Cerebrovascular reactivity determination method, device, equipment and medium

Abstract

The invention relates to the technical field of medical image processing, in particular to a cerebrovascular reactivity determination method, device and equipment and a medium, and the method comprises the steps: obtaining a cerebrovascular reactivity index image of a target brain tissue, the cerebrovascular reactivity index image comprises a cerebrovascular reactivity relative value of the target brain tissue in a resting state; acquiring a cerebral blood flow image and an artery passing time image of the target brain tissue; the cerebrovascular reactivity index image, the cerebral blood flow image and the artery passing time image are input into a pre-trained neural network model to be processed, a cerebrovascular reactivity image of the target brain tissue is obtained, and the cerebrovascular reactivity image comprises a cerebrovascular reactivity absolute value of the target brain tissue. According to the method disclosed by the invention, the CVR absolute value can be determined without using external equipment related to respiration control and without additional cooperation of a patient, so that the applicability of the CVR determination method to encephalopathy and the patient is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a method, device, equipment and medium for determining cerebrovascular reactivity. Background Art

[0002] Cerebrovascular reactivity (CVR) reflects the ability of blood vessels to dilate to adapt to the increased demand for tissue blood supply, and can be studied by measuring changes in cerebral blood flow (CBF) or cerebral blood volume (CBV). CVR is closely related to vascular health and is an important parameter for studying diseases such as stroke, moyamoya disease, and cerebral small vessel disease.

[0003] In order to measure CVR, artificial stimulation is generally required to induce vasodilation, including inhalation of carbon dioxide-rich gas, breath holding, hyperventilation, injection of acetazolamide (ACZ), etc. Since the partial pressure of carbon dioxide (PaCO2), which is directly related to the degree of vasodilation, is difficult to measure, the end tidal carbon dioxide tension (EtCO2) is usually used as a substitute to characterize the degree of vasodilation during the measurement of CVR. In current CVR-related research, the most commonly used method of inducing vasodilation is manual or computer-controlled gas delivery components to achieve the target EtCO2.

[0004] Imaging methods that can be used to measure CVR include positron emission computed tomography (PET), single-photon emission computed tomography (SPECT), computed tomography (CT), transcranial Doppler ultrasound and magnetic resonance imaging (MRI), among which the latter two do not have ionizing radiation. Transcranial Doppler ultrasound has a small field of view and can only measure local blood flow velocity, so its application is relatively limited. MRI can measure CBF and CBV through arterial spin labeling (ASL) perfusion imaging technology, and can also measure BOLD signals indirectly related to CBF and CBV through blood oxygen level dependent (BOLD) imaging technology.

[0005] Currently, the signal changes caused by vasodilator stimulation can be converted into quantitative or semiquantitative measurements reflecting CVR by the following methods: subtract the signals before and after vasodilator stimulation to obtain the CBF or BOLD signal changes caused by vasodilation, use EtCO2 as the independent variable and CBF or BOLD signal as the dependent variable to model, and use algorithms such as linear regression, cross-correlation, and nonlinear fitting to estimate the relevant parameters and then determine CVR.

[0006] However, these methods require artificial stimulation to induce vasodilation, require the use of external equipment related to breathing control and the cooperation of the patient, and the patient's comfort during the measurement process is poor. Summary of the invention

[0007] In view of the above-mentioned problems in the prior art, an object of the present invention is to provide a method, device, equipment and medium for determining cerebrovascular reactivity, which can improve the applicability of the CVR determination method for brain diseases and patients.

[0008] In order to solve the above problems, the present invention provides a method for determining cerebrovascular reactivity, comprising: Acquiring a cerebrovascular reactivity index image of a target brain tissue, wherein the cerebrovascular reactivity index image includes a relative value of cerebrovascular reactivity of the target brain tissue in a resting state; Acquiring cerebral blood flow images and arterial transit time images of target brain tissue; The cerebrovascular reactivity index image, cerebral blood flow image and arterial transit time image are input into a pre-trained neural network model for processing to obtain a cerebrovascular reactivity image of the target brain tissue, which includes the absolute value of the cerebrovascular reactivity of the target brain tissue.

[0009] Furthermore, obtaining a cerebrovascular reactivity index image of the target brain tissue includes: Acquire blood oxygen level-dependent imaging data of target brain tissue in a resting state; registering the blood oxygen level dependent imaging data to the standard space to obtain the blood oxygen level dependent imaging data in the standard space; A cerebrovascular reactivity index image of a target brain tissue is determined based on the blood oxygen level-dependent imaging data in a standard space.

[0010] Furthermore, the method further comprises: Preprocessing the blood oxygen level dependent imaging data to obtain preprocessed blood oxygen level dependent imaging data; wherein the preprocessing includes one or more of motion correction, linear component removal, spatial smoothing and bandpass filtering; Among them, the blood oxygen level dependent imaging data is registered to the standard space, including: The preprocessed blood oxygen level dependent imaging data is registered to the standard space to obtain the blood oxygen level dependent imaging data in the standard space.

[0011] Furthermore, a cerebral blood flow image and an arterial transit time image of the target brain tissue are obtained, including: Acquiring arterial spin labeling perfusion imaging data of target brain tissue at rest with multiple post-labeling delays; registering a plurality of delayed arterial spin labeling perfusion imaging data after labeling to a standard space respectively, to obtain a plurality of delayed arterial spin labeling perfusion imaging data after labeling in the standard space; determining an arterial transit time image of a target brain tissue based on arterial spin labeling perfusion imaging data at multiple post-labeling delays in standard space; A cerebral blood flow image of the target brain tissue is determined based on the arterial transit time image.

[0012] Further, determining the cerebral blood flow image of the target brain tissue according to the arterial transit time image includes: Calculating cerebral blood flow images corresponding to each delayed arterial spin labeling perfusion imaging data after labeling, respectively, according to the arterial transit time image and each delayed arterial spin labeling perfusion imaging data after labeling; The average image of the cerebral blood flow images corresponding to the arterial spin labeling perfusion imaging data after each labeling delay is calculated as the cerebral blood flow image of the target brain tissue.

