blood vessel of the patient in the medical image, and segmenting the medical image after the virtual treatment to obtain the second coronary artery model.

In an embodiment of the first aspect, the blood flow reserve fraction acquisition module includes: an actual blood flow acquisition unit, connected to the coronary artery model acquisition module, for acquiring a the actual maximum blood flow of the target blood supply area; the ideal blood flow acquisition unit, connected to the coronary artery model acquisition module, is used to acquire the ideal maximum blood flow of the target blood supply area according to the second coronary artery model; reserve fraction acquisition a unit, connected to the actual blood flow obtaining unit and the ideal blood flow obtaining unit, and configured to obtain the coronary blood flow reserve fraction according to the actual maximum blood flow and the ideal maximum blood flow of the target blood supply area.

In an embodiment of the first aspect, the blood flow reserve fraction acquisition module includes: an actual pressure acquisition unit, which is connected to the coronary artery model acquisition module, and is used for acquiring myocardial maximum value according to the first coronary artery model. The actual average pressure of the target position in the hyperemia state, wherein the target position refers to the distal end of coronary artery stenosis; the ideal pressure acquisition unit is connected to the coronary artery model acquisition module, and is used for obtaining according to the second coronary artery The model obtains the ideal average pressure of the target position under the state of maximum myocardial hyperemia; the reserve fraction obtaining unit is connected to the actual pressure obtaining unit and the ideal pressure obtaining unit, and is used for obtaining the actual average pressure and the ideal pressure according to the target position. The mean pressure was used to obtain the fractional coronary flow reserve.

In an embodiment of the first aspect, the system for obtaining fractional coronary blood flow reserve further includes: a user interaction module, connected to the medical image obtaining module and/or the coronary model obtaining module, for Displaying the medical image, the first coronary artery model and/or the second coronary artery model, and being used to obtain model generation instructions and/or model editing instructions input by the user; the coronary artery model obtaining module is based on the The model generating instruction acquires the second coronary artery model, and/or the coronary artery model acquiring module edits the first coronary artery model and/or the second coronary artery model according to the model editing instruction.

A second aspect of the present invention provides a method for obtaining fractional coronary blood flow reserve, the method for obtaining fractional coronary blood flow reserve includes: obtaining a medical image of a patient, the medical image including a heart region of the patient; The first coronary artery model and the second coronary artery model of the patient are obtained from the images; wherein, the first coronary artery model refers to the actual coronary artery model of the patient, and the second coronary artery model refers to at least the coronary artery model of the patient. Coronary artery model obtained after a lesion is repaired; the coronary blood flow reserve fraction of the patient is obtained according to the first coronary artery model and the second coronary artery model.

A third aspect of the present invention provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the method for obtaining fractional coronary blood flow reserve described in the second aspect of the present invention.

As mentioned above, a technical solution of the system, method and medium for obtaining fractional coronary blood flow reserve according to the present invention has the following beneficial effects:

The system for obtaining fractional coronary blood flow reserve of the present invention can obtain the patient's first coronary model according to the patient's medical image, and obtain the patient's second coronary model according to the patient's medical image or the first coronary model. Model. Wherein, the first coronary artery model refers to the actual coronary artery model of the patient, and the second coronary artery model refers to the coronary artery model obtained after repairing at least one lesion of the coronary artery of the patient. According to the first coronary artery model, the maximum blood flow Q _S that can be obtained in the myocardial region supplied by the blood vessel under the condition of coronary stenosis and lesions can be obtained, and according to the second coronary artery model, the theoretical normal condition of the same region can be obtained. The maximum blood flow Q _N that can be obtained, the coronary blood flow reserve fraction of the patient can be directly obtained based on Q _S and Q _N . Therefore, the system for obtaining the fractional coronary blood flow reserve of the present invention does not adopt the method for obtaining the fractional coronary blood flow reserve.

to replace

Therefore, no additional errors will be introduced. Therefore, the fractional coronary blood flow reserve obtained by the system for obtaining the fractional coronary blood flow reserve of the present invention has higher accuracy than the prior art, especially in the case of coronary stenosis. When the location is far away from the aorta, the advantages of the system for obtaining fractional coronary blood flow reserve of the present invention are more obvious.

Description of drawings

FIG. 1 shows a schematic structural diagram of a system for obtaining fractional coronary blood flow reserve according to the present invention in a specific embodiment.

2A is a schematic structural diagram of a coronary artery model acquisition module in a specific embodiment of the coronary blood flow fraction reserve acquisition system according to the present invention.

FIG. 2B is a schematic structural diagram of a second coronary artery model acquisition sub-module in an embodiment of the coronary blood flow fraction reserve acquisition system according to the present invention.

FIG. 2C is a diagram showing an example of part of blood vessels of the first coronary artery model of the system for obtaining fractional coronary flow reserve according to the present invention.

FIG. 2D is a diagram showing an example of part of blood vessels of the second coronary artery model of the system for obtaining fractional coronary flow reserve according to the present invention.

