A method and system for myocardial perfusion image sequence recognition

Abstract

The embodiment of the specification provides a myocardial perfusion image sequence identification method and system, the method comprising: acquiring a plurality of myocardial perfusion image sequences of a target object; for each myocardial perfusion image sequence, determining a first characteristic value corresponding to the myocardial perfusion image sequence, the first characteristic value being related to one or more blood flow characteristic parameters; and based on the first characteristic value of each myocardial perfusion image sequence, identifying a first image sequence and a second image sequence from the plurality of myocardial perfusion image sequences, the first image sequence being collected under a load state of the target object, and the second image sequence being collected under a resting state of the target object.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of medical image analysis, and in particular to a myocardial perfusion image sequence identification method and system. BACKGROUND

[0002] In recent years, myocardial perfusion imaging technology based on computed tomography (CT) has become an important method for evaluating heart disease. This technology can evaluate myocardial blood flow and ischemia by measuring myocardial perfusion and calculating various quantitative parameters, such as myocardial perfusion reserve index (MPRi) and the like.

[0003] Myocardial perfusion imaging sequences include resting myocardial perfusion image sequences, stress myocardial perfusion image sequences, and various other types. Traditional methods require users to manually mark the type of the current imaging sequence in order to perform subsequent myocardial perfusion parameter calculations. Furthermore, when performing myocardial perfusion parameter calculations, users need to sequentially perform operations such as reading data, finding the best phase, motion registration, myocardial segmentation, defining the aortic reference point, and previewing the results, which occupies a relatively long time and has human errors.

[0004] Therefore, an automated myocardial perfusion image sequence identification and myocardial perfusion parameter calculation method and system are provided to reduce human intervention and improve the accuracy of the calculation results. SUMMARY

[0005] One of the embodiments of the present specification provides a myocardial perfusion image sequence identification method, comprising: acquiring a plurality of myocardial perfusion image sequences of a target object; determining, for each of the myocardial perfusion image sequences, a first feature value corresponding to the myocardial perfusion image sequence, the first feature value being related to one or more blood flow characteristic parameters; and identifying, based on the first feature value of each of the myocardial perfusion image sequences, a first image sequence and a second image sequence from the plurality of myocardial perfusion image sequences, the first image sequence being acquired under a stress state of the target object, and the second image sequence being acquired under a resting state of the target object.

[0006] One of the embodiments of the present specification provides a myocardial perfusion image sequence identification system, comprising: an acquisition module configured to acquire a plurality of myocardial perfusion image sequences of a target object; a determination module configured to determine, based on each of the myocardial perfusion image sequences, a first feature value corresponding to each of the myocardial perfusion image sequences, the first feature value being related to one or more blood flow characteristic parameters; and an identification module configured to identify, based on the first feature value of each of the myocardial perfusion image sequences, a first image sequence and a second image sequence from the plurality of myocardial perfusion image sequences, the first image sequence being acquired under a stress state of the target object, and the second image sequence being acquired under a resting state of the target object. BRIEF DESCRIPTION OF DRAWINGS

[0007] The present specification will be further described in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. The embodiments are not restrictive, and in the embodiments, the same numbers represent the same structures or operations, and:

[0008] Figure 1 is a schematic diagram of an application scenario of a myocardial perfusion image sequence recognition system according to some embodiments of the present specification;

[0009] Figure 2 is an exemplary block diagram of a myocardial perfusion image sequence recognition system according to some embodiments of the present specification;

[0010] Figure 3 is an exemplary flowchart of a myocardial perfusion image sequence recognition method according to some embodiments of the present specification;

[0011] Figure 4 is a schematic diagram of an exemplary flow of recognizing a first image sequence and a second image sequence according to some embodiments of the present specification;

[0012] Figure 5 is a schematic diagram of another exemplary flow of recognizing a first image sequence and a second image sequence according to some embodiments of the present specification;

[0013] Figure 6 is a schematic diagram of still another exemplary flow of recognizing a first image sequence and a second image sequence according to some embodiments of the present specification;

[0014] Figure 7 is an exemplary flowchart of myocardial perfusion parameter calculation according to some embodiments of the present specification. DETAILED DESCRIPTION

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor. Unless it is obvious from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structures or operations.

[0016] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing 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.

[0017] The singular forms "a," "an," and "the" do not necessarily mean the same thing, and / or "the" do not necessarily refer to the singular unless the context clearly indicates so. Generally, the terms "comprise" and "comprising" are intended to be open-ended, meaning that they include the steps and elements that have been explicitly identified, but not to the exclusion of other steps or elements. The term "comprise" is used in its broadest sense, meaning that it includes the steps and elements that have been explicitly identified, but not to the exclusion of other steps or elements.

[0018] Flow diagrams are used in this specification to illustrate the operations according to embodiments of the present specification in a sequential order. It will be understood that some of the operations can be performed in an order different than that shown, and / or some operations can be performed concurrently, that the drawings shown in this specification are not necessarily to scale, and that some embodiments can include more or fewer operations than are indicated in this specification. In addition, operations can be repeated, and / or the operations can be performed in real time.

[0019] Figure 1 is a schematic diagram of an application scenario of a myocardial perfusion image sequence recognition system according to some embodiments of the present specification.

[0020] As shown in Figure 1 , the application scenario 100 of the myocardial perfusion image sequence recognition system can include a medical scanning device 110, a network 120, a terminal 130, a processing device 140, and a storage device 150. The components in the application scenario 100 can be connected in various ways. For example only, as shown in Figure 1 , the medical scanning device 110 can be connected to the processing device 140 through the network 120. The medical scanning device 110 can be directly connected to the processing device 140 (as shown by the bidirectional arrow in the dashed line connecting the medical scanning device 110 and the processing device 140). The storage device 150 can be directly or through the network 120 connected to the processing device 140.

[0021] The medical scanning device 110 can scan a scanned object and / or generate data about the scanned object. For example, the medical scanning device 110 can include a CT device, a PET / CT device, an MRI / CT device, etc. In this specification, the scanned object can also be referred to as a scanned subject, a target object, a target, or a detected object. In some embodiments, the target object can be a patient, an animal, etc. When the target object needs to be scanned, after the scanned object enters the scanning area 115 through the examination bed 116, the medical scanning device 110 can obtain the medical image corresponding to the target object. The medical image includes a single image, and can also include an image sequence sequentially acquired at multiple time points.