[0013] Furthermore, the method further comprises: Preprocessing a plurality of delayed arterial spin labeling perfusion imaging data after labeling to obtain a plurality of preprocessed delayed arterial spin labeling perfusion imaging data after labeling; wherein the preprocessing includes motion correction processing; Among them, multiple delayed arterial spin labeling perfusion imaging data after labeling are registered to the standard space respectively, including: The preprocessed multiple delayed arterial spin labeling perfusion imaging data are respectively registered to the standard space to obtain multiple delayed arterial spin labeling perfusion imaging data of the standard space.

[0014] Furthermore, the method also includes pre-training the neural network model, and the training process of the neural network model includes: Acquire a training sample data set, the training sample data set comprising first blood oxygen level-dependent imaging sample data, second blood oxygen level-dependent imaging sample data, third blood oxygen level-dependent imaging sample data and multiple labeled delayed arterial spin labeling perfusion imaging sample data of brain tissues of multiple patients with cerebrovascular reactivity-related encephalopathy; wherein the first blood oxygen level-dependent imaging sample data, the second blood oxygen level-dependent imaging sample data and the third blood oxygen level-dependent imaging sample data are blood oxygen level-dependent imaging sample data of the corresponding brain tissues under vasodilation stimulation, under no vasodilation stimulation and in a resting state, respectively; determining a corresponding cerebrovascular reactivity training image according to the first blood oxygen level dependent imaging sample data and the second blood oxygen level dependent imaging sample data of each brain tissue; Determine a corresponding cerebrovascular reactivity training index image according to the third blood oxygen level dependent imaging sample data of each brain tissue; Determine corresponding cerebral blood flow training images and arterial transit time training images according to a plurality of labeled delayed arterial spin labeling perfusion imaging sample data of each brain tissue; The cerebrovascular reactivity training index images, cerebral blood flow training images and arterial transit time training images corresponding to each brain tissue are used as input, and the corresponding cerebrovascular reactivity training images are used as supervision to perform supervised training on the preset neural network model to obtain a trained neural network model.

[0015] Another aspect of the present invention provides a device for determining cerebrovascular reactivity, comprising: A first acquisition module is used to acquire a cerebrovascular reactivity index image of a target brain tissue, wherein the cerebrovascular reactivity index image includes a relative value of cerebrovascular reactivity of the target brain tissue in a resting state; A second acquisition module is used to acquire a cerebral blood flow image and an arterial transit time image of a target brain tissue; The determination module is used to input the cerebrovascular reactivity index image, the cerebral blood flow image and the arterial transit time image into a pre-trained neural network model for processing to obtain a cerebrovascular reactivity image of the target brain tissue, wherein the cerebrovascular reactivity image includes an absolute value of the cerebrovascular reactivity of the target brain tissue.

[0016] Another aspect of the present invention provides an electronic device, including a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining cerebrovascular reactivity as described above.

[0017] Another aspect of the present invention provides a computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the method for determining cerebrovascular reactivity as described above.

[0018] Another aspect of the present invention provides a computer program product, comprising a computer program / instruction, which implements the above-mentioned method for determining cerebrovascular reactivity when executed by a processor.

[0019] Due to the above technical solution, the present invention has the following beneficial effects: According to the method for determining cerebrovascular reactivity of an embodiment of the present invention, the relative value of CVR is calculated by using BOLD imaging data at rest, the absolute values ​​of CBF and ATT are calculated by using ASL perfusion imaging data of multiple post-labeling delays (Multi-PLD) at rest, and the absolute value of CVR is calculated by using a neural network model. This method can be applied to various encephalopathy related to CVR, not just limited to the situation of local CVR decline. Moreover, the data required by this method can be collected completely at rest, without the need to use external equipment related to breathing control, and without the need for additional cooperation from the patient. Therefore, it can be applied to pediatric patients, comatose patients, etc., and has a wider range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments or prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 It is a schematic diagram of an implementation environment provided by an embodiment of the present invention; Figure 2 is a flow chart of a method for determining cerebrovascular reactivity provided by one embodiment of the present invention; Figure 3 is a flow chart of a model training process provided by an embodiment of the present invention; Figure 4 is a schematic diagram of a CVR training image provided by an embodiment of the present invention; Figure 5 is a schematic diagram of a CVR training index image provided by an embodiment of the present invention; Figure 6 is a schematic structural diagram of a device for determining cerebrovascular reactivity provided by an embodiment of the present invention; Figure 7 is a block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0024] The method provided by the embodiment of the present invention can be applied to the scenario of measuring the CVR of patients who may suffer from stroke, moyamoya disease, cerebral small vessel disease and other diseases. Studies in recent years have shown that the BOLD signal in the resting state can also reflect the CVR. This is because the carbon dioxide content in the blood will also change with natural breathing, which will cause corresponding fluctuations in CBF, CBV and BOLD signals. By measuring this fluctuation, the CVR can be obtained. The advantage of this method is that it does not require any artificial stimulation, so there is no need to use external equipment related to breathing control, and at the same time increases the comfort of the patient. However, the CVR obtained by this method is a relative value, not an absolute value in the conventional sense (ie, percentage of signal change / mmHg), which makes this method only applicable to situations where the local CVR decreases.

[0025] ASL perfusion imaging is also a common imaging method for measuring CVR. ASL perfusion imaging scans can be performed before and after vasodilation stimulation to calculate the percentage difference in CBF. The advantage of ASL perfusion imaging is that it can obtain the absolute value of CBF without invasion and radiation, and the ASL perfusion imaging data with multiple post-labeling delays (Multi-PLD) can also calculate the arterial transit time (ATT) parameter, which is very important in vascular occlusion / stenosis diseases (including stroke, moyamoya disease, etc.), and can also obtain more accurate CBF.

[0026] The embodiment of the present invention proposes a neural network model that can obtain CVR in units of signal change percentage / mmHg based on BOLD imaging data and Multi-PLD ASL perfusion imaging data at rest. The neural network model can be trained based on BOLD imaging data and Multi-PLD ASL perfusion imaging data at rest, using CVR obtained from BOLD imaging data before and after vasodilation stimulation as supervision.

[0027] In order to make the purpose, technical solution and advantages disclosed in the embodiments of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not used to limit the embodiments of the present invention.