FIG. 3A is a schematic structural diagram of a coronary artery model acquisition module in a specific embodiment of the coronary flow reserve fraction acquisition system according to the present invention.

3B is a schematic structural diagram of a second coronary artery model acquisition sub-module in a specific embodiment of the coronary blood flow fraction reserve acquisition system according to the present invention.

FIG. 4A is a schematic structural diagram of a coronary artery model acquisition module in a specific embodiment of the coronary flow reserve fraction acquisition system according to the present invention.

FIG. 4B is a schematic structural diagram of a coronary artery model acquisition module in a specific embodiment of the coronary blood flow fraction reserve acquisition system according to the present invention.

FIG. 5A is a schematic structural diagram of a fractional blood flow reserve acquisition module in a specific embodiment of the coronary blood flow fraction acquisition system according to the present invention.

FIG. 5B is a schematic structural diagram of a fractional blood flow reserve acquisition module in a specific embodiment of the coronary blood flow reserve fraction acquisition system according to the present invention.

FIG. 6 shows a flowchart of the method for obtaining fractional coronary flow reserve according to the present invention in a specific embodiment.

Component label description

1 Coronary flow reserve fraction acquisition system

11 Medical Image Acquisition Module

12 coronary artery model acquisition module

121 First coronary artery model acquisition sub-module

122 Second coronary artery model acquisition sub-module

1221 The first lesion parameter acquisition unit

1222 First Simulation Repair Unit

1223 Second lesion parameter acquisition unit

1224 Second Simulation Repair Unit

1225 image segmentation unit

123 Third coronary artery model acquisition sub-module

124 Fourth coronary artery model acquisition sub-module

13. Fractional flow reserve acquisition module

131 Actual blood flow acquisition unit

132 ideal blood flow acquisition unit

133 Reserve Score Acquisition Unit

134 Actual pressure acquisition unit

135 ideal pressure acquisition unit

136 Reserve Score Acquisition Unit

Steps from S61 to S63

Detailed ways

The embodiments of the present invention are described below through specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other under the condition of no conflict.

It should be noted that the drawings provided in the following embodiments are only to illustrate the basic concept of the present invention in a schematic way, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in actual implementation. For drawing, the type, quantity and proportion of each component can be arbitrarily changed during actual implementation, and the layout of components may also be more complicated. Furthermore, herein, relational terms such as "first," "second," etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply those entities or that there is any such actual relationship or sequence between operations.

In practical applications, both Q _S and Q _N are indicators that are difficult to obtain.

Therefore, in practice, it is often used

replace

To calculate FFR, at this time, the formula to obtain FFR is

Coronary Oral Aortic Mean Pressure

ratio. However, the inventors found in practical applications that using

replace

To calculate FFR relies on three assumptions, specifically:

At this time, assuming R _S =R _N , we get

Then suppose

get

final hypothesis

get

In the above formula, Q represents the maximum blood flow, P represents the pressure, R represents the coronary microcirculation resistance, the subscript S represents the actual situation of coronary stenosis, the subscript N represents the ideal situation of no coronary stenosis, and the superscript a indicates the proximal end of the stenosis (usually the aorta at the coronary ostium), the superscript d indicates the distal end of the stenosis, and the superscript v indicates the venous position, for example,

represents the actual venous pressure,

Indicates the pressure distal to the location of the coronary artery stenosis ideally. From this, it can be seen that using

replace

to calculate FFR depends on R _S =R _N ,

as well as

These three assumptions (equivalent to

This assumption), however, the above three assumptions are often not strictly true in practical applications, especially

this assumption. When the stenosis of the coronary artery is far from the aorta, it is assumed that

Larger errors will be introduced, resulting in lower accuracy of the obtained FFR.

In response to this problem, the present invention provides a system for obtaining fractional coronary blood flow reserve. The coronary blood flow reserve fraction acquisition system can acquire the patient's first coronary artery model according to the patient's medical image, and acquire the patient's second coronary artery model according to the patient's medical image or the first coronary artery model. Wherein, the first coronary artery model refers to the actual coronary artery model of the patient, and the second coronary artery model refers to the coronary artery model obtained after repairing at least one lesion of the coronary artery of the patient. According to the first coronary artery model, the maximum blood flow Q _S that can be obtained in the myocardial region supplied by the blood vessel under the condition of coronary stenosis and lesions can be obtained, and according to the second coronary artery model, the theoretical normal condition of the same region can be obtained. The maximum blood flow Q _N that can be obtained, the coronary blood flow reserve fraction of the patient can be directly obtained based on Q _S and Q _N . Therefore, the system for obtaining the fractional coronary blood flow reserve of the present invention does not adopt the method for obtaining the fractional coronary blood flow reserve.

to replace

Referring to FIG. 1 , in an embodiment of the present invention, the system 1 for acquiring fractional coronary blood flow reserve includes a medical image acquiring module 11 , a coronary artery model acquiring module 12 and a fractional blood flow reserve acquiring module 13 .