[0022] In some embodiments, the medical images can be a sequence of images of the target subject acquired in a resting state. In some embodiments, the medical images can also be a sequence of images of the target subject acquired in a stress state. For example, the sequence of images in the stress state can be acquired by the medical scanning device 110 after the target subject performs appropriate movements under the guidance of a medical professional.

[0023] The network 120 can include any suitable network that facilitates the exchange of information and / or data for the application scenario 100. In some embodiments, one or more components of the application scenario 100 (e.g., the medical scanning device 110, the terminal 130, the processing device 140, or the storage device 150) can communicate information and / or data with one or more other components of the application scenario 100 via the network 120. For example, the processing device 140 can obtain medical images of a scanned subject from the medical scanning device 110 via the network 120. In some embodiments, the network 120 can be any one or more of a wired network or a wireless network. In some embodiments, the network can be of various topologies, such as point-to-point, shared, hub-and-spoke, etc., or a combination of topologies.

[0024] The terminal 130 can include a mobile device 130-1, a tablet 130-2, a notebook computer 130-3, etc., or any combination thereof. In some embodiments, the terminal 130 can interact with other components in the application scenario 100 through the network 120. For example, the terminal 130 can receive data such as medical images sent by the medical scanning device 110. In some embodiments, the terminal 130 can receive information and / or instructions input by a user (e.g., a user of the medical scanning device 110, such as a doctor) and send the received information and / or instructions to the medical scanning device 110 or the processing device 140 via the network 120. For example, the doctor can input operation instructions for the medical scanning device 110 through the terminal 130. In some embodiments, the terminal 130 can display images reconstructed from scanning data.

[0025] In some embodiments, the terminal 130 can include a display interface for displaying information related to the medical images. For example, the type of the medical images (e.g., a stress sequence of images, a resting sequence of images) and the label of the type (e.g., a stress label, a resting label, etc.) can be displayed according to the display interface.

[0026] The processing device 140 can process data and / or information obtained from the medical scanning device 110, the terminal 130, and / or the storage device 150. For example, the processing device 140 can obtain medical images of a scanned subject.

[0027] In some embodiments, the processing device 140 can be a single server or a group of servers. The group of servers can be centralized or distributed. In some embodiments, the processing device 140 can be local or remote. The processing device 140 can be directly connected to the medical scanning device 110, the terminal 130, and the storage device 150 to access the stored or acquired information and / or data. In some embodiments, the processing device 140 can be implemented on a cloud platform. The cloud platform can include, by way of example only, a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an on-premise cloud, a multi-tier cloud, or the like, or any combination thereof.

[0028] The storage device 150 can store data and / or instructions. In some embodiments, the storage device 150 can store data acquired from the medical scanning device 110, the terminal 130, and / or the processing device 140. For example, the storage device 150 can store medical images obtained by the user scanning device, etc. In some embodiments, the storage device 150 can store data and / or instructions used by the processing device 140 to perform the exemplary methods described in this specification. For example, the storage device 150 can store instructions for the processing device 140 to perform the methods shown in the respective flowcharts. In some embodiments, the storage device 150 can include a mass storage device, a removable storage device, a volatile read / write memory, a read-only memory (ROM), or the like, or any combination thereof. In some embodiments, the storage device 150 can be implemented on a cloud platform.

[0029] In some embodiments, the storage device 150 can be connected to the network 120 to communicate with one or more components of the application scenario 100 (e.g., the medical scanning device 110, the terminal 130, the processing device 140, etc.). One or more components of the application scenario 100 can access the data or instructions stored in the storage device 150 via the network 120. In some embodiments, the storage device 150 can be directly connected to or in communication with one or more components of the application scenario 100. In some embodiments, the storage device 150 can be part of the processing device 140.

[0030] In some embodiments, multiple medical scanning devices 110 can be included in the application scenario 100, such as medical scanning devices 110 belonging to multiple hospitals or research institutions. The processing device 140 can remotely communicate with these medical scanning devices and analyze the myocardial perfusion image sequences collected by them. For example, the processing device 140 can be a computing device of a myocardial perfusion image analysis platform that can provide analysis and processing services for myocardial perfusion images for multiple hospitals or research institutions.

[0031] Figure 2 is an exemplary schematic diagram of modules of a myocardial perfusion image sequence recognition system according to some embodiments of the present specification.

[0032] As shown in Figure 2 The myocardial perfusion image sequence identification system 200 (hereinafter can be referred to as identification system 200) can include an acquisition module 210, a determination module 220, and an identification module 230.

[0033] The acquisition module 210 can be configured to acquire a plurality of myocardial perfusion image sequences of a target object.

[0034] In some embodiments, the acquisition module 210 can be further configured to acquire a plurality of initial scan sequences of the target object.

[0035] The determination module 220 can be configured to determine, based on each myocardial perfusion image sequence, a first feature value corresponding to each myocardial perfusion image sequence, the first feature value being related to one or more blood flow characteristic parameters.

[0036] In some embodiments, the determination module 220 can be further configured to determine, based on the identification result of the myocardial perfusion image sequence, a sequence label, the sequence label including a load label corresponding to the first image sequence and a rest label corresponding to the second image sequence.

[0037] The identification module 230 can be configured to identify, based on the first feature value of each myocardial perfusion image sequence, a first image sequence and a second image sequence from the plurality of myocardial perfusion image sequences, the first image sequence being acquired under a load state of the target object, and the second image sequence being acquired under a rest state of the target object.

[0038] In some embodiments, the one or more blood flow characteristic parameters include a plurality of blood flow characteristic parameters arranged in sequence, and the identification module 230 can be further configured to sequentially process the plurality of blood flow characteristic parameters as follows: determining whether the first feature value corresponding to the blood flow characteristic parameter satisfies a preset condition; in response to the first feature value corresponding to the blood flow characteristic parameter satisfying the preset condition, identifying the first image sequence and the second image sequence based on the first feature value of the blood flow characteristic parameter; and in response to the first feature value corresponding to the blood flow characteristic parameter not satisfying the preset condition, processing the next blood flow characteristic parameter.

[0039] In some embodiments, the identification module 230 can be further configured to, in response to the first feature values of the plurality of blood flow characteristic parameters all not satisfying the preset condition, acquire image metadata corresponding to each myocardial perfusion image sequence; determine, based on the image metadata corresponding to each myocardial perfusion image sequence, a second feature value of the target object, the second feature value being related to a heart rate; and determine the first image sequence and the second image sequence based on the second feature value corresponding to each myocardial perfusion image sequence.

