CN117976234A - System and method for determining fractional flow reserve - Google Patents

Abstract

Embodiments of the present disclosure provide a system and method for determining fractional flow reserve. The method comprises the following steps: obtaining a reference fractional flow reserve of a target position of a target object under a reference physiological parameter, the target position being a position on a target vessel of the target object; and processing the target physiological parameter of the target object, the reference physiological parameter and the reference fractional flow reserve by using a fractional flow reserve algorithm, and determining the target fractional flow reserve of the target position under the target physiological parameter.

Description

System and method for determining fractional flow reserve

Technical Field

The present disclosure relates to the field of medical technology, and in particular, to a system and method for determining fractional flow reserve.

Background

Fractional flow reserve (Fractional Flow Reserve, FFR) can be effective in assessing the extent of ischemia due to plaque stenosis, providing a reference for the direction of subsequent treatment. Fractional flow reserve (FFR obtained based on CT may be referred to as FFRCT) obtained based on medical imaging techniques (e.g., coronary CT scanning) is more patient friendly than invasive FFR due to the non-invasive examination means it uses.

It is now common practice to derive FFR by simulating by computational fluid dynamics (Computational Fluid Dynamics, CFD) to obtain blood flow characteristics (pressure, velocity, etc.). The CFD simulation has the advantages of high accuracy, high authenticity and the like. However, at the same time, the principle is to solve the Navier-Stokes equation, and the process involves very complex calculation, so that the CFD simulation calculation cost is large and the time is long. Although there have been recent studies to solve the problem of large CFD simulation overhead, it has been proposed to use a machine learning model to obtain FFR. The input of the machine learning model requires the inclusion of a large amount of data (e.g., physiological parameters, medical images, detailed anatomical features, etc.) related to the patient and thus still requires significant computational resources and is inefficient.

Physiological parameters of the patient (blood pressure, heart rate, cardiac output, etc.) are directly related to the calculation of FFR. In practice, FFR is typically determined using the same set of default physiological parameters for different subjects. When the FFR of the subject is subsequently required to be determined again, the actual FFR will change due to the change in the physiological parameter of the subject, and thus if the FFR determined before is directly used, the diagnosis of the patient will be adversely affected. If FFR is calculated again based on new physiological parameters, this will result in inefficiency and a significant amount of time and computing resources.

It is therefore desirable to provide a system and method for efficiently and accurately determining fractional flow reserve.

Disclosure of Invention

One of the embodiments of the present specification provides a method of determining fractional flow reserve. The method comprises the following steps: obtaining a reference fractional flow reserve of a target position of a target object under a reference physiological parameter, the target position being a position on a target vessel of the target object; and processing the target physiological parameter of the target object, the reference physiological parameter and the reference fractional flow reserve by using a fractional flow reserve algorithm, and determining the target fractional flow reserve of the target position under the target physiological parameter.

One of the embodiments of the present description provides a system for determining fractional flow reserve. The system includes an acquisition module configured to acquire a reference fractional flow reserve of a target location of a target subject under a reference physiological parameter, the target location being a location on a target vessel of the target subject; and determining a model configured to process a target physiological parameter of the target subject, the reference physiological parameter, and the reference fractional flow reserve using a fractional flow reserve algorithm to determine a target fractional flow reserve for the target location at the target physiological parameter.

One of the embodiments of the present description provides a system for determining fractional flow reserve. The system includes at least one memory device for storing computer instructions; at least one processor configured to execute the computer instructions to implement the method of determining fractional flow reserve described above.

Additional features of the application will be set forth in part in the description which follows. Additional features of the application will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following description and the accompanying drawings or may be learned from production or operation of the embodiments. The features of the present application can be implemented and obtained by practicing or using the various aspects of the methods, means, and combinations set forth in the detailed examples below.

Drawings

The present specification will be further elucidated by way of example embodiments, which will be described in detail by means of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:

FIG. 1 is a schematic illustration of an application scenario of an exemplary FFR determination system shown in accordance with some embodiments of the present description;

FIG. 2 is a schematic diagram of an exemplary FFR determination system shown in accordance with some embodiments of the present description;

FIG. 3 is a schematic diagram of an exemplary FFR determination process shown in accordance with some embodiments of the present disclosure;

Figure 4 is a flow diagram of an exemplary determination FFR algorithm shown in accordance with some embodiments of the present description.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.

It will be appreciated that "system," "apparatus," "unit" and/or "module" as used herein is one method for distinguishing between different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

A flowchart is used in this specification to describe the operations performed by the system according to embodiments of the present specification. It should be appreciated that the preceding or following operations are not necessarily performed in order precisely. Rather, the steps may be processed in reverse order or simultaneously. Also, other operations may be added to or removed from these processes.