The medical image acquisition module 11 is used for acquiring a medical image of a patient, and the medical image of the patient includes a heart region of the patient. The medical image is preferably a CT angiography (CTA, CT angiography) image, in addition, the medical image can also be a CT perfusion image, a plain CT image, a DSA angiography image, or an image obtained by using methods such as X-ray, nuclear magnetic resonance, The medical images of the heart part obtained by scanning imaging methods such as ultrasound, PET, SPECT, etc., may also be intravascular images (eg, optical coherence imaging, intravascular ultrasound), and the like.

The coronary artery model acquisition module 12 is connected with the medical image acquisition module 11, and is used to acquire the patient's first coronary artery model and the second coronary artery model according to the medical image of the patient, wherein the first coronary artery model The model is an actual coronary artery model of the patient, and the first coronary artery model contains at least one stenotic lesion of the patient. The second coronary artery model refers to a coronary artery model obtained after repairing at least one lesion of a patient's coronary artery. The coronary artery model acquisition module 12 may acquire the second coronary artery model according to the first coronary artery model, and may also acquire the second coronary artery model according to the medical image of the patient. In a specific application, the coronary artery model obtaining module 12 may repair at least one lesion of the patient's coronary vessels by using automatic algorithm repair, manual repair by a user, or a combination of the two.

The fractional blood flow reserve acquisition module 13 is connected to the coronary artery model acquisition module 12, and is configured to acquire the coronary blood flow reserve fraction of the patient according to the first coronary artery model and the second coronary artery model.

It can be seen from the above description that the system for obtaining fractional coronary blood flow reserve according to this embodiment obtains the fractional coronary blood flow reserve of the patient according to the first coronary model and the second coronary model, instead of using

to replace

to obtain the patient's coronary flow reserve fraction without introducing additional errors. Therefore, the fractional coronary blood flow reserve obtained in this embodiment has a higher accuracy than the prior art, especially when the stenosis position of the coronary artery is far away from the aorta, the coronary blood flow described in this embodiment is more accurate. The advantage of the reserve score acquisition system is even more obvious.

In an embodiment of the present invention, the medical image of the patient is a 3D image. The three-dimensional image obtained after processing and calculating the two-dimensional image (such as DSA). In this embodiment, the first coronary artery model and the second coronary artery model are both three-dimensional geometric models of the coronary artery.

In an embodiment of the present invention, the coronary artery model acquisition module includes a first coronary artery model acquisition sub-module and a second coronary artery model acquisition sub-module.

Optionally, please refer to FIG. 2A , the first coronary artery model acquisition sub-module 121 is connected to the medical image acquisition module 11, and is used for segmenting the medical image to acquire the first coronary artery model. The second coronary artery model acquisition sub-module 122 is connected to the first coronary artery model acquisition sub-module 121, and is used to simulate and repair at least one lesion of the coronary artery of the patient in the first coronary artery model, to Obtain the second coronary artery model.

The first coronary artery model obtaining sub-module 121 may use a neural network-based image segmentation model (eg, U-Net, V-Net, etc.) to segment the medical image to obtain the first coronary artery model. Specifically, the first coronary artery model obtaining sub-module 121 inputs the medical image into the image segmentation model, and the first coronary artery model can be obtained according to the output of the image segmentation model.

The first coronary artery model obtaining sub-module 121 may also use a threshold method to segment the medical image to obtain the first coronary artery model. Specifically, the first coronary artery model obtaining sub-module 121 obtains the gray value range of coronary blood vessels, and obtains all voxels (or pixel points) located within the gray value range from the medical image, The set composed of these voxel points (or pixel points) is the first coronary artery model.

Referring to FIG. 2B , an implementation structure of the second coronary artery model obtaining sub-module 122 includes a first lesion parameter obtaining unit 1221 and a first simulated repairing unit 1222 .

The first lesion parameter acquisition unit 1221 is connected to the first coronary artery model acquisition sub-module 121, and is configured to acquire lesion parameters related to coronary stenosis in the first coronary artery model. Wherein, the lesion parameters are, for example, the stenosis position of the coronary artery, the blood vessel centerline of the stenosis position, the cross-section of the proximal stenosis and the cross-section of the distal end of the stenosis (or, the diameter of the blood vessel at the proximal end of the stenosis and the diameter of the distal end of the stenosis). vessel diameter). In a specific application, the first lesion parameter obtaining unit 1221 may use a stenosis detection algorithm to obtain at least one stenosis position in the first coronary artery model, and use an existing geometric method to obtain the stenosis position Vessel centerline, cross-section proximal to the stenosis, and cross-section distal to the stenosis (or, vessel diameter proximal to the stenosis and vessel diameter distal to the stenosis).

An implementation manner of the stenosis detection algorithm is: according to the coronary vessels obtained by segmentation and their centerlines, calculate the diameters or cross-sections of different positions of the blood vessels along the centerline, and select the positions with diameters or cross-sections smaller than a threshold value as the coronary arteries. stenosis of the arteries.