[0040] In some embodiments, the identifying module 230 can be further configured to identify, for each blood flow feature parameter in the plurality of blood flow feature parameters, a candidate first image sequence and a candidate second image sequence based on the first feature value of the blood flow feature parameter; and determine the first image sequence and the second image sequence based on the candidate first image sequence and the candidate second image sequence corresponding to each blood flow feature parameter.

[0041] In some embodiments, the identifying module 230 can be further configured to process the first feature value of each myocardial perfusion image sequence using an identifying model to determine the first image sequence and the second image sequence, the identifying model being a machine learning model.

[0042] In some embodiments, the identifying module 230 can be further configured to identify a plurality of initial scan sequences to determine the plurality of myocardial perfusion sequences and the delayed enhancement sequence.

[0043] In some embodiments, the identifying system 200 can further include a computing module 240.

[0044] The computing module 240 can be configured to determine a myocardial perfusion parameter based on the first image sequence and the second image sequence.

[0045] In some embodiments, the identifying system 200 can further include a display module (not shown).

[0046] The display module can be configured to present the first image sequence and the stress label, and the second image sequence and the rest label on a user interface.

[0047] Further description of the above modules can be found elsewhere in this specification (e.g., in the detailed description), and will not be repeated here. Figure 3

[0048] It should be noted that the myocardial perfusion image sequence identifying system 200 is provided for illustrative purposes only and is not intended to limit the scope of this specification. Numerous modifications or variations can be made in light of this description. For example, the myocardial perfusion image sequence identifying system 200 can include other suitable component(s) to perform substantially the same or different functions. However, such variations and modifications do not depart from the scope of this specification.

[0049] Figure 3 is an exemplary flowchart of a myocardial perfusion image sequence identifying method according to some embodiments of the present specification.

[0050] In some embodiments, the flow 300 can be performed by the identifying system 200. As shown in Figure 3 , the flow 300 includes the following steps.

[0051] At step 310, a plurality of myocardial perfusion image sequences of a target subject are obtained.​

[0052] The target object can refer to an object that needs to be analyzed for perfusion, such as diagnosis of heart disease, assessment of myocardial blood flow and ischemia. The target object can be a human or an animal or a part thereof.

[0053] The myocardial perfusion image sequence can refer to a set of images that are acquired sequentially at multiple time points. For example, a myocardial perfusion image sequence can include a plurality of myocardial perfusion images acquired from a target object over a period of time (e.g., 5 min, 20 min).

[0054] The plurality of myocardial perfusion image sequences can include image sequences of multiple sequence types. For example, the sequence type of a myocardial perfusion image sequence can be related to the motion state, emotion, condition (e.g., stage of disease), contrast agent diffusion state, etc. of the target object when the sequence is acquired. For example, a myocardial perfusion image sequence can include a stress image sequence, a rest image sequence, etc.

[0055] The stress image sequence can also be referred to as a first image sequence, which refers to an image sequence acquired from a target object in a stress state. The stress state refers to a state in which the target object is stimulated (e.g., motion or a period of time after the end of motion). For example, the target object can be guided by a medical professional (e.g., a doctor) to perform a targeted (e.g., specific intensity and duration) limb movement (e.g., walking, running, arm stretching, etc.), and perfusion imaging is performed during or immediately after the movement to obtain the first image sequence. For another example, a specific drug can be injected into the target object to make its body in a stress state similar to that after exercise. The changes in myocardial oxygen consumption, heart rate, etc. of the target object in the stress state can be distinguished from those in other states (e.g., non-stress state).

[0056] The rest image sequence can also be referred to as a second image sequence, which refers to an image sequence acquired from a target object in a rest state. The rest state can represent a state in which the target object is not externally stimulated, for example, a state of peace when the target object is calm, which can be a state relative to the stress state. For example only, the second image sequence can be acquired before the first image sequence is acquired (i.e., before the target object is moved), or a long time after the first image sequence is acquired (e.g., several hours later).

[0057] It should be noted that the sequence type of at least part of the myocardial perfusion image sequences obtained in step 310 is unknown. For example, when the myocardial perfusion image analysis platform receives myocardial perfusion image sequences from a hospital or research institution, the sequence types of these sequences are unknown. Traditional methods rely on manual judgment of sequence types, which has problems such as low efficiency, high error rate, etc. There are also traditional methods that identify sequence types based on the heart rate value tags of the DICOM files corresponding to multiple sequences. However, problems such as missing or incorrect heart rate value tags can cause errors in the identification results. In addition, the identification method based on the heart rate value tag has a large error. Under stress, the patient's heart rate may not change much, but there are clinical symptoms such as increased cardiac output, decreased blood pressure, chest tightness, chest pain, etc. The present specification solves the above problems by performing steps 320 and 330 described below to achieve automatic identification of sequence types.

[0058] In some embodiments, step 310 can include obtaining a plurality of initial scan sequences of the target object, and identifying the plurality of initial scan sequences to determine the plurality of myocardial perfusion image sequences and the delayed enhancement image sequence.

[0059] The initial scan sequence can refer to the image sequence before identification. For example, the initial scan sequence can be a plurality of scan image sequences obtained by a medical scanning device.

[0060] The delayed enhancement image sequence can refer to an image sequence acquired within a certain period of time after a drug (such as a contrast agent) is injected into the target object. For example, the acquired image sequence is a delayed enhancement image sequence 5 minutes to 8 minutes after the drug is injected. The delayed enhancement image sequence can be used to evaluate the severity of myocardial fibrosis and the reversibility of myocardial ischemia of the target object.

[0061] For example, for each initial scan sequence, the number of phases of the sequence can be obtained; if the sequence contains more than a preset number (such as 9) of phases, the sequence is a perfusion image sequence; then, it is determined whether the images in the sequence contain a heart, and if all phase images contain a heart, the perfusion image sequence is a myocardial perfusion image sequence. Whether the image contains a heart can be determined using a part recognition model, which can be a trained machine learning or deep learning model.

[0062] For another example, if an initial scan sequence meets the preset conditions corresponding to the delayed enhancement image sequence, it is considered to be a delayed enhancement image sequence. The preset sequence conditions can include one or more of the following: the number of phases contained is less than a certain threshold (such as 4, 5), the acquisition time is within a certain period of time (such as 3 to 5 minutes) after contrast agent injection, etc.

[0063] In some embodiments, the identification system 200 can obtain the myocardial perfusion image sequences by scanning a region of interest (e.g., a thorax) of the target subject using a medical device (e.g., a CT device, a PET-CT device, etc.). The identification system 200 can also obtain the myocardial perfusion image sequences from a third-party platform (e.g., other medical institutions), which is not limited in the present specification.