Fig. 1 is a schematic illustration of an application scenario of an exemplary FFR determination system according to some embodiments of the present description. As shown in fig. 1, application scenario 100 of FFR determination system may include a medical device 110, a processing device 120, a terminal device 130, a storage device 140, and a network 150. In some embodiments, the processing device 120 may be part of the medical device 110. The connections between components in the application scenario 100 may be variable. As shown in fig. 1, medical device 110 may be connected to processing device 120 through a network 150. As another example, medical device 110 may be directly connected to processing device 120. For another example, the storage device 140 may be connected to the processing device 120 directly or through the network 150. As yet another example, terminal 130 may be directly connected to processing device 120 (as indicated by the dashed arrow connecting terminal 130 and processing device 120), or may be connected to processing device 120 via network 150.

Medical device 110 may be a non-invasive scanning imaging device for disease diagnosis or research purposes. In some embodiments, the medical device 110 may scan an object within a detection region or scanning region to obtain scan data for the object. In some embodiments, medical device 110 may include a single modality scanner and/or a multi-modality scanner. The single mode scanner may include, for example, an ultrasound scanner, an X-ray scanner, a Computed Tomography (CT) scanner, a Magnetic Resonance Imaging (MRI) scanner, an Optical Coherence Tomography (OCT) scanner, an Ultrasound (US) scanner, an intravascular ultrasound (IVUS) scanner, or the like, or any combination thereof. The multi-modality scanner may include, for example, an X-ray imaging-magnetic resonance imaging (X-ray-MRI) scanner, a positron emission tomography-X-ray imaging (PET-X-ray) scanner, a single photon emission computed tomography-magnetic resonance imaging (SPECT-MRI) scanner, a positron emission tomography-computed tomography (PET-CT) scanner, a digital subtraction angiography-magnetic resonance imaging (DSA-MRI) scanner, and the like. In some embodiments, the medical device 110 is a CT scanner. In some embodiments, the processing device 120 may be integrated on the medical device 110, or the medical device 110 and the processing device 120 may perform their functions by the same entity. The medical devices provided above are for illustrative purposes only and are not intended to limit the scope of the present description.

The processing device 120 may process data and/or information obtained from the medical device 110, the terminal device 130, the storage device 140, or other components of the application scenario 100. For example, the processing device 120 obtains a reference fractional flow reserve for the target location of the target subject at the reference physiological parameter. The target location is a location on a target vessel of the target object. The processing device 120 then processes the target physiological parameter, the reference physiological parameter, and the reference fractional flow reserve of the target subject using a fractional flow reserve algorithm to determine a target fractional flow reserve for the target location at the target physiological parameter.

In some embodiments, the processing device 120 may be local or remote. For example, the processing device 120 may access information and/or data from the medical device 110, the terminal device 130, and/or the storage device 140 via the network 150.

Terminal device 130 may include a mobile device 131, a tablet 132, a notebook 133, and the like, or any combination thereof. In some embodiments, terminal 130 may be part of processing device 120.

Storage device 140 may store data, instructions, and/or any other information. In some embodiments, the storage device 140 may store data obtained from the medical device 110, the processing device 120, and/or the terminal device 130, e.g., medical images generated by the medical device 110, etc.

Network 150 may include any suitable network capable of facilitating the exchange of information and/or data. In some embodiments, at least one component of the application scenario 100 (e.g., the medical device 110, the processing device 120, the terminal device 130, the storage device 140) may exchange information and/or data with at least one other component in the application scenario 100 via the network 150. For example, the processing device 120 may obtain medical images of the target object from the medical device 110 via the network 150. As another example, terminal device 130 may obtain FFR of the target object from processing device 120 via network 150.

It should be noted that the application scenario 100 is provided for illustrative purposes only and is not intended to limit the scope of the present description. Many modifications and variations will be apparent to those of ordinary skill in the art in light of the present description. For example, the application scenario 100 may also include an input device and/or an output device. As another example, application scenario 100 may implement similar or different functionality on other devices. However, such changes and modifications do not depart from the scope of the present specification.

Fig. 2 is a schematic diagram of an exemplary FFR determination 200 shown in accordance with some embodiments of the present description.

As shown in fig. 2, in some embodiments, FFR determination system 200 may include an acquisition module 201, a determination module 202. In some embodiments, FFR determination system 200 may further comprise training module 203. In some embodiments, the corresponding functions of FFR determination system 200 may be performed by processing device 120, e.g., acquisition module 201, determination module 202, and training module 203 may be modules in processing device 120.

The acquisition module 201 may be configured to acquire a reference fractional flow reserve of a target location of a target subject at a reference physiological parameter, the target location being a location on a target vessel of the target subject. For further description of obtaining a reference fractional flow reserve for a target location of a target subject under a reference physiological parameter, see elsewhere in this document (e.g., 310 in fig. 3), which is not repeated herein.

