Method, device and equipment for monitoring volume of focus of cerebral apoplexy and medium

Abstract

The invention discloses a cerebral apoplexy focus volume monitoring method, device and equipment and a storage medium. The method comprises the steps of determining a baseline lesion volume according to a first to-be-detected brain image corresponding to baseline acquisition time, determining blood vessel spatial-temporal characteristics corresponding to at least one interested blood vessel according to at least one interested blood vessel image sequence corresponding to a to-be-detected cerebral apoplexy type, and determining a first to-be-detected cerebral apoplexy type according to the determined blood vessel spatial-temporal characteristics; the spatial-temporal features of the blood vessels, the baseline focus volume and the volume monitoring duration are input into a focus volume prediction model obtained through pre-training, the output monitored focus volume corresponding to the target monitoring time is obtained, the volume monitoring duration represents the time length between the baseline collection time and the target monitoring time, and the volume monitoring duration represents the time length between the baseline collection time and the target monitoring time; according to the embodiment of the invention, the problem that the focus volume of the cerebral apoplexy needs to be monitored by medical imaging equipment is solved, and the medical cost and radiation dose of cerebral apoplexy condition monitoring are reduced.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular, to a method, device, equipment and storage medium for monitoring the volume of stroke lesions. Background Art

[0002] Cerebral stroke, also known as stroke, is an acute cerebrovascular disease. It is a disease caused by sudden rupture of blood vessels in the brain or blockage of blood vessels in the brain, resulting in blood not flowing into the brain and causing damage to brain tissue, including ischemic stroke and hemorrhagic stroke.

[0003] Cerebral stroke has a high mortality rate and a high recurrence rate, and is accompanied by the risk of other neurological disease complications. Clinical studies have shown that the progression of the lesion volume of cerebral stroke is positively correlated with neurological dysfunction.

[0004] In current clinical practice, it is necessary to use medical imaging equipment to monitor the volume change of stroke lesions. However, multiple medical imaging examinations will significantly increase medical costs, and some medical imaging examinations will bring a relatively high radiation dose to patients. Summary of the Invention

[0005] Embodiments of the present invention provide a method, device, equipment and storage medium for monitoring the volume of stroke lesions, so as to solve the problem that the volume of stroke lesions needs to be monitored by medical imaging equipment, and reduce the medical cost and radiation dose of stroke condition monitoring.

[0006] According to an embodiment of the present invention, a method for monitoring the volume of stroke lesions is provided. The method includes:

[0007] Obtaining a to-be-detected image dataset corresponding to the to-be-detected stroke type; wherein, the to-be-detected image dataset includes a first to-be-detected brain image corresponding to the baseline acquisition time and an image sequence of an interested blood vessel of at least one interested blood vessel;

[0008] Determining a baseline lesion volume corresponding to the baseline acquisition time according to the first to-be-detected brain image;

[0009] For each interested blood vessel, determining the vascular spatio-temporal feature of the interested blood vessel according to the image sequence of the interested blood vessel corresponding thereto;

[0010] Inputting the vascular spatio-temporal features, the baseline lesion volume and the volume monitoring duration into a pre-trained lesion volume prediction model, and obtaining a monitored lesion volume corresponding to the target monitoring time as an output;

[0011] Wherein, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time.

[0012] According to another embodiment of the present invention, a monitoring device for the volume of a stroke lesion is provided. The device includes:

[0013] A to-be-detected image dataset acquisition module, configured to acquire a to-be-detected image dataset corresponding to the type of to-be-detected stroke; wherein, the to-be-detected image dataset includes a first to-be-detected brain image corresponding to the baseline acquisition time and an image sequence of an interested blood vessel of at least one interested blood vessel;

[0014] A baseline lesion volume determination module, configured to determine a baseline lesion volume corresponding to the baseline acquisition time according to the first to-be-detected brain image;

[0015] A blood vessel spatio-temporal feature determination module, configured to determine the spatio-temporal feature of each interested blood vessel according to the image sequence of the interested blood vessel corresponding to the interested blood vessel;

[0016] A monitored lesion volume output module, configured to input each of the blood vessel spatio-temporal features, the baseline lesion volume, and the volume monitoring duration into a pre-trained lesion volume prediction model to obtain an output monitored lesion volume corresponding to the target monitoring time;

[0017] Wherein, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time.

[0018] According to another embodiment of the present invention, an electronic device is provided. The electronic device includes:

[0019] At least one processor; and

[0020] A memory communicatively connected to the at least one processor;

[0021] Wherein, the memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the method for monitoring the volume of a stroke lesion according to any embodiment of the present invention.

[0022] According to another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method for monitoring the volume of a stroke lesion according to any embodiment of the present invention is implemented.