[0064] At step 320, for each myocardial perfusion image sequence, a first feature value corresponding to the myocardial perfusion image sequence is determined, which is related to one or more blood flow characteristic parameters.

[0065] The blood flow characteristic parameter can refer to a characteristic parameter that can reflect a blood condition. For example, the blood flow characteristic parameter can include blood pressure, blood flow, blood volume, peak time, capillary blood volume, tissue permeability, etc.

[0066] In some embodiments, the one or more blood flow characteristic parameters can include at least one of myocardial blood flow (MBF) and myocardial blood volume (MBV).

[0067] In some embodiments, the myocardial perfusion image sequence can be used to determine values of the blood flow characteristic parameter at a plurality of location points of the target subject. The first feature value of the blood flow characteristic parameter can be determined based on each image or partial image in the sequence.

[0068] In some embodiments, the first feature value can be one or a combination of a minimum value, a maximum value, or an average value of the blood flow characteristic parameter. Taking the first feature value corresponding to the myocardial blood flow as an example, the identification system 200 can determine parameter values (hereinafter referred to as myocardial blood flow values) of the myocardial blood flow at a plurality of location points on the target subject based on each image or partial image in the myocardial perfusion image sequence, which can form a parameter map corresponding to the myocardial blood flow. The first feature value can be one or a combination of a minimum value, a maximum value, or an average value of the parameter map.

[0069] The first feature value can also be a vector or an array, etc., which can represent a combination of the minimum value, the maximum value, or the average value. For example, for a myocardial perfusion image sequence, the first feature value corresponding to the myocardial blood flow can be represented as a vector (min, max, avg), where the elements min, max, and avg of the vector represent the minimum value, the maximum value, and the average value of the myocardial blood flow values, respectively. The representation of the first feature value is not limited in the present specification.

[0070] In some embodiments, for one myocardial perfusion image sequence, the identification system 200 can obtain parameter values corresponding to one or more blood flow characteristic parameters in the myocardial perfusion image sequence to determine a first characteristic value corresponding to the myocardial perfusion image sequence. For example, if the first characteristic value is related to two blood flow characteristic parameters, such as myocardial blood flow and myocardial blood volume, the first characteristic value of the myocardial perfusion image sequence includes one or a combination of a minimum value, a maximum value, or an average value of the myocardial blood flow value, and one or a combination of a minimum value, a maximum value, or an average value of the myocardial blood volume. The first characteristic values corresponding to multiple blood flow characteristic parameters can be represented in the form of a sequence, a vector, or a vector matrix.

[0071] At step 330, based on the first characteristic value of each myocardial perfusion image sequence, the identification system 200 identifies a first image sequence and a second image sequence from the multiple myocardial perfusion image sequences, the first image sequence being acquired under a stress state of the target object, and the second image sequence being acquired under a resting state of the target object.

[0072] As described above, the first image sequence can refer to a myocardial perfusion image sequence acquired under a stress state of the target object.

[0073] The second image sequence can refer to a myocardial perfusion image sequence acquired under a resting state of the target object.

[0074] In some embodiments, the identification system 200 can identify the first image sequence and the second image sequence from the multiple myocardial perfusion image sequences based on the first characteristic value of the myocardial perfusion image sequence; or further identify the first image sequence and the second image sequence from the multiple myocardial perfusion image sequences in combination with the second characteristic value corresponding to the image metadata. For details about identifying the first image sequence and the second image sequence based on the first characteristic value and the second characteristic value, see Figure 4 and the description thereof.

[0075] In some embodiments, for each blood flow characteristic parameter in the multiple blood flow characteristic parameters, the identification system 200 can identify a candidate first image sequence and a candidate second image sequence based on the first characteristic value of the blood flow characteristic parameter, and determine the first image sequence and the second image sequence based on the candidate first image sequence and the candidate second image sequence corresponding to each blood flow characteristic parameter. For details, see Figure 5 and the description thereof.

[0076] In some embodiments, the identification system 200 can further process the first characteristic value of each myocardial perfusion image sequence using an identification model to determine the first image sequence and the second image sequence, the identification model being a machine learning model. For details about the identification model, see Figure 6 and the description thereof.

[0077] In some embodiments, the recognition system 200 can determine a sequence label based on the recognition result of the myocardial perfusion image sequence. The recognition result can include the first image sequence and the second image sequence, and the sequence label can include a stress label corresponding to the first image sequence and a rest label corresponding to the second image sequence. The recognition system 200 can further present the first image sequence and the stress label, and the second image sequence and the rest label on a user interface.

[0078] The user interface can refer to an interface developed based on computer programming technology, which can be a client display interface or a web-based display interface.

[0079] In some embodiments, the recognition result of the myocardial perfusion image sequence can further include a recognition result of a delay enhancement image sequence, and the sequence label can further include a delay enhancement label corresponding to the delay enhancement image sequence.

[0080] It should be noted that the recognition result and the label corresponding to the recognition result can also be a recognition result and a label of other types of sequences.

[0081] The sequence label can refer to annotation information of a certain myocardial perfusion image sequence, which can be presented in the form of text, symbols, color annotations, etc. on a user interface.

[0082] In some embodiments, the recognition system 200 can further determine a myocardial perfusion parameter based on the first image sequence and the second image sequence. For details about determining the myocardial perfusion parameter, see Figure 7 and the description thereof.

[0083] Some embodiments of the present specification can accurately recognize the rest image sequence and the stress image sequence by automatically processing the myocardial perfusion image sequence. At the same time, by automatically calculating and analyzing the perfusion quantitative parameters of the rest image sequence and the stress image sequence, the accuracy of the calculation results can be improved, and the time, effort and errors consumed by manual processing can be reduced. In addition, through the user interface, the aforementioned image sequences and the calculation results of the perfusion quantitative parameters can be more intuitively compared and presented, reducing manual operations.

[0084] Figure 4 is an exemplary schematic diagram of a process of recognizing the first image sequence and the second image sequence according to some embodiments of the present specification.

[0085] In some embodiments, the process 400 as shown in Figure 4 may be performed by the recognition system 200. The process 400 can include the following steps.

[0086] Step 410, determining a blood flow characteristic parameter set {P1, …, P n}.

[0087] The blood flow characteristic parameter set refers to the set of one or more blood flow characteristic parameters described in step 320, where n represents the number of blood flow characteristic parameters (e.g., 1, 2, 4, etc.). As an example only, the blood flow characteristic parameter set P... n This can include myocardial blood flow (MBF) and myocardial blood volume (MBV), where n is 2.