[0028] Reference Manual Attached Figure 1 , which shows a schematic diagram of an implementation environment provided by an embodiment of the present invention. Figure 1 As shown, the implementation environment may include at least one medical scanning device 110 and a computer device 120. The computer device 120 and each medical scanning device 110 may be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment of the present invention.

[0029] The medical scanning device 110 may be, but is not limited to, a magnetic resonance imaging device, etc. The computer device 120 may be, but is not limited to, various servers, personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be an independent server or a server cluster or distributed system composed of multiple servers. It may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0030] The computer device 120 can obtain medical imaging data of the target brain tissue in a resting state scanned by the medical scanning device 110 (such as blood oxygen level-dependent imaging data and arterial spin labeling perfusion imaging data after multiple labeling delays), and determine the absolute value of the cerebrovascular reactivity of the target brain tissue through the cerebrovascular reactivity determination method provided in an embodiment of the present invention.

[0031] It should be noted that Figure 1 It is just an example. Those skilled in the art will understand that although Figure 1Only one medical scanning device 110 is shown, but this does not constitute a limitation on the embodiment of the present invention, and more or less medical scanning devices 110 than shown in the figure may be included.

[0032] Reference Manual Attached Figure 2 , which shows the process of a method for determining cerebrovascular reactivity provided by an embodiment of the present invention, which can be applied to Figure 1 In the computer device 120. Specifically, Figure 2 As shown, the method may include the following steps: S210: Acquire a cerebrovascular reactivity index image of the target brain tissue, wherein the cerebrovascular reactivity index image includes a relative value of cerebrovascular reactivity of the target brain tissue in a resting state.

[0033] In an embodiment of the present invention, the image of the target brain tissue in a resting state can be collected by BOLD imaging technology to obtain BOLD imaging data in a resting state, and then the CVR index image can be calculated using the BOLD imaging data. Among them, the CVR index image can include the CVR relative value corresponding to each voxel in the target brain tissue, that is, the dimensionless CVR value. The target brain tissue may be a patient who may suffer from CVR-related encephalopathy (such as stroke, moyamoya disease, cerebral small vessel disease, dementia, etc.).

[0034] It should be noted that the source of the BOLD imaging data in the resting state can be directly imported related data, or obtained from other resource libraries through real-time configuration connection, or obtained from a stored image database after searching according to the patient's name and other information. The embodiment of the present invention does not limit this.

[0035] In some embodiments, obtaining a cerebrovascular reactivity index image of a target brain tissue may include: obtaining blood oxygen level-dependent imaging data of the target brain tissue in a resting state; registering the blood oxygen level-dependent imaging data to a standard space to obtain blood oxygen level-dependent imaging data of the standard space; and determining a cerebrovascular reactivity index image of the target brain tissue based on the blood oxygen level-dependent imaging data of the standard space.

[0036] Specifically, after the BOLD imaging data is acquired, it can be registered to the standard space to obtain the BOLD imaging data in the standard space to simplify the subsequent processing process.

[0037] It should be noted that the BOLD imaging data can be registered to the standard space using the registration method in the prior art. For example, the BOLD imaging data can be directly registered to the standard space, or the T1 structural image of the target brain tissue can be first scanned, and the BOLD imaging data can be registered to the T1 structural image, and then registered to the standard space. The embodiment of the present invention is not limited to this.

[0038] Specifically, BOLD imaging data can include the BOLD signal corresponding to each voxel in the target brain tissue. Since the lesions of most brain diseases are in the brain, in order to avoid the influence, the spatial mean of the BOLD signal in the cerebellum can be used (recorded as the reference signal BOLD ref ) is the independent variable, the motion parameter is the covariate, and the BOLD signal is the dependent variable. The generalized linear model of the following formula (1) is used to calculate the coefficient of the BOLD signal changing with the reference signal to obtain the dimensionless CVR value:

[0039] in, Motion represents motion parameters, a, b, and c are coefficients, and coefficient a is the dimensionless CVR value.

[0040] It should be noted that the motion parameter may be a parameter extracted from the BOLD imaging data for measuring the motion of the patient during the scanning process. The motion of the patient during the scanning process may include but is not limited to head movement, movement caused by breathing, and pulsation, etc., which is not limited in the embodiment of the present invention. The specific process of using a generalized linear model to calculate the coefficient of the change of the BOLD signal with the reference signal can refer to the existing model construction method, and the embodiment of the present invention will not be repeated here.

[0041] It should be noted that the above method of determining the CVR index image using BOLD imaging data can be executed by a computer device that implements the method provided in the embodiment of the present invention, or can be executed by other devices, and the obtained CVR index image is sent to the computer device, and the embodiment of the present invention is not limited to this.

[0042] In some embodiments, the method may further include: preprocessing the blood oxygen level dependent imaging data to obtain preprocessed blood oxygen level dependent imaging data; wherein the preprocessing includes one or more of motion correction, removal of linear components, spatial smoothing and bandpass filtering.

[0043] Specifically, before registering the BOLD imaging data to the standard space, the BOLD imaging data may be preprocessed. The preprocessing of the BOLD imaging data may include, but is not limited to, one or more of motion correction, linear component removal, spatial smoothing, and bandpass filtering. The specific processing method may refer to the prior art, and the embodiments of the present invention will not be described in detail herein.

[0044] Accordingly, registering the blood oxygen level dependent imaging data to the standard space may include: registering the preprocessed blood oxygen level dependent imaging data to the standard space to obtain blood oxygen level dependent imaging data in the standard space. The specific registration method may refer to the relevant content in step S110, and the embodiments of the present invention will not be described in detail here.

[0045] It can be understood that by preprocessing the BOLD imaging data, the problem of image intensity distribution being non-uniform due to different inherent settings of different medical scanning devices and different preprocessing parameters can be avoided.

[0046] S220: Acquire a cerebral blood flow image and an arterial transit time image of the target brain tissue.

[0047] In an embodiment of the present invention, the ASL perfusion imaging technology can be used to collect images of target brain tissues in a resting state, and Multi-PLD ASL perfusion imaging data in a resting state can be obtained, and then the CBF image and ATT image can be calculated using the Multi-PLD ASL perfusion imaging data. The target brain tissue may be a patient who may suffer from CVR-related encephalopathy (such as stroke, moyamoya disease, cerebral small vessel disease, etc.).