The determination module 202 may be configured to process the target physiological parameter of the target subject, the reference physiological parameter, and the reference fractional flow reserve using a fractional flow reserve algorithm to determine a target fractional flow reserve for the target location at the target physiological parameter. For more description of determining the fractional flow reserve of the target location at the target physiological parameter, see elsewhere in this specification (e.g., 320 in fig. 3), and will not be described in detail herein.

The training module 203 may be configured to perform model training based on the sample data to obtain a trained machine learning model (e.g., the machine learning model used to determine fractional flow reserve in step 320). For further description of training of machine learning models for determining fractional flow reserve, see elsewhere in this document (e.g., 430 in fig. 4), which is not repeated herein.

For further description of determining the fractional flow reserve of the target location at the target physiological parameter, see elsewhere in this document (e.g., 320 in fig. 3), and will not be described in detail herein.

It should be understood that the system shown in fig. 2 and its modules may be implemented in a variety of ways. For example, in some embodiments the system and its modules may be implemented in hardware, software, or a combination of software and hardware.

It should be noted that the above description of the system and its modules is for descriptive convenience only and is not intended to limit the present disclosure to the scope of the illustrated embodiments. It will be appreciated by those skilled in the art that, given the principles of the system, various modules may be combined arbitrarily or a subsystem may be constructed in connection with other modules without departing from such principles. For example, in some embodiments, the above modules disclosed in fig. 2 may be different modules in one system, or may be one module to implement the functions of two or more modules described above. For example, each module may share one memory module, or each module may have a respective memory module. Such variations are within the scope of the present description.

Fig. 3 is a flow diagram illustrating an exemplary determination of a target FFR according to some embodiments of the present description. In some embodiments, one or more steps of flow 300 may be implemented at FFR determination system 100 shown in fig. 1 or performed by FFR determination system 200 shown in fig. 2. For example, the process 300 may be performed by a module of the processing device 120. As shown in fig. 3, the process 300 may include the following steps.

Step 310, obtaining a reference FFR of a target position of a target object under a reference physiological parameter, wherein the target position is a position on a target blood vessel of the target object. In some embodiments, step 310 may be performed by processing device 120 or acquisition module 201.

The target object may comprise a human body, an animal or a part thereof. Hereafter, description will be made taking a human body as an example.

The target vessel may include various vessels such as coronary, aortic, pulmonary, etc. The application is not limited to the type of target vessel.

The target location may correspond to one or more points on a target vessel of the target object. In some embodiments, the target location corresponds to a point in the target vessel near a lesion (e.g., thrombus). For example, the target location corresponds to a point in the target vessel a distance (e.g., 2 cm) upstream and downstream of the lesion. In some embodiments, the target location is specified by a user (e.g., a physician). In some embodiments, the processing device 120 may select a plurality of target locations equally spaced along the direction of extension of the target vessel. Optionally, the target locations near the user-specified location are more dense upstream and downstream of the lesion than the target locations in the other areas. For each target location, steps 310-320 may be performed to determine its corresponding target FFR.

In some embodiments, the physiological parameters (e.g., reference physiological parameters, sample physiological parameters, target physiological parameters) of the present application include at least one of blood pressure, cardiac output, heart rate, etc. Wherein the blood pressure may be the mean arterial pressure, i.e. the average of the sum of the systolic and diastolic pressures.

The reference fractional flow reserve refers to the fractional flow reserve that the target subject has determined in the historical examination, and the reference physiological parameter refers to the physiological parameter of the target subject in the historical examination. In some embodiments, the reference physiological parameter may be a default physiological parameter. As described elsewhere in this specification, in practice, FFR is typically determined using a set of the same default physiological parameters for different subjects. The default physiological parameter should correspond to the physiological condition of most people and is thus typically selected within normal physiological parameters. For example, a large number of normal human physiological parameters may be counted and the mean or mode of these physiological parameters selected as the default physiological parameter. Default physiological parameters may also be determined empirically by the user. The application is not limited in this regard. In some embodiments, the reference physiological parameter and the reference FFR of the target subject are stored in a storage device (e.g., storage device 140 or an external storage device) from which the processing device 120 can retrieve the reference physiological parameter and the reference FFR.

In some embodiments, a reference FFR of the target subject at the reference physiological parameter may be determined based on the medical image of the target subject and the reference physiological parameter using one of a computational fluid dynamics (Computational Fluid Dynamics, CFD) equation, a centralized parameter model, and a machine learning model. The medical image may be an image obtained by scanning a region of the human body containing a target blood vessel, for example, a CT angiographic image, a CT pan-scan image, or the like. In some embodiments, a scan of a target vessel of a target object may be performed by medical device 110 to acquire a medical image. In some embodiments, the medical image may be generated in advance and stored in a storage device (e.g., storage device 140 or an external storage device) from which the processing device 120 may obtain the medical image.