[0088] In some embodiments, one or more blood flow feature parameters may include only one blood flow feature parameter, i.e., n equals 1. The identification system 200 may perform the first identification process only on this blood flow feature parameter. In some embodiments, one or more blood flow feature parameters may include multiple blood flow feature parameters arranged in sequence, i.e., n is greater than or equal to 2. The sequential arrangement may be based on a preset priority of the blood flow feature parameters (e.g., descending order). For example, myocardial blood flow and myocardial blood volume have decreasing priorities, so myocardial blood flow and myocardial blood volume are the first and second parameters in the set of blood flow feature parameters, respectively. The order of the multiple blood flow feature parameters may be randomly determined or manually specified. The identification system 200 may perform the first identification process on the multiple blood flow feature parameters sequentially. Figure 4 As shown, the first identification process may include one of steps 420, 431, 441, and 442.

[0089] For illustrative purposes, the following description uses the blood flow characteristic parameter Pi as an example to illustrate the specific implementation of the first identification process.

[0090] Step 420: Determine the blood flow characteristic parameter P i The corresponding first eigenvalue V i .

[0091] i can be equal to the number of times the first identification process is executed. For example, when the first identification process is executed for the first time, i equals 1, and step 420 determines the blood flow feature parameter set P. n The first characteristic value V1 of the first blood flow characteristic parameter P1 (such as myocardial blood flow) in the dataset. When the first identification process is executed for the second time, i=2, step 420 will determine the blood flow characteristic parameter set P. n The first characteristic value V2 of the second blood flow characteristic parameter P2 (such as myocardial blood volume).

[0092] First eigenvalue V i This can include blood flow characteristic parameters P for each myocardial perfusion image sequence. i The corresponding characteristic values. Taking myocardial blood flow as an example, the first characteristic value V i It is a set of eigenvalues ​​{V} containing n values. i1 ,…,V in}, where V ijrepresents a feature value related to myocardial blood flow of the jth (j is less than or equal to n) myocardial perfusion image sequence. For example, V ij may be one or a combination of minimum value, maximum value and average value related to myocardial blood flow. For the first feature value and its representation, please refer to Figure 3 and its description.

[0093] Step 431, judge whether the first feature value V i satisfies the preset condition.

[0094] In some embodiments, the preset condition can include a load preset condition and a resting preset condition.

[0095] The load preset condition can refer to a condition that the first feature value satisfies the load sequence; the resting preset condition can refer to a condition that the first feature value satisfies the resting sequence.

[0096] The preset condition can be pre-set for the first feature value corresponding to each blood flow feature parameter. For example, for myocardial blood flow, the load preset condition can be that the myocardial blood flow is greater than or equal to a blood flow threshold (such as 150 ml / 100ml / min); the resting preset condition can be that the myocardial blood flow is less than the blood flow threshold. For myocardial blood volume, the resting preset condition can be that the myocardial blood volume is greater than or equal to a blood volume threshold (such as 21.6 ml / 100ml); the resting preset condition can be that the myocardial blood volume is less than the blood volume threshold.

[0097] In some embodiments, the preset condition can be related to the difference between the maximum value and the minimum value of the first feature value V i . For example, the preset condition can be that the difference between the maximum value and the minimum value is greater than a preset threshold. Taking myocardial blood flow as an example, V i may include feature values (such as average values) of myocardial blood flow corresponding to a plurality of myocardial perfusion image sequences, including V i1 to V in . The difference between the maximum value and the minimum value of the first feature value V i1 to V in may be determined. When the difference is greater than a preset threshold, it can be considered that the preset condition is satisfied.

[0098] In response to the first feature value Vi satisfying the preset condition, step 441 can be performed. Step 441 includes identifying the first image sequence and the second image sequence based on the first feature value of the blood flow feature parameter.

[0099] For example, if the first feature value of a certain myocardial perfusion image sequence satisfies the load preset condition, the myocardial perfusion image sequence is the first image sequence; if the first feature value of a certain myocardial perfusion image sequence satisfies the resting preset condition, the myocardial perfusion image sequence is the second image sequence. For another example, if the first feature value Vi If the difference between the maximum value and the minimum value is greater than a preset threshold, the myocardial perfusion image sequence corresponding to the maximum value can be taken as the first image sequence, and the myocardial perfusion image sequence corresponding to the minimum value can be taken as the second image sequence.

[0100] In response to the first feature value V i If the preset condition is not met, step 442 can be performed. Step 442 can include determining whether i is less than the number n of blood flow feature parameters.

[0101] When i is less than n, it indicates that there are still blood flow feature parameters in the blood flow feature parameter set that can be processed, and the recognition system 200 can set i = i + 1, and acquire the next blood flow feature parameter P i in the blood flow feature parameter set, and repeat the above steps 420, 441, and 442 to perform the next first recognition processing.

[0102] When i is not less than n, it indicates that all blood flow feature parameters in the blood flow feature parameter set have been processed, and the first feature value V i of all blood flow feature parameters does not meet the preset condition, indicating that the first recognition processing cannot identify the first image sequence and the second sequence image.

[0103] In some embodiments, in response to the first feature values corresponding to the sequentially arranged plurality of blood flow feature parameters all not meeting the preset condition (i.e., the first recognition processing cannot identify the first image sequence and the second sequence image), the recognition system 200 can perform a recognition processing (hereinafter can be referred to as a second recognition processing) based on the image metadata corresponding to the myocardial perfusion image sequence. As shown in FIG. 4B, the second recognition processing can include the following steps 451 to 453. Figure 4

[0104] Step 451, acquiring image metadata corresponding to each myocardial perfusion image sequence.

[0105] The image metadata of the myocardial perfusion image sequence can include acquisition parameters corresponding to the image sequence, such as acquisition time, physiological parameters (such as heart rate) of the target object at the time of acquisition, etc. In some embodiments, the myocardial perfusion image sequence can be a DICOM file, and the image metadata can include tag data of the DICOM file. In some embodiments, the image metadata at least includes heart rate information of the target object at the time of image acquisition.

[0106] Step 452, determining a second feature value of the target object based on the image metadata corresponding to each myocardial perfusion image sequence.

[0107] ​The second characteristic value can be related to a physiological characteristic of the target object at the time of the myocardial perfusion image sequence acquisition. For example, the second characteristic value can be related to a heart rate. The second characteristic value can include various forms such as a maximum value, a minimum value, or an average value. For example, the second characteristic value can be a maximum value, a minimum value, or an average value of the heart rate.