[0048] It should be noted that the source of the Multi-PLD ASL perfusion imaging data in the resting state can be directly imported related data, or obtained from other resource libraries through real-time configuration connection, or obtained from the stored image database after searching according to the patient's name and other information. The embodiment of the present invention does not limit this.

[0049] In some embodiments, obtaining a cerebral blood flow image and an arterial transit time image of a target brain tissue includes: obtaining a plurality of labeling-delayed arterial spin labeling perfusion imaging data of the target brain tissue in a resting state; respectively registering the plurality of labeling-delayed arterial spin labeling perfusion imaging data to a standard space to obtain a plurality of labeling-delayed arterial spin labeling perfusion imaging data of the standard space; determining an arterial transit time image of the target brain tissue based on the plurality of labeling-delayed arterial spin labeling perfusion imaging data of the standard space; and determining a cerebral blood flow image of the target brain tissue based on the arterial transit time image.

[0050] Specifically, after the Multi-PLD ASL perfusion imaging data is acquired, it can be registered to the standard space to obtain the Multi-PLD ASL perfusion imaging data of the standard space, so as to simplify the subsequent processing process.

[0051] It should be noted that the Multi-PLD ASL perfusion imaging data can be registered to the standard space using the registration method in the prior art. For example, the Multi-PLD ASL perfusion imaging data can be directly registered to the standard space, or the T1 structural image of the target brain tissue can be first scanned, and the Multi-PLD ASL perfusion imaging data can be registered to the T1 structural image, and then registered to the standard space. The embodiment of the present invention is not limited to this.

[0052] Specifically, ASL perfusion imaging data may include proton density (PD) weighted images, control images (i.e., non-labeled images), and labeled images (label images). Subtracting the control image corresponding to each PLD from the labeled image can obtain the ASL image corresponding to each PLD. The weighted delay (WD) image can be calculated using the ASL image corresponding to each PLD, and then the ATT image can be determined based on the WD image. Specifically, the process of calculating the WD image can refer to the following formula:

[0053] Where N is the number of PLDs, w(i) is the value of PLD, and ΔS(i) is the signal difference between the control image and the marker image (i.e., the signal value of the ASL image).

[0054] Specifically, after the WD image is obtained, the ATT value corresponding to the WD value of each voxel in the WD image can be determined based on a pre-established corresponding curve between the ATT value and the WD value, thereby obtaining an ATT image.

[0055] The corresponding curve between the ATT value and the WD value is a monotonically increasing curve, that is, the larger the WD value is, the larger the corresponding ATT value is. In some possible embodiments, the corresponding curve between the ATT value and the WD value can be obtained in the following manner: first, the value range of the ATT value is preset, and according to the preset ASL signal model, the WD values ​​corresponding to all ATT values ​​within the value range are calculated according to a certain step length, and the corresponding curve between the ATT value and the WD value can be obtained. The value range can be set according to actual needs, and the embodiment of the present invention does not limit this; the preset ASL signal model can be composed of formula (2) and the following formula (3):

[0056] Where w(i) is the value of PLD, ΔS(i) is the signal difference between the control image and the labeled image (i.e., the signal value of the ASL image), λ is the brain-blood distribution coefficient, which is usually simplified to be 0.9 mL / g, S0 is the voxel signal value of the proton density weighted image, a is the labeling efficiency, CBF(i) is the CBF value corresponding to each PLD, and R 1,a is the T1 relaxation rate of arterial blood, that is, 1 / T 1,a , τ is the duration of the marking sequence.

[0057] Specifically, determining the cerebral blood flow image of the target brain tissue based on the arterial transit time image can include: calculating each cerebral blood flow image corresponding to each delayed arterial spin labeling perfusion imaging data after labeling based on the arterial transit time image and each delayed arterial spin labeling perfusion imaging data after labeling; and calculating an average image of the cerebral blood flow images corresponding to each delayed arterial spin labeling perfusion imaging data after labeling as the cerebral blood flow image of the target brain tissue.

[0058] Specifically, after determining the ATT image, the CBF value of each voxel in the CBF image corresponding to each PLD can be calculated using formula (3) based on the ATT image and the ASL image corresponding to each PLD, thereby obtaining the CBF image corresponding to each PLD. Then, the average image of the CBF images corresponding to each PLD is calculated as the CBF image of the target brain tissue.

[0059] The average image may be a weighted average image, and the weights of the CBF images corresponding to the respective PLDs may be preset, for example, may be set to 1 / N (N is the number of PLDs), which is not limited in the embodiment of the present invention. Exemplarily, when the weights of the CBF images corresponding to the respective PLDs are set to 1 / N, the average image of the CBF images corresponding to the respective PLDs may be calculated by the following formula:

[0060] It should be noted that the above method of calculating CBF images and ATT images using ASL perfusion imaging data can be executed by a computer device that implements the method provided in the embodiment of the present invention, or can be executed by other devices, and the obtained CBF images and ATT images are sent to the computer device, and the embodiment of the present invention does not limit this.

[0061] In some embodiments, the method may further include: preprocessing a plurality of delayed arterial spin labeling perfusion imaging data after labeling to obtain a plurality of preprocessed delayed arterial spin labeling perfusion imaging data after labeling; wherein the preprocessing includes motion correction processing.

[0062] Specifically, before registering the Multi-PLD ASL perfusion imaging data to the standard space, the Multi-PLD ASL perfusion imaging data may be preprocessed. Preprocessing the Multi-PLD ASL perfusion imaging data may include but is not limited to motion correction and other processing. The specific processing method can refer to the prior art, and the embodiments of the present invention will not be repeated here.

[0063] Accordingly, registering a plurality of delayed arterial spin labeling perfusion imaging data after labeling to the standard space may include: registering a plurality of delayed arterial spin labeling perfusion imaging data after preprocessing to the standard space to obtain a plurality of delayed arterial spin labeling perfusion imaging data after labeling in the standard space. The specific registration method may refer to the relevant content in step S210, and the embodiment of the present invention will not be repeated here.

[0064] It can be understood that by preprocessing the Multi-PLD ASL perfusion imaging data, the problem of non-uniform image intensity distribution due to different inherent settings of different medical scanning devices and different preprocessing parameters can be avoided.