Specifically, the processing device 120 determines a vessel segmentation image of the target object based on the medical image. The vessel segmentation image refers to an image generated after segmenting a target vessel in the medical image, which may be indicative of the target vessel of the subject. In some embodiments, different vessel branches may be displayed in different ways in the vessel segmentation image. A vessel branch may include a segment of a vessel between two adjacent bifurcation points in the vessel, a segment of a vessel between a vessel start point and its adjacent bifurcation point, and a segment of a vessel between each vessel end and its adjacent bifurcation point. For example, different blood vessel branches may be displayed in the blood vessel segmentation image using different colors. For another example, different branch labels (e.g., labels "1", "2", …) may be used to display different vessel branches in the vessel segmentation image. In some embodiments, the vessel segmentation image may be manually derived by a user (e.g., a imaging physician) from a medical image to delineate the target vessel. In some embodiments, the vessel segmentation image may be automatically derived by the processing device 120 from segmenting the target vessel from the medical image. For example, the processing device 120 may segment the target vessel from the medical image using an image segmentation algorithm or a machine learning model to obtain a vessel segmentation image.

Then, the processing device 120 performs three-dimensional reconstruction based on the medical image and the blood vessel segmentation image, obtains a three-dimensional model of the target blood vessel, and determines a reference FFR based on the three-dimensional model. Specifically, in some embodiments, the processing device 120 may determine a reference boundary condition of the target object based on the reference physiological parameter. Boundary conditions in the present application refer to boundary conditions of blood vessels, which are directly related to physiological parameters of the patient (blood pressure, heart rate, cardiac output, etc.). Further, processing device 120 may determine the reference FFR based on a reference boundary condition of the target object using one of a computational fluid dynamics CFD equation, a centralized parameter model, a machine learning model, and the like. For example, first, the processing device 120 determines an aortic inlet boundary condition (e.g., heart cycle of 1 second, cardiac output of 83 ml/s) of the target object, an aortic outlet boundary condition resistance, and determines a boundary condition resistance of the target vessel according to a predetermined boundary condition resistance equation. For example, the processing device 120 may determine the aortic outlet boundary condition resistance and the boundary condition resistance of the target vessel according to the following formulas (1) and (2), respectively:

Wherein, R ₁ is the aortic outlet boundary condition resistance, R ₂ is the boundary condition resistance of one lumen of the target vessel (one lumen of the vessel refers to the vessel segment through which blood passes from the beginning of the inflow vessel to one end of the outflow vessel), P _mean is the blood pressure, Q is the cardiac output, i is the lumen number of the vessel, n is the total number of lumens, l _i is the branch vessel weight corresponding to the i-th lumen end, l _e is the branch vessel weight corresponding to the current lumen end, and β and γ are hyper-parameters.

The processing device 120 then determines an equivalent voltage for each point on the target vessel using the CFD equation based on the aortic inlet boundary condition, the aortic outlet boundary condition, and the boundary condition resistance of the target vessel of the target subject. Finally, the processing device 120 determines a reference FFR for each point on the target vessel by normalizing the equivalent voltage for each point on the target vessel based on the target vessel inlet equivalent voltage, and obtains a reference FFR for each target location.

For another example, the processing device 120 determines an equivalent resistance of the target vessel based on the medical image using a first machine learning model and determines a boundary condition resistance of the target vessel using a second machine learning model. Wherein the equivalent resistance of the target vessel comprises resistance values of a plurality of points on the target vessel, and the boundary condition resistance is used for describing boundary conditions of blood flowing out of the tail end of the target vessel. Further, the processing device 120 determines a reference FFR for each point on the target vessel based on the equivalent resistance and the boundary condition resistance of the target vessel, and acquires a reference FFR corresponding to each target location.

Step 320, the target physiological parameter of the target object, the reference physiological parameter and the reference FFR are processed by using the FFR algorithm, and a target FFR of the target position under the target physiological parameter is determined. In some embodiments, step 320 may be performed by processing device 120 or determination module 202.

The target physiological parameter may be a current physiological parameter of the target subject. In some embodiments, the target physiological parameter and the reference physiological parameter are of the same type. For example, the reference physiological parameters include blood pressure and cardiac output, and the target physiological parameters also include blood pressure and cardiac output.