[0108] At step 453, the first image sequence and the second image sequence are determined based on the second characteristic value corresponding to each myocardial perfusion image sequence.

[0109] For example, the myocardial perfusion image sequence with the maximum second characteristic value can be determined as the first image sequence, and the myocardial perfusion image sequence with the minimum second characteristic value can be determined as the second image sequence. For another example, the myocardial perfusion image sequence with the second characteristic value greater than or equal to a first preset heart rate threshold can be determined as the first image sequence, and the myocardial perfusion image sequence with the second characteristic value less than a second preset heart rate threshold can be determined as the second image sequence. The first and second preset heart rate thresholds can be equal or unequal, which can be determined based on medical experience. For example, the first preset heart rate threshold can be 100 bpm.

[0110] It should be noted that the above description of the process 400 is merely for example and illustration, and does not limit the scope of the present specification. Various modifications and changes can be made to the process under the guidance of the present specification. For example, when there is only one blood flow characteristic parameter, step 442 can be omitted. For another example, the second characteristic value is not limited to a heart rate value, and can also be other physiological characteristic values. For another example, when the first characteristic values of all blood flow characteristic parameters do not satisfy the preset condition, multiple myocardial perfusion image sequences can be sent to a user for manual judgment, instead of performing steps 451-453. For another example, the first identification process can be performed simultaneously on multiple blood flow characteristic parameters. For example, the first identification process can be performed on myocardial blood flow and myocardial blood volume in a blood flow characteristic parameter set. At this time, the blood flow characteristic parameters P i The multi-dimensional vector or matrix can be used.

[0111] In some embodiments of the present specification, through the first identification process, the first image sequence and the second image sequence can be automatically determined in combination with multiple blood flow characteristic parameters. Through comprehensive evaluation of multiple sequentially arranged blood flow characteristic parameters, the identification results of the first image sequence and the second image sequence are more accurate. In addition, in the case that the first identification process fails to successfully identify, the second identification process can be performed according to the information of the image metadata, improving the flexibility of the identification process. Through the combination of the first identification process and the second identification process, the unreliable problem caused by identification based on the heart rate alone can be avoided, and the robustness of the identification results of the first image sequence and the second image sequence is ensured.

[0112] Figure 5 is an exemplary schematic diagram of another method of identifying a first image sequence and a second image sequence according to some embodiments of the present specification.

[0113] In some embodiments, the identification system 200 can determine a plurality of candidate first image sequences and candidate second image sequences based on processing of a plurality of blood flow characteristic parameters, and determine the first image sequence and the second image sequence based on the plurality of candidate first image sequences and candidate second image sequences.

[0114] A candidate first image sequence corresponding to a blood flow characteristic parameter can refer to a load sequence identified based on a first characteristic value of the blood flow characteristic parameter.

[0115] A candidate second image sequence corresponding to a blood flow characteristic parameter can refer to a rest sequence identified based on a first characteristic value of the blood flow characteristic parameter.

[0116] For example, according to a first characteristic value corresponding to myocardial blood flow, the identification result obtained can include a candidate first image sequence A1 and a candidate second image sequence B1. According to a first characteristic value corresponding to myocardial blood volume, the identification result obtained can also include a candidate first image sequence A2 and a candidate second image sequence B2. According to a first characteristic value corresponding to a combination of a plurality of blood flow characteristic parameters (such as a combination of myocardial blood flow and myocardial blood volume), the identification result obtained can also include a candidate first image sequence A3 and a candidate second image sequence B3.

[0117] It should be noted that for a blood flow characteristic parameter, the identification system 200 can determine the load sequence and the rest sequence (i.e., the candidate first image sequence and the candidate second image sequence) corresponding to the blood flow characteristic parameter based on the sequence identification method disclosed elsewhere in the present specification. For example, the identification system 200 can determine the candidate first image sequence or the candidate second image sequence through Figure 4 the first identification process or the second identification process. For details of the first identification process and the second identification process, please refer to Figure 4 and the description thereof.

[0118] In some embodiments, the identification system 200 can determine the first image sequence and the second image sequence according to the plurality of candidate first image sequences and candidate second image sequences.

[0119] For example, the myocardial perfusion image sequence with the highest frequency of occurrence in the candidate first image sequence can be taken as the first image sequence, and the myocardial perfusion image sequence with the highest frequency of occurrence in the candidate second image sequence can be taken as the second image sequence.

[0120] In some embodiments, the recognition system 200 can determine the first image sequence and the second image sequence according to the feature weight coefficients of different blood flow feature parameters.

[0121] The feature weight coefficient can represent the contribution degree of different blood flow feature parameters to the recognition result. The greater the feature weight coefficient, the greater the contribution degree, and the corresponding recognition result can be used as the final recognition result (e.g., the first image sequence and the second image sequence). The feature weight coefficient can be pre-set according to actual needs. For example, it can be determined according to medical experience, and can also be determined according to the basic information (e.g., age, gender, and disease) of different target objects. In some embodiments, different blood flow feature parameters can correspond to different feature weight coefficients. For example, the feature weight coefficients of the myocardial blood flow and the myocardial blood volume can be set to 0.6 and 0.4, respectively.

[0122] For example, as shown in Figure 5 , the recognition system 200 can obtain a plurality of first feature values 520 corresponding to a plurality of blood flow feature parameters 510.

[0123] The plurality of blood flow feature parameters 510 can include a blood flow feature parameter P1, a blood flow feature parameter P2, …, and a blood flow feature parameter P n N. The plurality of first feature values 520 can include a first feature value V1 corresponding to the blood flow feature parameter P1, a first feature value V2 corresponding to the blood flow feature parameter P2, …, and a first feature value V n N corresponding to the blood flow feature parameter P n N. For more information about the first feature value, see Figure 3 and the description thereof.

[0124] The recognition system 200 can determine a corresponding candidate recognition result 530 according to each first feature value 520. As shown in Figure 5 , the candidate recognition result 530 can include a candidate first image sequence S1 and a candidate second image sequence S1' corresponding to the first feature value V1, …, a candidate first image sequence Sn and a candidate second image sequence Sn' corresponding to the first feature value V n N.

[0125] Further, the recognition system 200 can determine the first image sequence and the second image sequence based on the alternative recognition result 530. For example, assuming there are m myocardial perfusion image sequences, the xth myocardial perfusion image sequence is the most frequent as the first alternative image sequence, then this myocardial perfusion image sequence can be identified as the first image sequence. For another example, one or more blood flow characteristic parameters identify a certain myocardial perfusion image sequence as the first alternative image sequence, the sum of the characteristic weight coefficients of these blood flow characteristic parameters can be determined. If the sum of the characteristic weight coefficients corresponding to the xth myocardial perfusion image sequence is the largest, then this myocardial perfusion image sequence can be identified as the first image sequence.