[0065] S230: Inputting the cerebrovascular reactivity index image, the cerebral blood flow image and the arterial transit time image into a pre-trained neural network model for processing to obtain a cerebrovascular reactivity image of the target brain tissue, wherein the cerebrovascular reactivity image includes an absolute value of the cerebrovascular reactivity of the target brain tissue.

[0066] In an embodiment of the present invention, the CVR index image, CBF image and ATT image can be combined to predict the CVR image of the target brain tissue using a machine learning algorithm. The CVR image can include the CVR absolute value corresponding to each voxel in the target brain tissue, that is, the CVR value in units of signal change percentage / mmHg.

[0067] Specifically, the neural network model can analyze and process the input CVR index image, CBF image and ATT image to determine the CVR absolute value corresponding to each voxel in the target brain tissue, thereby determining the CVR image of the target brain tissue.

[0068] In an embodiment of the present invention, the neural network model can be obtained by supervised training of a preset neural network model using a CVR training index image calculated by BOLD imaging sample data at rest, a CBF training image and an ATT training image calculated by Multi-PLD ASL perfusion imaging sample data as input, and a CVR training image calculated by BOLD imaging sample data with and without vasodilation stimulation as supervision. The preset neural network model may include but is not limited to machine learning models commonly used in the prior art, such as generative adversarial networks, conditional generative adversarial networks, and the like. The embodiment of the present invention does not impose specific restrictions on the structure of the model. The specific model training method will be described in detail later.

[0069] In summary, the method for determining cerebrovascular reactivity according to the embodiment of the present invention calculates the relative value of CVR by using the BOLD imaging data at rest, calculates the absolute values ​​of CBF and ATT by using the ASL perfusion imaging data of multiple post-labeling delays (Multi-PLD) at rest, and then calculates the absolute value of CVR by using the neural network model. This method can be applied to various encephalopathy related to CVR, not just limited to the situation of local CVR decline. Moreover, the data required by this method can be collected completely at rest, without the need to use external equipment related to breathing control, and does not require additional cooperation from the patient. Therefore, it can be applied to pediatric patients, comatose patients, etc., and has a wider range of applications.

[0070] The model training method in the embodiment of the present invention is described in detail below.

[0071] Reference Manual Attached Figure 3 , which shows the process of a training method for a neural network model provided by an embodiment of the present invention, specifically as follows Figure 3 As shown, the method may include the following steps: S310: Acquire a training sample data set, which includes first blood oxygen level dependent imaging sample data, second blood oxygen level dependent imaging sample data, third blood oxygen level dependent imaging sample data and multiple post-labeling delayed arterial spin labeling perfusion imaging sample data of brain tissues of multiple patients with cerebrovascular reactivity-related encephalopathy.

[0072] Among them, the first blood oxygen level dependent imaging sample data, the second blood oxygen level dependent imaging sample data and the third blood oxygen level dependent imaging sample data are the blood oxygen level dependent imaging sample data of the corresponding brain tissue under vasodilation stimulation, without vasodilation stimulation and in a resting state, respectively.

[0073] In an embodiment of the present invention, BOLD imaging sample data of brain tissue of patients with multiple CVR-related encephalopathy (e.g., stroke, moyamoya disease, cerebral small vessel disease, etc.) under vasodilation stimulation, without vasodilation stimulation, and at rest, as well as Multi-PLD ASL perfusion imaging sample data at rest, can be collected to form a training sample data set. Each training sample data in the training sample data set can be an image sample with excellent scanning quality and no motion artifacts.

[0074] It should be noted that, in actual applications, image samples corresponding to one or more types of patients with CVR-related encephalopathy can be selected to form a training sample data set. When image samples corresponding to patients with a type of CVR-related encephalopathy are selected to form a training sample data set, the trained neural network model is only applicable to determining the cerebrovascular reactivity images of the brain tissue of patients with this type of encephalopathy. When image samples corresponding to patients with multiple types of CVR-related encephalopathy are selected to form a training sample data set, the trained neural network model can be applicable to determining the cerebrovascular reactivity images of the brain tissue of patients with these multiple types of encephalopathy. In other words, for different types of encephalopathy, the neural network model corresponding to each type can be trained separately, or a general neural network model can be trained, and the embodiments of the present invention do not limit this.

[0075] In some embodiments, BOLD imaging sample data of multiple normal brain tissues (i.e., brain tissues of healthy people) under vasodilation stimulation, without vasodilation stimulation, and at rest, as well as Multi-PLD ASL perfusion imaging sample data at rest can also be collected to form a training sample data set together with the training sample data of patients with CVR-related encephalopathy.

[0076] Exemplarily, BOLD imaging sample data of brain tissue of at least 150 healthy people and 150 patients with CVR-related encephalopathy under vasodilation stimulation, without vasodilation stimulation and at rest, as well as Multi-PLD ASL perfusion imaging sample data at rest can be collected to form a training sample data set.

[0077] Specifically, a device that can accurately control the target EtCO2 or a method of manually controlling EtCO2 can be used. The subjects are allowed to breathe air for the first 4 minutes, and the carbon dioxide concentration is adjusted to produce a 9 mmHg change in EtCO2 for the next 6 minutes. During this period, BOLD imaging scans are performed simultaneously. The data in the transition period (the 5th and 6th minutes) are discarded to obtain BOLD imaging sample data with / without vasodilation stimulation. Among them, the BOLD imaging sample data can be four-dimensional image data with a time resolution of 1 to 3 seconds, which records the BOLD signals of each voxel in the brain tissue corresponding to different moments in the whole process.

[0078] Specifically, although the above-mentioned first 4 minutes of breathing air is similar to the resting state, considering that the subjects will wear breathing masks, etc., which is slightly different from the natural breathing state, it is possible to collect natural BOLD imaging sample data in the resting state. During the collection process, the subjects do not wear any breathing equipment, breathe freely and remain awake, and perform BOLD imaging scans continuously for 5 minutes to obtain BOLD imaging sample data in the resting state.

[0079] Specifically, when the subjects are not wearing any respiratory equipment, breathing freely and remaining awake, a Multi-PLD ASL perfusion imaging scan is performed to obtain Multi-PLD ASL perfusion imaging sample data in a resting state.