The FFR algorithm may be an algorithm for determining a changed FFR based on the FFR value corresponding to the original physiological parameter and the change in the physiological parameter. In some embodiments, the FFR algorithm includes a machine learning model or a multidimensional function. The multidimensional function is used to characterize the effect of changes in physiological parameters on FFR. For example, where the FFR value for a subject is a for a first set of physiological parameters and B for a second set of physiological parameters, it is understood that as the physiological parameters of this subject change from a to B for the first set of physiological parameters, the FFR value changes from a to B, i.e., a multidimensional function may be used to characterize the effect of the change from the first set of physiological parameters to the second physiological parameters on the FFR. The machine learning model (which may also be referred to as a target machine learning model) may include a support vector machine model, a random forest model, a deep learning model, and the like. Specifically, if the FFR algorithm is a multi-dimensional function, processing device 120 may substitute the target physiological parameter, the reference physiological parameter, and the reference FFR into the multi-dimensional function for calculation to determine the target FFR. If the FFR algorithm is a machine learning model, processing device 120 can input the target physiological parameter, the reference physiological parameter, and the reference FFR as model inputs and the model inputs into the machine learning model, which can output the target FFR.

In some embodiments, the model input may further include a medical image of the target object, a segmented image of the target vessel, anatomical features of the target location, depth features indicative of a relationship between anatomical features of the target location, equivalent capacitance or equivalent resistance corresponding to the target location, and the like. The anatomical features include the cross-sectional diameter and area of the target site in the direction perpendicular to the blood flow, the length of the lumen in which the target site is located, the length of the lumen upstream of the target site, the location and number of bifurcation points upstream and downstream of the target site, the distance of the lesion from the target site along the target vessel, the stenosis rate of the target site, etc.

In some embodiments, the processing device 120 may determine the anatomical feature of the target location based on a medical image of the target object and/or a segmented image of the target vessel. In some embodiments, the processing device 120 may determine the equivalent capacitance or equivalent resistance corresponding to the target location using a focused parameter model (e.g., WINDKESSEL MODEL). In some embodiments, the processing device 120 may input anatomical features of the target location into a reference model, designating the output of intermediate layers of the reference model as depth features indicative of the relationship between the anatomical features of the target location. The reference model may be a machine learning model for determining FFR based on physiological parameters and anatomical features of the subject. In some embodiments, the processing device 120 may perform pre-processing such as normalization, de-outlier, etc. on the obtained features to improve the accuracy of the features. In this way, a more accurate target FFR can be obtained by adding various other features related to FFR determination as model inputs.

In some embodiments, processing device 120 may retrieve the FFR algorithm from a storage device (e.g., storage device 140). In some embodiments, the FFR algorithm may be generated based on multiple sets of sample physiological parameters. The sample physiological parameter and the reference physiological parameter (or target physiological parameter) are of the same type. For example, the reference physiological parameters include blood pressure and cardiac output, and the sample physiological parameters also include blood pressure and cardiac output. In particular, multiple sets of sample physiological parameters may be acquired and for each sample object of the one or more sample objects, a sample FFR for that sample object under each set of sample physiological parameters is determined. Further, an FFR algorithm may be determined based on the plurality of sets of sample physiological parameters and the corresponding sample FFR. For more description of the FFR generation algorithm, please refer to fig. 4 of the present application, and the description thereof is omitted.

It should be noted that the above description of the process 300 is for purposes of example and illustration only and is not intended to limit the scope of applicability of the present disclosure. Various modifications and changes to flow 300 will be apparent to those skilled in the art in light of the present description. However, such modifications and variations are still within the scope of the present description.

Figure 4 is a flow diagram of an exemplary determination FFR algorithm shown in accordance with some embodiments of the present description. In some embodiments, the process 400 may be performed by the processing device 120 or the first one or more modules shown in fig. 2 (e.g., the training module 203). As shown in fig. 4, the flow 400 may include the following steps.

In step 410, a plurality of sets of sample physiological parameters are obtained.

In some embodiments, the plurality of sets of sample physiological parameters are within a range of physiological parameters of a normal person. In some embodiments, the plurality of sets of sample physiological parameters may be random values within a range of normal physiological parameters. In some embodiments, the sets of sample physiological parameters may be determined by sampling at regular, e.g., equidistant, intervals. For example, a normal human blood pressure range is between 70mmHg and 105mmHg, and blood pressure values in a set of sample physiological parameters are determined every 5mmHg from 70mmHg to 105 mmHg. The cardiac output of a normal human ranges from 66-116mL/s, and the cardiac output is from 66mL/s to 116mL/s, and the cardiac output value in a group of sample physiological parameters is determined every 5 mL/s. Thus, 8 blood pressure values and 11 heart output values were obtained. And combining the 8 blood pressure values and the 11 heart output values in pairs to obtain 88 groups of sample physiological parameters.

Step 420, for each sample object of the one or more sample objects, determines a sample FFR for that sample object under each set of sample physiological parameters.