[0126] In some embodiments of the present specification, the first image sequence and the second image sequence can be determined by synthesizing the multiple alternative first image sequences and the alternative second image sequences corresponding to the multiple blood flow characteristic parameters, thereby improving the accuracy of the sequence recognition result. At the same time, by setting different characteristic weight coefficients for different blood flow characteristic parameters, the recognition result can be more in line with the actual situation (such as the actual situation of the patient).

[0127] Figure 6 is another example schematic diagram of a method for identifying a first image sequence and a second image sequence according to some embodiments of the present specification.

[0128] In some embodiments, the recognition system 200 can use a recognition model to process the first characteristic values of each myocardial perfusion image sequence to determine the first image sequence and the second image sequence. The recognition model is a machine learning model.

[0129] The recognition model can refer to a model for identifying the first image sequence and the second image sequence. In some embodiments, the recognition model can be a trained machine learning model, for example, it can be a convolutional neural network (CNN) model, or other custom deep learning network model.

[0130] As shown in Figure 6 , the input of the recognition model 630 can include the first characteristic values 620 corresponding to the myocardial perfusion image sequence 610, and the output of the recognition model 630 can include the recognition result of the first image sequence and the second image sequence. For example only, the myocardial perfusion image sequence 610 can include m myocardial perfusion image sequences, each myocardial perfusion sequence has n first characteristic values corresponding to blood flow characteristic parameters, and the recognition result can include the sequence number corresponding to the first image sequence and the second image sequence.

[0131] In some embodiments, the identification model 630 can be obtained by training. The training samples can include first feature values corresponding to a plurality of sample myocardial perfusion image sequences and gold standard identification results. The gold standard identification results can include the numbers of image sequences belonging to the stress sequences and the rest sequences in the plurality of sample myocardial perfusion image sequences. The gold standard identification results can be determined artificially and used as labels for model training.

[0132] In some embodiments of the present specification, the first image sequence and the second image sequence are determined by the identification model based on the plurality of feature values of the plurality of perfusion image sequences. In the identification process, the identification model can mine the deep relationship between different perfusion image sequences and different feature values, so as to make the identification result more accurate. In addition, the characteristics of the stress image sequence and the rest image sequence (such as the change rule of the characteristics such as blood vessels and heart shape in an image sequence, the difference between the feature values of different sequence images, etc.) are learned by the artificial intelligence algorithm, so as to achieve the automatic identification of the stress image sequence and the rest image sequence, and reduce the loss of energy and time caused by manual identification.

[0133] Figure 7 is an exemplary flowchart of a process for determining a myocardial perfusion parameter according to some embodiments of the present specification.

[0134] In some embodiments, the identification system 200 can determine a myocardial perfusion parameter based on the first image sequence and the second image sequence. For related content of the first image sequence and the second image sequence, refer to Figure 3 and the description thereof.

[0135] The process 700 can be performed by the computing module 240. As Figure 7 indicated, the process 700 includes the following steps.

[0136] Step 710, identifying a first region and a second region corresponding to a region of interest from the first image sequence and the second image sequence.

[0137] The region of interest can refer to a preset part or tissue that needs to be labeled and analyzed. For example, the region of interest can be the left ventricular region of the heart.

[0138] The first region can refer to a region corresponding to the region of interest in the first image sequence. In some embodiments, the first region can be a region in a first representative image of the first image sequence. The first representative image can be a reference image in a plurality of images of the first image sequence.

[0139] The second region can refer to a region corresponding to the part of interest in the second image sequence. In some embodiments, the second region can be a region in a second representative image of the second image sequence. The second representative image can be a reference image in the plurality of images of the second image sequence. In some embodiments, the first representative image and the second representative image can be images with the same sequence number in the two image sequences. In some embodiments, the first representative image and the second representative image can correspond to the same heartbeat phase. For example, a reference phase can be selected from the plurality of heartbeat phases, and the first representative image and the second representative image both correspond to the reference phase.

[0140] At step 720, the first region and the second region are registered.

[0141] The registration can be based on any registration algorithm. The registration can be used to determine the correspondence and the position transformation between points in the first region and the second region. For example, the registration can be used to determine a pair of elements (pixel points or voxel points) in the first region and the second region corresponding to the same position on the same heart vessel, and the coordinate transformation between the pair of elements.

[0142] At step 730, a myocardial perfusion parameter is determined based on the values of the at least one blood flow feature parameter of the registered first region and the registered second region.

[0143] The myocardial perfusion parameter can refer to a parameter that can be used to evaluate the state of the myocardium.

[0144] The myocardial perfusion parameter can be a myocardial perfusion parameter under a stress state of the target object, or a myocardial perfusion parameter under a resting state of the target object. For example, the myocardial perfusion parameter can include myocardial blood flow of the target object under the stress state, or can include myocardial blood flow of the target object under the resting state.

[0145] In some embodiments, the myocardial perfusion parameter can reflect the difference between the myocardial state of the target object under the stress state and the resting state. For example, the myocardial perfusion parameter can include a myocardial perfusion reserve index (MPRi). The MPRi can represent the ratio of the myocardial blood flow of the target object under the stress state to the myocardial blood flow of the target object under the resting state.

[0146] In some embodiments, the identification system 200 can determine the myocardial perfusion parameter based on the values of the at least one blood flow feature parameter of the registered first region and the registered second region.

[0147] For example, the identification system 200 can determine the myocardial blood flow of the plurality of first elements in the registered first region, which can be referred to as the stress myocardial blood flow, and determine the myocardial blood flow of the plurality of second elements in the registered second region, which can be referred to as the rest myocardial blood flow. The identification system 200 can obtain the first element and the second element corresponding to the same position point of the region of interest according to the registration result of the first region and the second region, and determine the ratio of the stress myocardial blood flow of the first element to the rest myocardial blood flow of the second element to determine the myocardial perfusion reserve index of the position point.

[0148] Accordingly, the myocardial perfusion reserve indices corresponding to the plurality of position points of the region of interest can be determined. The identification system 200 can further obtain the myocardial perfusion reserve index information of the target object at the region of interest, which can be a numerical sequence of the myocardial perfusion reserve index or an average value.