[0080] It can be understood that by training a neural network model for determining CVR images based on sample data of normal brain tissue and patient brain tissue, the generalization ability of the model in the patient category can be improved to avoid overfitting.

[0081] S320: Determine a corresponding cerebrovascular reactivity training image according to the first blood oxygen level dependent imaging sample data and the second blood oxygen level dependent imaging sample data of each brain tissue.

[0082] In the embodiment of the present invention, the corresponding CVR training image can be calculated based on the BOLD imaging sample data with / without vasodilation stimulation.

[0083] Specifically, the BOLD imaging sample data with / without vasodilation stimulation can be first registered to the standard space to obtain the BOLD imaging sample data with / without vasodilation stimulation in the standard space to simplify the subsequent processing process. The specific registration method can refer to the relevant content of registering the BOLD imaging data to the standard space in step S210, and the embodiment of the present invention will not be repeated here.

[0084] Specifically, the CVR training image can be calculated by the following formula (5):

[0085] Among them, BOLD0 represents the mean BOLD signal without vasodilation stimulation, and BOLD S represents the mean BOLD signal with vasodilation stimulation, EtCO20 represents the mean EtCO2 without vasodilation stimulation, and EtCO2 S It represents the mean EtCO2 value when there is vasodilation stimulation, and the unit of CVR is % / mmHg.

[0086] For example, in conjunction with the reference specification Figure 4, which shows a schematic diagram of a CVR training image provided by an embodiment of the present invention. Figure 4 As shown, the CVR value in the CVR training image is an absolute value, and the unit is % / mmHg.

[0087] In some embodiments, before registering the BOLD imaging sample data with / without vasodilation stimulation to the standard space, the BOLD imaging sample data may be preprocessed. The preprocessing of the BOLD imaging data may include, but is not limited to, one or more of motion correction, removal of linear components, spatial smoothing, and bandpass filtering. The specific processing method can refer to the prior art, and the embodiments of the present invention will not be repeated here.

[0088] It can be understood that by preprocessing the BOLD imaging sample data in the training sample data set, the problem of non-uniform image intensity distribution due to different inherent settings of different medical scanning devices and different preprocessing parameters can be avoided.

[0089] S330: Determine a corresponding cerebrovascular reactivity training index image according to the third blood oxygen level-dependent imaging sample data of each brain tissue.

[0090] In the embodiment of the present invention, the corresponding CVR training index image can be calculated based on the BOLD imaging sample data in the resting state. The specific calculation method can refer to the relevant content of calculating the CVR index image in step S210, and the embodiment of the present invention will not be repeated here.

[0091] For example, in conjunction with the reference specification Figure 5 , which shows a schematic diagram of a CVR training index image provided by an embodiment of the present invention. Figure 5 As shown in FIG. 1 , the CVR value in the CVR training index image is a relative value, that is, a dimensionless CVR value.

[0092] S340: Determine corresponding cerebral blood flow training images and arterial transit time training images according to a plurality of labeled delayed arterial spin labeling perfusion imaging sample data of each brain tissue.

[0093] In the embodiment of the present invention, the corresponding CBF training image and ATT training image can be calculated based on the Multi-PLD ASL perfusion imaging sample data in the resting state. The specific calculation method can refer to the relevant content of calculating the CBF image and the ATT image in step S220, and the embodiment of the present invention will not be repeated here.

[0094] S350: Using the cerebrovascular reactivity training index images, cerebral blood flow training images and arterial transit time training images corresponding to each brain tissue as input and the corresponding cerebrovascular reactivity training images as supervision, a preset neural network model is subjected to supervised training to obtain a trained neural network model.

[0095] In the embodiment of the present invention, an initial neural network model can be pre-constructed, and the neural network model may include but is not limited to the commonly used machine learning models in the prior art, for example, it may include but is not limited to the generative adversarial network, conditional generative adversarial network and other models, and the embodiment of the present invention does not specifically limit the structure of the model. The construction process of the initial neural network model can refer to the prior art, and the embodiment of the present invention will not be repeated here.

[0096] In an embodiment of the present invention, a CVR training image obtained from BOLD imaging sample data with / without vasodilation stimulation can be used as supervision, and a CVR training index image obtained from BOLD imaging sample data at rest, and a CBF training image and an ATT training image obtained from Multi-PLD ASL perfusion imaging sample data at rest can be used as input data. The initial neural network model can be supervised trained using a machine learning algorithm to obtain a trained neural network model.

[0097] Specifically, during the model training process, each training sample data in the training sample data set can be cross-divided into a training set and a test set (for example, the training sample data can be divided into five groups, and any four groups are taken as training sets each time, and the remaining group is a test set), and the training sample data in the training set is used to train the neural network model, and the training sample data in the test set is used to test the performance of the trained neural network model, evaluate the performance of the model, and finally obtain a reliable model. The specific content of the model training process can refer to the prior art, and the embodiments of the present invention will not be repeated here.

[0098] In summary, the embodiment of the present invention uses the CVR training image obtained from the BOLD imaging sample data with / without vasodilation stimulation as supervision, and uses the CVR training index image obtained from the BOLD imaging sample data at rest, the CBF training image and the ATT training image obtained from the Multi-PLD ASL perfusion imaging sample data at rest as input, to perform machine learning model training to obtain a neural network model for determining the absolute value of CVR, which can improve the accuracy and reliability of the model. The trained neural network model can be applied to various brain diseases related to CVR, and is not limited to the situation of local CVR decline, and has a wider range of applications.

[0099] It should be noted that other contents in the embodiments of the present invention can be referred to Figure 2The relevant contents of the method for determining cerebrovascular reactivity provided by the illustrated embodiment will not be repeated in detail in the embodiment of the present invention.

[0100] Reference Manual Attached Figure 6 , which shows the structure of a cerebrovascular reactivity determination device 600 provided by one embodiment of the present invention. Figure 6 As shown, the apparatus 600 may include: A first acquisition module 610 is used to acquire a cerebrovascular reactivity index image of a target brain tissue, wherein the cerebrovascular reactivity index image includes a relative value of cerebrovascular reactivity of the target brain tissue in a resting state; A second acquisition module 620 is used to acquire a cerebral blood flow image and an arterial transit time image of a target brain tissue; The determination module 630 is used to input the cerebrovascular reactivity index image, the cerebral blood flow image and the arterial transit time image into a pre-trained neural network model for processing to obtain a cerebrovascular reactivity image of the target brain tissue, wherein the cerebrovascular reactivity image includes an absolute value of the cerebrovascular reactivity of the target brain tissue.