Another implementation manner of the stenosis detection algorithm is: using an AI stenosis detection model to process the patient's first coronary artery model to obtain the coronary stenosis position. The AI stenosis detection model is a trained deep learning network model, and its training data includes multiple coronary images marked with stenosis positions. In specific applications, the stenosis positions of coronary images can be manually marked ; The training of the AI stenosis detection model can be implemented by using an existing training method, which will not be repeated here.

The first simulated repair unit 1222 is connected to the first lesion parameter acquisition unit 1221, and is configured to modify the stenosis of the first coronary artery model according to the lesion parameters, so as to obtain the stenosis position when the stenosis does not appear. The ideal blood vessel in stenotic lesions is obtained, and then the second coronary artery model is obtained.

Specifically, for any stenosis position B, the first simulated repair unit 1222 may take the blood vessel centerline of the stenosis position B as the axis of symmetry, and the cross-section of the proximal stenosis and the cross-section of the distal stenosis as the The end face generates a geometry of a specific shape, and uses the geometry of the specific shape to replace the blood vessel at the stenosis position B to obtain the second coronary artery model. Alternatively, the first simulated repairing unit 1222 can use the blood vessel centerline of the stenosis position B as the axis of symmetry, and generate a geometry with a specific shape according to the blood vessel diameter at the proximal end of the stenosis and the blood vessel diameter at the distal end of the stenosis, and use the The geometry of the specific shape replaces the blood vessel at the stenosis position B to obtain the second coronary artery model. For example, please refer to Fig. 2C and Fig. 2D, wherein Fig. 2C shows an example diagram of a vessel at a stenosis position in the first coronary artery model, and Fig. 2D shows the result of modifying the vessel at the stenosis position shown in Fig. 2C.

Optionally, please refer to FIG. 3A , the first coronary artery model obtaining sub-module 121 is connected to the medical image obtaining module 11 , and is used for segmenting the medical image to obtain the first coronary artery model. The second coronary artery model acquisition sub-module 122 is connected to the medical image acquisition module 11, and is used to simulate and repair at least one lesion of the patient's coronary vessels in the medical image, and to simulate and repair the medical image after the repair. Segmentation is performed to obtain the second coronary artery model.

Referring to FIG. 3B , an implementation structure of the second coronary artery model acquisition sub-module 122 includes a second lesion parameter acquisition unit 1223 , a second simulated repair unit 1224 and an image segmentation unit 1225 .

The second lesion parameter acquisition unit 1223 is connected to the image acquisition module 11, and is configured to acquire lesion parameters related to coronary stenosis in the medical image. Wherein, the lesion parameters are, for example, the stenosis position of the coronary artery, the blood vessel centerline of the stenosis position, the cross-section of the proximal stenosis and the cross-section of the distal end of the stenosis (or, the diameter of the blood vessel at the proximal end of the stenosis and the diameter of the distal end of the stenosis). vessel diameter). In a specific application, the second lesion parameter obtaining unit 1223 may use a stenosis detection algorithm to obtain at least one stenosis position in the medical image, and use an existing geometric method to obtain the blood vessel centerline of the stenosis position , a cross-section of the proximal end of the stenosis and a cross-section of the distal end of the stenosis (or, the vessel diameter at the proximal end of the stenosis and the vessel diameter at the distal end of the stenosis).

The second simulated repair unit 1224 is connected to the second lesion parameter acquisition unit 1223, and is configured to modify the stenosis of the patient's coronary blood vessels in the medical image according to the lesion parameters, so as to obtain the simulated repair after medical images. The modification performed by the second simulation repair unit 1224 on the stenosis of the coronary blood vessel of the patient is similar to that of the first simulation unit 1222 , and details are not described here.

The image segmentation unit 1225 is connected to the second simulated restoration unit 1224, and is used for segmenting the medical image after the simulated restoration to obtain the second coronary artery model. Specifically, the image segmentation unit 1225 may use a neural network-based image segmentation model or a threshold method to segment the medical image after the simulated restoration.

As can be seen from the above description, this embodiment provides a method for automatically repairing a patient's coronary vascular disease by using an algorithm to obtain a second coronary model. In specific applications, the second coronary model can be directly used to obtain the coronary blood flow of the patient. The reserve fraction can also be further revised manually on the basis of the second coronary artery model to obtain a more accurate second coronary artery model.

In an embodiment of the present invention, the coronary artery model acquisition module includes a third coronary artery model acquisition sub-module and a fourth coronary artery model acquisition sub-module.

Optionally, referring to FIG. 4A , the third coronary artery model obtaining sub-module 123 is connected to the medical image obtaining module 11 , and is used for segmenting the medical image to obtain the first coronary artery model. The fourth coronary artery model acquisition sub-module 124 is connected to the third coronary artery model acquisition sub-module 123, and is used to perform virtual treatment on at least one lesion of the coronary artery of the patient in the first coronary artery model, so as to Obtain the second coronary artery model. The manner of segmenting the medical image to obtain the first coronary artery model is similar to that of the first coronary artery model obtaining sub-module 121 , and details are not described here.