[0023] In the technical solution of the embodiment of the present invention, the baseline lesion volume is determined according to the first brain image to be detected corresponding to the baseline acquisition time, and the vascular spatio-temporal features corresponding to at least one interested blood vessel are determined according to at least one interested blood vessel image sequence corresponding to the type of stroke to be detected. Then, each vascular spatio-temporal feature, the baseline lesion volume, and the volume monitoring duration are input into a pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time as the output. Herein, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time. This embodiment utilizes the correlation between the vascular spatio-temporal features and the change of the lesion volume in stroke, solves the problem that the lesion volume in stroke needs to be monitored by medical imaging equipment, and reduces the medical cost and radiation dose of stroke condition monitoring.

[0024] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 It is a flowchart of a method for monitoring the lesion volume of stroke provided by an embodiment of the present invention;

[0027] Figure 2 It is a model architecture diagram of an autoencoder provided by an embodiment of the present invention;

[0028] Figure 3 It is a network architecture diagram of a fully connected network provided by an embodiment of the present invention;

[0029] Figure 4 It is a flowchart of another method for monitoring the lesion volume of stroke provided by an embodiment of the present invention;

[0030] Figure 5 It is a schematic diagram of the process from an interested blood vessel image to a straightened blood vessel image provided by an embodiment of the present invention;

[0031] Figure 6 It is a flowchart of another method for monitoring the lesion volume of stroke provided by an embodiment of the present invention;

[0032] Figure 7Flowchart of a specific example of a method for monitoring the volume of a stroke lesion provided by an embodiment of the present invention;

[0033] Figure 8 Structural schematic diagram of a device for monitoring the volume of a stroke lesion provided by an embodiment of the present invention;

[0034] Figure 9 Structural schematic diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0035] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0036] It should be noted that the terms "first", "second", "original", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances 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 inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0037] Figure 1 Flowchart of a method for monitoring the volume of a stroke lesion provided by an embodiment of the present invention. This embodiment is applicable to the situation of monitoring the progression change of the volume of a stroke lesion. This method can be executed by a device for monitoring the volume of a stroke lesion, and the device for monitoring the volume of a stroke lesion can be implemented in the form of hardware and / or software, and the device for monitoring the volume of a stroke lesion can be configured in a terminal device. As Figure 1 shown, the method includes:

[0038] S110. Obtain a dataset of images to be detected corresponding to the type of stroke to be detected.

[0039] In this embodiment, the type of stroke to be detected is ischemic stroke or hemorrhagic stroke. Among them, ischemic stroke refers to the necrosis of brain tissue caused by the stenosis or occlusion of the blood supply arteries of the brain (such as the carotid artery and / or vertebral artery) and insufficient blood supply to the brain. Hemorrhagic stroke, also known as cerebral hemorrhage, parenchymal hemorrhage of the brain, intracerebral hemorrhage, etc., refers to the brain tissue damage caused by the rupture and bleeding of intracranial blood vessels due to non-traumatic reasons.

[0040] In this embodiment, the image dataset to be detected includes a first brain image to be detected corresponding to the baseline acquisition time and an image sequence of the blood vessels of interest of at least one blood vessel of interest.

[0041] Specifically, the first brain image to be detected is used to represent the brain image containing the stroke lesion acquired at the baseline acquisition time, and the baseline lesion volume is used to represent the lesion volume of the stroke in the first brain image to be detected.

[0042] Specifically, the image sequence of the blood vessels of interest includes at least two images of the blood vessels of interest based on the time series. The image of the blood vessel of interest is used to represent the blood vessel image containing the blood vessel of interest corresponding to the type of stroke to be detected. Exemplarily, at least one blood vessel of interest corresponding to ischemic stroke includes, but is not limited to, the carotid artery, vertebral artery, etc. At least one blood vessel of interest corresponding to hemorrhagic stroke includes, but is not limited to, the inferior anastomotic vein (Labbe), the superficial middle cerebral vein (SMCV), and the basal vein of Rosenthal (BVR), etc. The corresponding relationship between the type of stroke to be detected and each blood vessel of interest is not limited here, and the blood vessels of interest can be custom-selected according to actual needs.

[0043] Among them, the image type of the first brain image to be detected and the image type of the image of the blood vessel of interest can be the same or different. Exemplarily, the image type of the first brain image to be detected or the image of the blood vessel of interest includes, but is not limited to, direct digital radiography (DR), computed tomography (CT), magnetic resonance imaging (MRI), positron emission computed tomography (PET), or ultrasound image, etc. The image type of the first brain image to be detected or the image of the blood vessel of interest is not limited here, and it can be custom-set according to actual needs.

[0044] S120. Determine the baseline lesion volume corresponding to the baseline acquisition time according to the first brain image to be detected.

[0045] In an optional embodiment, determining the baseline lesion volume corresponding to the baseline acquisition time according to the first brain image to be detected includes: inputting the first brain image to be detected into a pre-trained volume extraction model to obtain the output baseline lesion volume corresponding to the baseline acquisition time.