[0101] In a possible embodiment, the device 600 may further include a model training module, which may include: A training sample data set acquisition unit is used to acquire a training sample data set, wherein the training sample data set includes first blood oxygen level-dependent imaging sample data, second blood oxygen level-dependent imaging sample data, third blood oxygen level-dependent imaging sample data, and multiple labeled delayed arterial spin labeling perfusion imaging sample data of brain tissues of patients with cerebrovascular reactivity-related encephalopathy; wherein the first blood oxygen level-dependent imaging sample data, the second blood oxygen level-dependent imaging sample data, and the third blood oxygen level-dependent imaging sample data are blood oxygen level-dependent imaging sample data of the corresponding brain tissues under vasodilation stimulation, under no vasodilation stimulation, and in a resting state, respectively; A first training image determining unit, configured to determine a corresponding cerebrovascular reactivity training image according to first blood oxygen level dependent imaging sample data and second blood oxygen level dependent imaging sample data of each brain tissue; a training index image determination unit, configured to determine a corresponding cerebrovascular reactivity training index image according to the third blood oxygen level dependent imaging sample data of each brain tissue; A second training image determination unit is used to determine a corresponding cerebral blood flow training image and an arterial transit time training image according to a plurality of labeled delayed arterial spin labeling perfusion imaging sample data of each brain tissue; The model training unit is used to use the cerebrovascular reactivity training index images, cerebral blood flow training images and arterial transit time training images corresponding to each brain tissue as input, and the corresponding cerebrovascular reactivity training images as supervision to perform supervised training on the preset neural network model to obtain the trained neural network model.

[0102] It should be noted that the device provided in the above embodiment, when implementing its functions, only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the corresponding method embodiment belong to the same concept, and the specific implementation process is detailed in the corresponding method embodiment, which will not be repeated here.

[0103] An embodiment of the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining cerebrovascular reactivity as provided in the above method embodiment.

[0104] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0105] In conjunction with the reference manual Figure 7 , which is a block diagram of an electronic device 700 according to an embodiment of the present invention. The electronic device 700 may include one or more processors 702, a system control logic 708 connected to at least one of the processors 702, a system memory 704 connected to the system control logic 708, a non-volatile memory (NVM) 706 connected to the system control logic 708, and a network interface 710 connected to the system control logic 708.

[0106] The processor 702 may include one or more single-core or multi-core processors. The processor 702 may include any combination of general-purpose processors and special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In the embodiments of this document, the processor 702 may be configured to execute the following: Figures 2 to 5 One or more of the various embodiments shown.

[0107] In some embodiments, system control logic 708 may include any suitable interface controller to provide any suitable interface to at least one of processors 702 and / or any suitable device or component in communication with system control logic 708 .

[0108] In some embodiments, the system control logic 708 may include one or more memory controllers to provide an interface to the system memory 704. The system memory 704 may be used to load and store data and / or instructions. In some embodiments, the memory 704 of the device 700 may include any suitable volatile memory, such as a suitable dynamic random access memory (DRAM).

[0109] NVM / memory 706 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions. In some embodiments, NVM / memory 706 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as at least one of a HDD (Hard Disk Drive), a CD (Compact Disc) drive, and a DVD (Digital Versatile Disc) drive.

[0110] NVM / storage 706 may include a portion of storage resources installed on a device of apparatus 700, or it may be accessible to the apparatus but not necessarily a part of the apparatus. For example, NVM / storage 706 may be accessed over a network via network interface 710.

[0111] In particular, system memory 704 and NVM / storage 706 may include, respectively, a temporary copy and a permanent copy of instructions 720. Instructions 720 may include instructions that, when executed by at least one of processors 702, cause device 700 to perform the following operations: Figures 2 to 5 Instructions of the cerebrovascular reactivity determination method shown. In some embodiments, the instructions 720 , hardware, firmware and / or software components thereof may additionally / alternatively be placed in the system control logic 708 , the network interface 710 and / or the processor 702 .

[0112] The network interface 710 may include a transceiver for providing a radio interface for the device 700 to communicate with any other suitable device (such as a front-end module, an antenna, etc.) through one or more networks. In some embodiments, the network interface 710 may be integrated with other components of the device 700. For example, the network interface 710 may be integrated with at least one of a communication module of the processor 702, a system memory 704, an NVM / storage 706, and a firmware device (not shown) having instructions. When at least one of the processors 702 executes the instructions, the device 700 implements Figures 2 to 5 One or more of the various embodiments shown.

[0113] The network interface 710 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, the network interface 710 may be a network adapter, a wireless network adapter, a telephone modem and / or a wireless modem.

[0114] In one embodiment, at least one of the processors 702 may be packaged together with logic for one or more controllers of the system control logic 708 to form a system in package (SiP). In one embodiment, at least one of the processors 702 may be integrated on the same die with logic for one or more controllers of the system control logic 708 to form a system on chip (SoC).

[0115] The device 700 may further include an input / output (I / O) device 712. The I / O device 712 may include a user interface to enable a user to interact with the device 700; the design of the peripheral component interface enables the peripheral components to also interact with the device 700. In some embodiments, the device 700 further includes a sensor for determining at least one of an environmental condition and location information related to the device 700.

[0116] In some embodiments, the user interface may include, but is not limited to, a display (e.g., an LCD display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., an LED flash), and a keyboard.

[0117] In some embodiments, the peripheral component interface may include, but is not limited to, a non-volatile memory port, an audio jack, and a power interface.

[0118] In some embodiments, the sensors may include, but are not limited to, gyroscope sensors, accelerometers, proximity sensors, ambient light sensors, and positioning units. The positioning unit may also be part of or interact with the network interface 710 to communicate with components of a positioning network (e.g., global positioning system (GPS) satellites).