After acquiring the first coronary artery model, the fourth coronary artery model acquisition sub-module 124 uses a stenosis detection algorithm to acquire the stenosis position of the coronary artery in the first coronary artery model. The arterial model obtaining sub-module 124 performs virtual treatment on at least one stenosis position of the coronary blood vessel of the patient. The virtual treatment method is, for example, a virtual stent implantation technique or a virtual balloon dilation technique. The coronary artery obtained after the virtual treatment is completed. The model is the second coronary artery model.

Specifically, the virtual stent implantation technology refers to that the fourth coronary artery model acquisition sub-module 124 implants a virtual stent into the stenosis position, so that the blood vessel in the stenosis position can be supported by the virtual stent to restore the normal state, so as to realize the virtual treatment.

The virtual balloon dilation technique means that the fourth coronary artery model acquisition sub-module 124 implants a virtual balloon implant into the stenosis position, and the virtual balloon implant has a load inside, and when After the virtual balloon implant is implanted into the three-dimensional model of the blood vessel, its internal load will expand under the action of external force, which will cause the balloon implant to produce plastic deformation, thereby supporting the blood vessel at the stenotic position. The virtual treatment is realized by external expansion and eventually return to normal state. Wherein, the method for load expansion in the virtual balloon implant includes, but is not limited to, inflating it.

Optionally, referring to FIG. 4B , the third coronary artery model obtaining sub-module 123 is connected to the medical image obtaining module 11 , and is used for segmenting the medical image to obtain the first coronary artery model. The fourth coronary artery model acquisition sub-module 124 is connected to the medical image acquisition module 11, and is used to perform virtual treatment on at least one lesion of the coronary artery of the patient in the medical image, and perform virtual treatment on the medical image after virtual treatment. Segmentation is performed to obtain the second coronary artery model.

Referring to FIG. 5A , in an embodiment of the present invention, the fractional blood flow reserve acquiring module 13 includes an actual blood flow acquiring unit 131 , an ideal blood flow acquiring unit 132 and a fractional reserve acquiring unit 133 .

The actual blood flow acquisition unit 131 is connected to the coronary artery model acquisition module 12, and is used for acquiring the actual maximum blood flow Q _S of the target blood supply area according to the first coronary artery model, wherein the actual blood flow acquisition unit 131 Methods for obtaining Q _S include, but are not limited to, fluid dynamics simulation, deep learning, and the like.

The ideal blood flow obtaining unit 132 is connected to the coronary artery model obtaining module 12, and is configured to obtain the ideal maximum blood flow Q _N of the target blood supply region according to the second coronary artery model, wherein the ideal blood flow The method by which the obtaining unit 132 obtains Q _N includes, but is not limited to, fluid dynamics simulation, deep learning, and the like.

The reserve fraction obtaining unit 133 is connected to the actual blood flow obtaining unit 131 and the ideal blood flow obtaining unit 132, and is used to obtain the coronary artery according to the actual maximum blood flow and the ideal maximum blood flow of the target blood supply area. Fractional flow reserve, wherein the fractional coronary flow reserve

According to the above description, this embodiment can directly obtain the actual maximum blood flow Q _S and the ideal maximum blood flow Q _N of the target blood supply region, and obtain the coronary blood flow reserve fraction according to Q _S and Q _N . During this process, no

to replace

Thus no additional error is introduced.

The inventor discovered through research and practice,

The value is equal to the ratio of the actual mean pressure in the coronary artery at the distal end of the stenosis under the state of maximum myocardial hyperemia to the mean pressure at this position when there is no stenotic lesion under ideal conditions. The ratio of the average pressure of the

to replace

For this problem, please refer to FIG. 5B . In an embodiment of the present invention, the score obtaining module includes 13 an actual pressure obtaining unit 134 , an ideal pressure obtaining unit 135 and a reserve score obtaining unit 136 .

The actual pressure obtaining unit 134 is connected to the coronary artery model obtaining module 12, and is used for obtaining the actual average pressure of the target position in the maximum myocardial hyperemia state according to the first coronary artery model

Wherein, the target position refers to the distal end of coronary artery stenosis, and the actual pressure acquisition unit 134 acquires

The methods include but are not limited to fluid dynamics simulation, deep learning, etc.

The ideal pressure obtaining unit 135 is connected to the coronary artery model obtaining module 12, and is used to obtain the ideal average pressure of the target position in the maximum myocardial hyperemia state according to the second coronary artery model

The ideal pressure acquisition unit 135 acquires

The reserve score obtaining unit 136 is connected with the actual pressure obtaining unit 134 and the ideal pressure obtaining unit 135, and is used for obtaining the actual average pressure according to the target position

and ideal mean pressure

Obtain the coronary blood flow reserve fraction, wherein the coronary blood flow reserve fraction

According to the above description, this embodiment can directly obtain the actual average pressure of the target position in the maximum myocardial hyperemia state

and mean pressure

and according to

and

to obtain the fractional coronary flow reserve. During this process, no

to replace

Thus no additional error is introduced.

In an embodiment of the present invention, the system for obtaining fractional coronary blood flow reserve further includes a user interaction module.