The sample object is the same type as the target object. For example, the sample object and the target object are both human or the same animal. In some embodiments, the sample FFR of a sample object under each set of sample physiological parameters comprises the sample FFR of a plurality of sample locations of a sample vessel of the sample object. In particular, for each sample object, the processing device 120 may acquire a sample medical image of the sample object. The sample medical image contains a sample vessel of the sample object. The sample vessel is of the same target vessel type as the target object in fig. 3. For example, both the sample vessel and the target vessel are coronary. In some embodiments, at least a portion of the sample subject's sample blood vessel contains a lesion (e.g., thrombus). The processing device 120 may then determine a sample vessel segmentation image of the sample object based on the sample medical image and reconstruct a three-dimensional model of the sample vessel of the sample object further based on the sample vessel segmentation image. The processing device 120 may determine a plurality of sample locations of the sample vessel based on the three-dimensional model of the sample vessel. In some embodiments, the sample location comprises a point in the sample vessel near the lesion. For example, the sample location includes a point in the sample vessel a distance (e.g., 2 cm) upstream and downstream of the lesion. In some embodiments, the sample location is specified by a user (e.g., a physician). In some embodiments, the processing device 120 may select a plurality of sample locations at equal intervals along the sample vessel. Optionally, the sample locations near the user-specified location are more dense upstream and downstream of the lesion than the sample locations in other areas. Then, for each sample object, processing device 120 can determine a sample FFR for each sample location of the sample object under each set of sample physiological parameters.

In some embodiments, processing device 120 may determine the sample FFR for each sample object in a manner similar to the determination of the reference FFR for the target object described in step 310. For example, the processing device 120 can determine sample boundary conditions of the sample object based on the sample physiological parameter, and determine a sample FFR of the sample object based on the sample boundary conditions of the sample object using one of a CFD equation, a focused parameter model, a machine learning model, and the like. For example, for 50 sample objects and 88 sets of sample physiological parameters, the processing device 120 needs to perform 50 x 89 calculations by CFD equations, centralized parameter models, machine learning models, etc. using 88 sets of sample physiological parameters to obtain the sample FFR for all sample objects.

Step 430, determining an FFR algorithm based on the plurality of sets of sample physiological parameters and the sample FFR.

In some embodiments, processing device 120 may determine multiple sets of sample data based on multiple sets of sample physiological parameters and sample FFR, and determine an FFR algorithm based on the multiple sets of sample data. Each set of sample data corresponds to a sample position of a sample object in the one or more sample objects and includes a first set of sample physiological parameters, a second set of sample physiological parameters, a first sample FFR with its corresponding sample position at the first set of sample physiological parameters, and a second sample FFR with its corresponding sample position at the second set of sample physiological parameters. For example, if the FFR of the sample at the sample position a of one sample subject is 0.7 at a blood pressure of 90mmHg and a cardiac output of 80mL/s and the FFR of the sample at the sample position a is 0.8 at a blood pressure of 100mmHg and a cardiac output of 85mL/s, FFR values of 0.7 for the first set of sample physiological parameters (blood pressure of 90mmHg and cardiac output of 80 mL/s) and FFR values of 0.8 for the second set of sample physiological parameters (blood pressure of 100mmHg and cardiac output of 85 mL/s) may be used as a set of sample data.

In some embodiments, when the FFR algorithm is a multi-dimensional function, processing device 120 may obtain an initial multi-dimensional function and generate the multi-dimensional function by fitting the initial multi-dimensional function based on the sets of sample data. For example only, the processing device 120 may obtain the following initial multidimensional function (1):

Wherein, P ₀、Q₀ is a first set of sample physiological parameters, P ₁、Q₁ is a second set of sample physiological parameters, FFR ₀ is a first sample FFR at the first set of sample physiological parameters, b ₀-b_n is a fitting parameter, y is a second sample FFR at the second set of sample physiological parameters.

The processing device 120 may substitute the first set of sample physiological parameters, the second set of sample physiological parameters, the first sample FFR, the second sample FFR in each set of sample data into equation (1), respectively, to obtain a plurality of equations. In some embodiments, the processing device 120 may solve the fitting parameter b ₀-b_n using a least squares method, a linear algebraic method (e.g., normal equations), an optimization algorithm (e.g., gradient descent) through the obtained plurality of equations, resulting in a fitted multidimensional function.

In some embodiments, when the FFR algorithm is a machine learning model, processing device 120 may generate a training input and training labels based on each set of sample data and generate a machine learning model by training an initial model based on the obtained training input and training labels. The training input for each set of sample data includes a first set of sample physiological parameters, a second set of sample physiological parameters, a first sample FFR in the set of sample data, and the training tag includes a second sample FFR in the sample data.