[0119] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the electronic device 700. In other embodiments of the present invention, the electronic device 700 may include more or fewer components than shown in the figure, or combine some components, or separate some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0120] An embodiment of the present invention also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a method for determining cerebrovascular reactivity. The at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining cerebrovascular reactivity provided by the above method embodiment.

[0121] Optionally, in an embodiment of the present invention, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0122] One embodiment of the present invention further provides a computer program product or a computer program, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the cerebrovascular reactivity determination method provided in the above-mentioned various optional implementation examples.

[0123] It should be noted that the sequence of the embodiments of the present invention described above is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0124] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0125] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0126] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining cerebrovascular reactivity, characterized in that: include: Acquiring a cerebrovascular reactivity index image of a target brain tissue, wherein the cerebrovascular reactivity index image includes a relative value of cerebrovascular reactivity of the target brain tissue in a resting state; Acquiring a cerebral blood flow image and an arterial transit time image of the target brain tissue; The cerebrovascular reactivity index image, the cerebral blood flow image and the arterial passage time image are input into a pre-trained neural network model for processing to obtain a cerebrovascular reactivity image of the target brain tissue, wherein the cerebrovascular reactivity image includes an absolute value of the cerebrovascular reactivity of the target brain tissue.

2. The method according to claim 1, characterized in that The step of obtaining the cerebrovascular reactivity index image of the target brain tissue includes: Acquiring blood oxygen level-dependent imaging data of the target brain tissue in a resting state; registering the blood oxygen level dependent imaging data to a standard space to obtain blood oxygen level dependent imaging data in the standard space; A cerebrovascular reactivity index image of the target brain tissue is determined based on the blood oxygen level-dependent imaging data of the standard space.

3. The method according to claim 2, characterized in that The method further comprises: Preprocessing the blood oxygen level dependent imaging data to obtain preprocessed blood oxygen level dependent imaging data; wherein the preprocessing includes one or more of motion correction, linear component removal, spatial smoothing and bandpass filtering; Wherein, registering the blood oxygen level dependent imaging data to a standard space comprises: The preprocessed blood oxygen level dependent imaging data is registered to the standard space to obtain the blood oxygen level dependent imaging data in the standard space.

4. The method according to claim 1, characterized in that: The step of acquiring the cerebral blood flow image and the arterial transit time image of the target brain tissue comprises: Acquiring arterial spin labeling perfusion imaging data of a plurality of labeling delays of the target brain tissue in a resting state; Registering the plurality of delayed arterial spin labeling perfusion imaging data after labeling to a standard space respectively, to obtain a plurality of delayed arterial spin labeling perfusion imaging data after labeling in the standard space; determining an arterial transit time image of the target brain tissue according to the arterial spin labeling perfusion imaging data of a plurality of post-labeling delays in the standard space; A cerebral blood flow image of the target brain tissue is determined based on the arterial transit time image.

5. The method according to claim 4, characterized in that Determining the cerebral blood flow image of the target brain tissue according to the arterial transit time image comprises: Calculating cerebral blood flow images corresponding to each delayed arterial spin labeling perfusion imaging data after labeling respectively according to the arterial transit time image and each delayed arterial spin labeling perfusion imaging data after labeling; An average image of the cerebral blood flow images corresponding to the arterial spin labeling perfusion imaging data after each labeling delay is calculated as the cerebral blood flow image of the target brain tissue.

6. The method according to claim 4, characterized in that The method further comprises: Preprocessing the plurality of delayed arterial spin labeling perfusion imaging data after labeling to obtain preprocessed plurality of delayed arterial spin labeling perfusion imaging data after labeling; wherein the preprocessing includes motion correction processing; Wherein, the step of registering the plurality of delayed arterial spin labeling perfusion imaging data after labeling to a standard space respectively includes: The preprocessed multiple delayed arterial spin labeling perfusion imaging data are respectively registered to the standard space to obtain multiple delayed arterial spin labeling perfusion imaging data in the standard space.

7. The method according to claim 1, characterized in that The method further includes pre-training the neural network model, and the training process of the neural network model includes: Acquire a training sample data set, wherein the training sample data set includes first blood oxygen level-dependent imaging sample data, second blood oxygen level-dependent imaging sample data, third blood oxygen level-dependent imaging sample data, and multiple labeled delayed arterial spin labeling perfusion imaging sample data of brain tissues of multiple patients with cerebrovascular reactivity-related encephalopathy; wherein the first blood oxygen level-dependent imaging sample data, the second blood oxygen level-dependent imaging sample data, and the third blood oxygen level-dependent imaging sample data are blood oxygen level-dependent imaging sample data of the corresponding brain tissues under vasodilation stimulation, under no vasodilation stimulation, and in a resting state, respectively; determining a corresponding cerebrovascular reactivity training image according to the first blood oxygen level dependent imaging sample data and the second blood oxygen level dependent imaging sample data of each of the brain tissues; Determining a corresponding cerebrovascular reactivity training index image according to the third blood oxygen level-dependent imaging sample data of each of the brain tissues; Determine a corresponding cerebral blood flow training image and an arterial transit time training image according to the plurality of labeled delayed arterial spin labeling perfusion imaging sample data of each brain tissue; The cerebrovascular reactivity training index images, cerebral blood flow training images and arterial transit time training images corresponding to each of the brain tissues are used as inputs, and the corresponding cerebrovascular reactivity training images are used as supervision to perform supervised training on the preset neural network model to obtain a trained neural network model.

8. A device for determining cerebrovascular reactivity, characterized in that: include: A first acquisition module is used to acquire a cerebrovascular reactivity index image of a target brain tissue, wherein the cerebrovascular reactivity index image includes a relative value of cerebrovascular reactivity of the target brain tissue in a resting state; A second acquisition module is used to acquire a cerebral blood flow image and an arterial transit time image of the target brain tissue; The determination module is used to input the cerebrovascular reactivity index image, the cerebral blood flow image and the arterial passage time image into a pre-trained neural network model for processing to obtain a cerebrovascular reactivity image of the target brain tissue, wherein the cerebrovascular reactivity image includes an absolute value of the cerebrovascular reactivity of the target brain tissue.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining cerebrovascular reactivity as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the method for determining cerebrovascular reactivity according to any one of claims 1 to 7.

Images