Optionally, the user interaction module is connected to the medical image acquisition module and/or the coronary artery model acquisition module, and is configured to display the medical image and/or the first coronary artery model. By observing the medical image and/or the first coronary artery model, the user inputs corresponding parameter labeling instructions using the tool provided by the user interaction module. The coronary artery model acquiring module acquires the lesion parameters related to coronary artery stenosis in the medical image and/or the first coronary artery model according to the parameter labeling instruction input by the user. For example, the user can input parameter annotation instructions by using tools such as brushes and erasers provided by the user interaction module through an input device such as a mouse (for example, by clicking a corresponding tool icon with a mouse, and dragging, clicking, or selecting a box, etc.) inputting the parameter labeling instructions) in the medical image and/or the first coronary artery model for stenosis position, stenosis proximal end, stenosis distal end, vessel centerline, cross section of stenosis proximal end and/or stenosis The parameters such as the cross section of the distal end are marked, and the coronary artery model obtaining module can obtain the corresponding lesion parameters according to the marking result of the user.

In addition, when the coronary artery model acquisition module adopts an algorithm to automatically acquire the lesion parameters, the user interaction module is further configured to mark the lesion parameters on the medical image and/or the first coronary artery model, and the user can observe the lesion parameters by observing For the lesion parameters marked in the medical image and/or the first coronary artery model marking, a corresponding parameter modification instruction is input using the tool provided by the user interaction module. The coronary artery model obtaining module modifies the lesion parameters automatically obtained by the algorithm according to the parameter modification instruction input by the user. For example, when the user observes that the blood vessel centerline automatically obtained by the algorithm is not accurate enough, the user can use the input device such as a mouse to input a parameter modification instruction by using tools such as a brush and an eraser provided by the user interaction module (for example, by using a mouse to click on the corresponding tool icon, and input the parameter modification instruction by dragging, clicking or box selection) to adjust the blood vessel centerline in the medical image and/or the first coronary artery model. Preferably, the user interaction module further provides a control point for the user, and the user can edit the spline curve, surface, solid, etc. by selecting and dragging the control point with the mouse.

Optionally, the user interaction module is connected to the medical image acquisition module for displaying the medical image. By observing the medical image, the user inputs corresponding model generation instructions using the tools provided by the user interaction module. The coronary artery model generation module segments the medical image according to the model generation instruction input by the user, so as to obtain the first coronary artery model. For example, the user can input model generation instructions by using tools such as a brush, an eraser, etc. provided by the user interaction module through an input device such as a mouse (for example, by using a mouse to click on the corresponding tool icon, and dragging, clicking, or box-selecting, etc. to segment the medical image, thereby segmenting coronary vessels from the medical image to obtain the first coronary artery model.

In addition, when the coronary artery model obtaining module adopts an algorithm to automatically obtain the first coronary artery model, the user interaction interface is further configured to display the first coronary artery model. By observing the first coronary artery model, the user inputs corresponding model editing instructions using the tool provided by the user interaction module. The coronary artery model generation module edits the first coronary artery model according to a model editing instruction input by a user. For example, when the user observes that the boundary of the first coronary artery model is inaccurate, an input device such as a mouse can be used to input model editing instructions by using tools such as a brush, an eraser and the like provided by the user interaction module (for example, by using a mouse Click the corresponding tool icon, and input the model editing instruction by dragging, clicking, or box selection) to adjust the boundary of the first coronary artery model.

Optionally, the user interaction module is connected to the medical image acquisition module for displaying the medical image. By observing the medical image, the user inputs corresponding model generation instructions using the tools provided by the user interaction module. The coronary artery model generation module repairs at least one lesion of the patient's coronary blood vessel in the medical image according to the model generation instruction input by the user, and segments the repaired medical image to obtain the second coronary artery Model. For example, when a user observes a stenotic lesion in the medical image, the model generation instruction can be input by using an input device such as a mouse, and tools such as a brush, an eraser, etc. provided by the user interaction module (for example, by using a mouse to click the corresponding tool icon, and input the model generation instruction by dragging, clicking or box selection) to modify the blood vessel centerline, blood vessel cross-section and/or blood vessel wall of at least one stenosis position in the medical image , so as to restore the blood vessel centerline, blood vessel cross-section and/or blood vessel wall of the at least one stenosis position to a normal state, so as to repair the lesion at the at least one stenosis position; after the repair is completed, the user can An input device such as a mouse, using tools such as brushes and erasers provided by the user interaction module to continue to input the model generation instructions to segment the medical image, thereby segmenting coronary vessels from the medical image, The second coronary artery model is obtained.