In some embodiments, the training input further includes an anatomical feature of a sample location corresponding to the set of sample data, a depth feature indicative of a relationship between the anatomical features of the sample location, an equivalent capacitance or equivalent resistance corresponding to the sample location, a sample medical image containing the sample vessel, a segmented image of the sample vessel, and the like. In some embodiments, similar to the anatomical features of the target location described in step 320, the anatomical features of the sample location may include a cross-sectional diameter and area of the sample location in a direction perpendicular to blood flow, a lumen length in which the sample location is located, an upstream lumen length of the sample location, a location and number of bifurcation points upstream and downstream of the sample location, a distance of a lesion from the sample location along a sample vessel, a stenosis rate of the sample location, and the like. In some embodiments, the processing device 120 may determine the anatomical features of the sample locations corresponding to each set of sample data, depth features indicative of the relationship between the anatomical features of the sample locations, equivalent capacitances or equivalent resistances corresponding to the sample locations, a sample medical image containing the sample blood vessel, a segmented image of the sample blood vessel, and so forth, in a manner similar to that described in step 320. For example, the processing device 120 may determine anatomical features of the sample location based on the sample medical image and/or the segmented image of the sample vessel. For another example, the processing device 120 may determine the equivalent capacitance or equivalent resistance corresponding to the sample location using an existing lumped parameter model (e.g., WINDKESSEL MODEL). For another example, the processing device 120 may input anatomical features of the sample locations into a reference model, designating outputs of intermediate layers of the reference model as depth features indicative of relationships between the anatomical features of the sample locations. In some embodiments, the processing device 120 may perform pre-processing such as normalization, de-outlier, etc. on the obtained features to improve the accuracy of the features. In this way, in the model training process, various other features related to FFR determination are added as training inputs, and the influence of various features on FFR can be learned in the model training process, so that a machine learning model with higher accuracy is obtained.

In some embodiments, the structure and type of the initial model may be the same as the type of the target machine learning model, and the related description may refer to step 320, which is not described herein. In some embodiments, processing device 120 may obtain the initial model from one or more components of FFR determination system 100 or external devices over a network (e.g., network 150).

Training of the initial model may include one or more iterations, in each of which updating model parameters of the initial model based on a set of sample data may be included. In some embodiments, the optimization objective of the initial model training may include adjusting the model parameters such that the value of the loss function becomes smaller (e.g., minimizing the value of the loss function). The loss function may be used to characterize the difference between the FFR predicted by the initial model and the training label. Illustratively, the loss function may include a focal point loss function, a logarithmic loss function, a cross entropy loss, and the like. In some embodiments, the initial model may cease training if it satisfies the termination condition in a certain iteration. Illustratively, the termination conditions may include any one or a combination of the following: the value of the loss function obtained in a certain iteration is smaller than a threshold, a certain number of iterations have been performed, the loss function converges (e.g., the difference between the value of the loss function obtained in the previous iteration and the value of the loss function obtained in the current iteration is within a preset threshold), etc. In some embodiments, when the iteration does not meet the termination condition, the processing device may further update the initial model for the next iteration according to a preset algorithm (e.g., a back-propagation algorithm). If the termination condition is met in the current iteration, the processing device may complete training of the initial model.

In some embodiments of the present description, the target physiological parameter of the target subject, the reference physiological parameter, and the reference FFR may be processed using an FFR algorithm to determine a target FFR for the target location at the target physiological parameter. Possible benefits of embodiments of the present description include, but are not limited to: (1) In the prior art, after the physiological parameters of the target object are changed, the FFR determined based on the original physiological parameters is still used as the FFR of the current state of the target object, compared with the FFR, the FFR obtained by the application is the FFR obtained based on the actual physiological parameters of the target object (namely, under the target physiological parameters), so that the FFR is more in line with the current physiological conditions of the target object, and the accuracy is higher; (2) In some prior art, after the physiological parameters of the target object are changed, based on the new physiological parameters, FFR is calculated again through CFD simulation or the existing machine learning model, which requires processing a large amount of data, resulting in inefficiency and consuming a large amount of time and computing resources. Compared with the method, the multidimensional function or the target machine learning model can greatly reduce the data to be processed, even only the target physiological parameter, the reference physiological parameter and the reference FFR are needed, the processing process is simple and quick, the FFR obtaining efficiency can be greatly improved, and the computing resource is saved; (3) The FFR algorithm obtained by the application can be applied to any object containing the same type of blood vessel, and has wide application range; (4) According to some embodiments of the present application, during machine learning model training, by adding various other features related to FFR determination (e.g., anatomical features, sample medical images, etc.) as training inputs, the impact of the various features on FFR may be learned during model training, thereby obtaining a more accurate machine learning model. Accordingly, these features can be added as model inputs to achieve a more accurate target FFR in practical applications.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present invention.