[0149] It should be noted that the calculation process of the myocardial perfusion reserve index herein is only an example, and other myocardial perfusion parameters can also be used to comprehensively evaluate the myocardial state of the target object under stress and rest conditions.

[0150] In some embodiments of the present specification, after the first image sequence and the second image sequence are automatically identified, the regions corresponding to the region of interest can be identified from the first image sequence and the second image sequence, respectively, and the myocardial perfusion parameter (MPRi) can be automatically calculated. Since the myocardial perfusion parameter is determined based on the first image sequence and the second image sequence at the same time, it can comprehensively evaluate the myocardial state of the target object, making the evaluation result more accurate and comprehensive.

[0151] It should be noted that the above description of the process is only for example and illustration, and does not limit the scope of the present specification. Those skilled in the art can make various modifications and changes to the process under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification.

[0152] One of the embodiments of the present specification provides a myocardial perfusion image sequence identification device, which comprises a processor configured to perform the myocardial perfusion image sequence identification method described above.

[0153] The above detailed disclosure is only an example, and obviously, the above detailed disclosure does not limit the present specification. Although it is not explicitly stated herein, those skilled in the art can make various modifications, improvements and modifications to the present specification. Such modifications, improvements and modifications are suggested in the present specification, so such modifications, improvements and modifications are still within the spirit and scope of the exemplary embodiments of the present specification.

[0154] Also, the use of "a" or "an" or "the" are intended to include "at least one" and "one or more", unless otherwise used. Also, the use of "comprise", "comprises", "comprising", "contain", "contains", "containing", "include", "includes", "including" are intended to be open and permissive and not exclusive or exhaustive. As used in the description of the embodiments of the application, the following terms shall have the following meanings.

[0155] Finally, it is to be understood that the embodiments described herein are only illustrative of the principles of the present application. Further modifications that do not depart from the scope of the present application will be apparent to those skilled in the art. For example, while the present application has been described in terms of particular embodiments, it is not intended that the scope of the present application be limited to such. On the contrary, it is intended to cover all alternatives, modifications and equivalents falling within the scope of the present application. Accordingly, the application is not to be limited by what has been particularly shown and described.

Claims

1. A method of myocardial perfusion image sequence recognition, characterized in that, The method comprises: obtaining a plurality of myocardial perfusion image sequences of a target object; for each of the myocardial perfusion image sequences, determining a first feature value corresponding to the myocardial perfusion image sequence, the first feature value being related to one or more blood flow characteristic parameters, the one or more blood flow characteristic parameters including at least one of myocardial blood flow and myocardial blood volume; and based on the first feature value of each of the myocardial perfusion image sequences, identifying a first image sequence and a second image sequence from the plurality of myocardial perfusion image sequences, the first image sequence being acquired under a stress state of the target object, and the second image sequence being acquired under a resting state of the target object.

2. The method of claim 1, wherein, The one or more blood flow characteristic parameters include a plurality of blood flow characteristic parameters arranged in sequence, and the identifying the first image sequence and the second image sequence from the plurality of myocardial perfusion image sequences comprises: processing the plurality of blood flow characteristic parameters in sequence as follows: determining whether the first feature value corresponding to the blood flow characteristic parameter satisfies a preset condition; in response to the first feature value corresponding to the blood flow characteristic parameter satisfying the preset condition, identifying the first image sequence and the second image sequence based on the first feature value of the blood flow characteristic parameter; in response to the first feature value corresponding to the blood flow characteristic parameter not satisfying the preset condition, processing a next blood flow characteristic parameter.

3. The method of claim 2, wherein, The identifying the first image sequence and the second image sequence from the plurality of myocardial perfusion image sequences further comprises: in response to the first feature values of the plurality of blood flow characteristic parameters all not satisfying the preset condition, obtaining image metadata corresponding to each of the myocardial perfusion image sequences; based on the image metadata corresponding to each of the myocardial perfusion image sequences, determining a second feature value of the target object, the second feature value being related to heart rate; and based on the second feature value corresponding to each of the myocardial perfusion image sequences, determining the first image sequence and the second image sequence.

4. The method of claim 1, wherein, The one or more blood flow characteristic parameters include a plurality of blood flow characteristic parameters, and the identifying the first image sequence and the second image sequence from the plurality of myocardial perfusion image sequences comprises: for each of the plurality of blood flow characteristic parameters, identifying a candidate first image sequence and a candidate second image sequence based on the first feature value of the blood flow characteristic parameter; based on the candidate first image sequence and the candidate second image sequence corresponding to each of the blood flow characteristic parameters, determining the first image sequence and the second image sequence.

5. The method of claim 1, wherein, The identifying the first image sequence and the second image sequence from the plurality of myocardial perfusion image sequences based on the first feature value of each of the myocardial perfusion image sequences further comprises: processing the first feature value of each of the myocardial perfusion image sequences by using an identification model to determine the first image sequence and the second image sequence, the identification model being a machine learning model.

6. The method of claim 1, wherein, The method further comprises: based on the first image sequence and the second image sequence, determining a myocardial perfusion parameter.

7. The method of claim 1, wherein, The obtaining a plurality of myocardial perfusion image sequences of a target object comprises: obtaining a plurality of initial scan sequences of the target object; and The plurality of initial scan sequences are identified to determine the plurality of myocardial perfusion image sequences and delayed enhancement image sequences.

8. The method of claim 1, wherein, The method further comprises: based on the identification result of the myocardial perfusion image sequences, determining a sequence label, the sequence label comprising a stress label corresponding to the first image sequence and a rest label corresponding to the second image sequence; presenting the first image sequence and the stress label, and the second image sequence and the rest label on a user interface.

9. The method of claim 1, wherein, The identification of the first image sequence and the second image sequence from the plurality of myocardial perfusion image sequences comprises: determining a feature weight coefficient of each of the one or more blood flow feature parameters; based on the feature weight coefficient, determining the first image sequence and the second image sequence.

10. A myocardial perfusion image sequence recognition system, characterized by comprise: an acquisition module configured to acquire a plurality of myocardial perfusion image sequences of a target object; a determination module configured to determine, for each of the myocardial perfusion image sequences, a first feature value corresponding to the myocardial perfusion image sequence, the first feature value being related to one or more blood flow feature parameters, the one or more blood flow feature parameters comprising at least one of myocardial blood flow and myocardial blood volume; an identification module configured to identify, based on the first feature value of each of the myocardial perfusion image sequences, a first image sequence and a second image sequence from the plurality of myocardial perfusion image sequences, the first image sequence being acquired under a stress state of the target object, and the second image sequence being acquired under a rest state of the target object.

Images