Optionally, the user interaction module is connected to the coronary artery model obtaining module, and is configured to display the first coronary artery model. The first coronary artery module may be automatically generated by an algorithm, or manually generated by a user through a model generation instruction. By observing the first coronary artery model, the user inputs corresponding model generation instructions using the tool provided by the user interaction module. The coronary artery model generation module repairs at least one lesion of the coronary artery of the patient in the first coronary artery model according to the model generation instruction input by the user, so as to obtain the second coronary artery model. For example, when the user observes a stenosis lesion in the first coronary artery model, the model generation instruction can be input by using an input device such as a mouse, using tools such as a brush and an eraser provided by the user interaction module (for example, by using a mouse Click the corresponding tool icon, and input the model generation instruction by dragging, clicking or box-selecting), to analyze the blood vessel centerline, blood vessel cross-section and /or the vessel wall is modified, so that the centerline of the vessel, the vessel cross-section and/or the vessel wall of the at least one stenosis position is restored to a normal state, so as to achieve the repair of the lesion at the at least one stenosis position, and after the repair The second coronary artery model can be obtained.

Optionally, the user interaction module is connected to the coronary artery model obtaining module, and is configured to display the second coronary artery model. Wherein, the second coronary artery model is automatically acquired by the coronary artery model acquiring module using an algorithm. By observing the second coronary artery model, the user inputs corresponding model editing instructions using the tool provided by the user interaction module. The coronary artery model generation module edits the second coronary artery model according to the model editing instruction input by the user. For example, when the user observes that the boundary of the second coronary artery model is inaccurate, an input device such as a mouse can be used to input model editing instructions by using tools such as a brush and an eraser provided by the user interaction module (for example, by using a mouse Click the corresponding tool icon, and input the model editing instruction by dragging, clicking or box selection) to adjust the boundary of the second coronary artery model.

In an embodiment of the present invention, the user interaction module is further configured to receive an automatic repair instruction input by the user, so as to trigger the coronary artery model acquisition module to perform the first coronary artery model and/or the medical image. Automatic repair. For example, when the user inputs the automatic repair instruction by clicking a certain stenosis position C with the mouse, the simulated repair module starts to repair the stenotic lesion at the stenosis position C according to the automatic repair instruction.

Based on the above description of the system for obtaining fractional coronary blood flow reserve, the present invention also provides a method for obtaining fractional coronary blood flow reserve. The score acquisition system is implemented. Referring to FIG. 6, in an embodiment of the present invention, the method for obtaining fractional coronary blood flow reserve includes:

S61. Acquire a medical image of a patient, where the medical image includes a heart region of the patient.

S62, obtaining a first coronary artery model and a second coronary artery model of the patient according to the medical image; wherein, the first coronary artery model refers to the actual coronary artery model of the patient, and the second coronary artery model refers to the Coronary artery model obtained after at least one lesion of the coronary artery of the patient is repaired.

S63: Obtain the coronary blood flow reserve fraction of the patient according to the first coronary artery model and the second coronary artery model.

Based on the above description of the method for obtaining the fractional coronary blood flow reserve, the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, realizes the coronary artery shown in FIG. 6 . Methods of obtaining fractional blood flow reserve.

The protection scope of the coronary blood flow fraction calculation method of the present invention is not limited to the execution sequence of the steps listed in this embodiment, and any solution implemented by adding or subtracting steps and replacing steps in the prior art according to the principles of the present invention All are included in the protection scope of the present invention.

The present invention also provides a system for obtaining fractional coronary blood flow reserve, which can realize the method for obtaining fractional coronary blood flow reserve of the present invention, but the The devices for realizing the fractional flow reserve acquisition method include, but are not limited to, the structures of the coronary blood flow fractional fractional acquisition system listed in this embodiment. within the scope of protection of the invention.

to replace

To sum up, the present invention effectively overcomes various shortcomings in the prior art and has high industrial utilization value.

The above-mentioned embodiments merely illustrate the principles and effects of the present invention, but are not intended to limit the present invention. Anyone skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed in the present invention should still be covered by the claims of the present invention.

Claims

A system for obtaining fractional coronary blood flow reserve, characterized in that the system for obtaining fractional coronary blood flow reserve comprises:

a medical image acquisition module for acquiring a medical image of a patient, the medical image including the patient's heart region;

The coronary artery model acquisition module is connected to the medical image acquisition module, and is used for acquiring the first coronary artery model and the second coronary artery model of the patient according to the medical image; wherein, the first coronary artery model refers to the patient's The actual coronary artery model, the second coronary artery model refers to the coronary artery model obtained after repairing at least one lesion of the coronary artery of the patient;

The blood flow reserve fraction acquisition module is connected to the coronary artery model acquisition module, and is used for acquiring the coronary blood flow reserve fraction of the patient according to the first coronary artery model and the second coronary artery model.
The system for obtaining fractional coronary blood flow reserve according to claim 1, wherein the coronary model obtaining module comprises a first coronary model obtaining sub-module and a second coronary model obtaining sub-module;

The first coronary artery model acquisition sub-module is connected to the medical image acquisition module, and is used for segmenting the medical image to acquire the first coronary artery model;

The second coronary artery model acquisition sub-module is connected with the first coronary artery model acquisition sub-module, and is used to simulate and repair at least one lesion of the coronary artery of the patient in the first coronary artery model, so as to obtain all the lesions. the second coronary model described above; or,