Meanwhile, the specification uses specific words to describe the embodiments of the specification. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present description. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.

Furthermore, the order in which the elements and sequences are processed, the use of numerical letters, or other designations in the description are not intended to limit the order in which the processes and methods of the description are performed unless explicitly recited in the claims. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of various examples, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the present disclosure. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.

Likewise, it should be noted that in order to simplify the presentation disclosed in this specification and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure does not imply that the subject matter of the present description requires more features than are set forth in the claims. Indeed, less than all of the features of a single embodiment disclosed above.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the number allows for a 20% variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., referred to in this specification is incorporated herein by reference in its entirety. Except for application history documents that are inconsistent or conflicting with the content of this specification, documents that are currently or later attached to this specification in which the broadest scope of the claims to this specification is limited are also. It is noted that, if the description, definition, and/or use of a term in an attached material in this specification does not conform to or conflict with what is described in this specification, the description, definition, and/or use of the term in this specification controls.

Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.

Claims (10)

1. A method of determining fractional flow reserve, the method comprising:

obtaining a reference fractional flow reserve of a target position of a target object under a reference physiological parameter, the target position being a position on a target vessel of the target object; and

And processing the target physiological parameter of the target object, the reference physiological parameter and the reference blood flow reserve score by using a blood flow reserve score algorithm, and determining the target blood flow reserve score of the target position under the target physiological parameter.

2. The method of claim 1, wherein the fractional flow reserve algorithm comprises a machine learning model or a multidimensional function.

3. The method of claim 1, wherein the fractional flow reserve algorithm is generated according to the following process:

Obtaining a plurality of groups of sample physiological parameters;

determining, for each sample object of one or more sample objects, a sample fractional flow reserve for the sample object under each set of sample physiological parameters; and

The fractional flow reserve algorithm is determined based on the plurality of sets of sample physiological parameters and the sample fractional flow reserve.

4. The method of claim 3, wherein the fractional flow reserve of the sample object at each set of sample physiological parameters comprises fractional flow reserve of the sample object at a plurality of sample locations of the sample vessel, the determining the fractional flow reserve algorithm based on the plurality of sets of sample physiological parameters and the fractional flow reserve comprises:

Determining a plurality of sets of sample data based on the plurality of sets of sample physiological parameters and the sample fractional flow reserve, wherein each set of sample data corresponds to a sample location of a sample object in the one or more sample objects and includes a first set of sample physiological parameters, a second set of sample physiological parameters, a first sample fractional flow reserve for its corresponding sample location under the first set of sample physiological parameters, and a second sample fractional flow reserve for its corresponding sample location under the second set of sample physiological parameters; and

The fractional flow reserve algorithm is determined based on the plurality of sets of sample data.

5. The method of claim 4, wherein the fractional flow reserve algorithm is a multi-dimensional function for characterizing an effect of a change in a physiological parameter on fractional flow reserve, the determining the fractional flow reserve algorithm based on the plurality of sets of sample data comprising:

Acquiring an initial multidimensional function; and

The multi-dimensional function is generated by fitting the initial multi-dimensional function based on the plurality of sets of sample data.

6. The method of claim 4, wherein the fractional flow reserve algorithm is a machine learning model, and wherein determining the fractional flow reserve algorithm based on the plurality of sets of sample data comprises:

generating a training input and a training tag based on each set of the sample data, the training input comprising the first set of sample physiological parameters, the second set of sample physiological parameters, the first sample fractional flow reserve in the sample data, the training tag comprising the second sample fractional flow reserve in the sample data; and

The machine learning model is generated by training an initial model based on the training input and training labels for each set of sample data.

7. The method of claim 6, wherein the training input further comprises one or more of an anatomical feature of a sample location corresponding to the sample data, a depth feature indicative of a relationship between the anatomical features, an equivalent capacitance or equivalent resistance corresponding to the sample location, a medical image containing the sample vessel, a segmented image of the sample vessel.

8. The method of claim 7, wherein the depth features are obtained by:

inputting an anatomical feature of the sample location into a reference model, the reference model being used to determine fractional flow reserve; and

The output of the middle layer of the reference model is designated as the depth feature.

9. A system for determining fractional flow reserve, comprising:

An acquisition module configured to acquire a reference fractional flow reserve of a target location of a target subject under a reference physiological parameter, the target location being a location on a target vessel of the target subject; and

A determination model configured to process a target physiological parameter of the target subject, the reference physiological parameter, and the reference fractional flow reserve using a fractional flow reserve algorithm to determine a target fractional flow reserve for the target location at the target physiological parameter.

10. A system for determining fractional flow reserve, comprising:

at least one memory device for storing computer instructions;

At least one processor configured to execute the computer instructions to implement the method of any one of claims 1-8.