The second coronary artery model acquisition sub-module is connected to the medical image acquisition module, and is used for simulating repair of at least one lesion of the coronary artery of the patient in the medical image, and segmenting the medical image after the simulating repair to obtain the second coronary artery model.
The system for obtaining fractional coronary blood flow reserve according to claim 2, wherein the second coronary model obtaining sub-module comprises:

a first lesion parameter acquisition unit, connected to the first coronary artery model acquisition sub-module, for acquiring the lesion parameters related to coronary stenosis in the first coronary artery model;

The first simulated repair unit is connected to the first lesion parameter acquisition unit, and is configured to modify the stenosis of the first coronary artery model according to the lesion parameter to acquire the second coronary artery model.
The system for obtaining fractional coronary blood flow reserve according to claim 2, wherein the second coronary model obtaining sub-module comprises:

a second lesion parameter acquisition unit, connected to the image acquisition module, for acquiring lesion parameters related to coronary stenosis in the medical image;

The second simulated repair unit is connected to the second lesion parameter acquisition unit, and is configured to modify the stenosis of the patient's coronary blood vessel in the medical image according to the lesion parameter, so as to obtain the medical image after the simulated repair ;

The image segmentation unit is connected to the second simulated repair unit, and is used for segmenting the medical image after the simulated repair to obtain the second coronary artery model.
The system for obtaining fractional coronary blood flow reserve according to claim 1, wherein the coronary model obtaining module comprises a third coronary model obtaining sub-module and a fourth coronary model obtaining sub-module;

The third coronary artery model acquisition sub-module is connected to the medical image acquisition module, and is used for segmenting the medical image to acquire the first coronary artery model;

The fourth coronary artery model acquisition sub-module is connected with the third coronary artery model acquisition sub-module, and is used to perform virtual treatment on at least one lesion of the coronary artery of the patient in the first coronary artery model, so as to obtain all the lesions. the second coronary model described above; or

The fourth coronary artery model acquisition sub-module is connected to the medical image acquisition module, and is used to perform virtual treatment on at least one lesion of the coronary blood vessel of the patient in the medical image, and segment the medical image after the virtual treatment to obtain the second coronary artery model.
The system for obtaining fractional coronary blood flow reserve according to claim 1, wherein the fractional blood flow reserve obtaining module comprises:

an actual blood flow acquisition unit, connected to the coronary artery model acquisition module, and configured to acquire the actual maximum blood flow of a target blood supply area according to the first coronary artery model;

an ideal blood flow acquisition unit, connected to the coronary artery model acquisition module, and configured to acquire the ideal maximum blood flow of the target blood supply region according to the second coronary artery model;

a fractional reserve acquisition unit, connected to the actual blood flow acquisition unit and the ideal blood flow acquisition unit, and configured to acquire the coronary blood flow reserve fraction according to the actual maximum blood flow and the ideal maximum blood flow of the target blood supply area .
The system for obtaining fractional coronary blood flow reserve according to claim 1, wherein the fractional blood flow reserve obtaining module comprises:

The actual pressure acquisition unit is connected to the coronary artery model acquisition module, and is used for acquiring the actual average pressure of the target position in the maximum myocardial hyperemia state according to the first coronary artery model, wherein the target position refers to the coronary artery. narrow distal end;

an ideal pressure acquisition unit, connected to the coronary artery model acquisition module, and configured to acquire the ideal average pressure of the target position under the maximum myocardial hyperemia state according to the second coronary artery model;

A fractional reserve acquisition unit, connected to the actual pressure acquisition unit and the ideal pressure acquisition unit, and configured to acquire the coronary blood flow reserve fraction according to the actual average pressure and the ideal average pressure at the target position.
The system for obtaining fractional coronary blood flow reserve according to claim 1, wherein the system for obtaining fractional coronary blood flow reserve further comprises:

A user interaction module, connected with the medical image acquisition module and/or the coronary artery model acquisition module, for displaying the medical image, the first coronary artery model and/or the second coronary artery model, and using to obtain the model generation instruction and/or model editing instruction input by the user; the coronary artery model acquisition module acquires the second coronary artery model according to the model generation instruction, and/or the coronary artery model acquisition module acquires the coronary artery model according to the The model editing instruction edits the first coronary artery model and/or the second coronary artery model.
A method for obtaining fractional coronary blood flow reserve, wherein the method for obtaining fractional coronary blood flow reserve comprises:

obtaining a medical image of the patient, the medical image including the patient's heart region;

Obtain the first coronary artery model and the second coronary artery model of the patient according to the medical image; wherein, the first coronary artery model refers to the actual coronary artery model of the patient, and the second coronary artery model refers to the coronary artery model of the patient. Coronary artery model obtained after at least one lesion of the blood vessel is repaired;

The coronary blood flow reserve fraction of the patient is obtained according to the first coronary artery model and the second coronary artery model.
A computer-readable storage medium on which a computer program is stored, characterized in that: when the computer program is executed by a processor, the method for obtaining fractional coronary blood flow reserve according to claim 9 is implemented.