[0046] Specifically, the method further includes: inputting the training brain image into an untrained volume extraction model to obtain the output predicted volume features, determining the first loss function according to the predicted volume features and the standard volume features, and adjusting the model parameters of the volume extraction model according to the first loss function; until the first loss function converges, using the volume extraction model in the current iteration process as the trained volume extraction model.

[0047] Exemplarily, the function type of the first loss function includes but is not limited to square loss function, logarithmic loss function, exponential loss function, mean square error loss function, logistic regression loss function, Huber loss function, cross-entropy loss function or Kullback-Leibler divergence loss function, etc. The function type of the first loss function is not limited here, and can be specifically customized according to actual needs.

[0048] In another optional embodiment, determining the baseline lesion volume corresponding to the baseline acquisition time according to the first brain image to be detected includes: performing a segmentation operation on the first brain image to be detected to obtain a stroke lesion image, and determining the baseline lesion volume corresponding to the baseline acquisition time according to the lesion voxel size corresponding to the stroke lesion image.

[0049] Exemplarily, the segmentation tool used for the segmentation operation can be The Medical Imaging Interaction Toolkit (MITK). The segmentation tool used for the segmentation operation is not limited here, and can be specifically customized according to actual needs.

[0050] S130. For each blood vessel of interest, determine the spatio-temporal characteristics of the blood vessel of interest according to the sequence of images of the blood vessel of interest corresponding thereto.

[0051] Specifically, the spatio-temporal characteristics of the blood vessel are used to characterize the temporal characteristics and spatial characteristics of the blood vessel of interest. Exemplarily, the temporal characteristics of the blood vessel include but are not limited to hemodynamic characteristics, and the spatial characteristics of the blood vessel include but are not limited to vascular morphological characteristics, vascular structural characteristics and vascular elasticity characteristics, etc.

[0052] In an alternative embodiment, according to the sequence of images of the blood vessel of interest corresponding to the blood vessel of interest, the spatio-temporal characteristics of the blood vessel of interest are determined, including: inputting the sequence of images of the blood vessel of interest corresponding to the blood vessel of interest into a pre-trained feature extraction model to obtain the spatio-temporal characteristics of the output blood vessel of interest.

[0053] Exemplarily, the model type of the feature extraction model can be the encoder in an autoencoder. Specifically, the method further includes: inputting the sequence of training blood vessel images into an untrained autoencoder, and through the encoder in the autoencoder, performing feature extraction on the input sequence of training blood vessel images to obtain spatio-temporal characteristics of the blood vessel; through the decoder in the autoencoder, performing a decoding operation on the input spatio-temporal characteristics of the blood vessel to obtain a predicted sequence of blood vessel images; determining a second loss function according to the predicted sequence of blood vessel images and the sequence of training blood vessel images, and adjusting the model parameters of the autoencoder according to the second loss function; until the second loss function converges, taking the encoder in the autoencoder in the current iteration process as the feature extraction model.

[0054] Exemplarily, the function type of the second loss function includes but is not limited to square loss function, logarithmic loss function, exponential loss function, mean square error loss function, logistic regression loss function, Huber loss function, cross-entropy loss function or Kullback-Leibler divergence loss function, etc. The function type of the second loss function is not limited here, and can be specifically customized according to actual needs.

[0055] Figure 2 FIG. is a model architecture diagram of an autoencoder provided by an embodiment of the present invention. Specifically, Figure 2 the left model structure in is the encoder in the autoencoder, and the right model structure is the decoder in the autoencoder. The model structures of the encoder and the decoder are basically symmetric and consist of a stack-shaped structure of multiple convolutional components. The decoder has one more convolutional output layer than the encoder.

[0056] S140: Input the spatio-temporal characteristics of each blood vessel, the baseline lesion volume, and the volume monitoring duration into a pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time as the output.

[0057] In this embodiment, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time. Exemplarily, the volume monitoring duration can be 1 day, 7 days or 30 days. The volume monitoring duration is not limited here, and can be specifically customized according to actual needs.

[0058] Exemplarily, the model types of the lesion volume prediction model include, but are not limited to, CNN network (Convolutional Neural Networks), FCN network (Fully Convolutional Networks), ResNet, DNN network (Deep Neural Networks), RNN network (Recurrent Neural Network), or Transformer network, etc. The model type of the lesion volume prediction model is not limited here, and can be specifically customized according to actual needs.

[0059] In an alternative embodiment, the model type of the lesion volume prediction model is a fully connected network.

[0060] Figure 3 This is a network architecture diagram of a fully connected network provided by an embodiment of the present invention. Specifically, the fully connected network is composed of 5 fully connected layers.

[0061] Specifically, the method further includes: inputting the training volume features, volume training duration, and the vascular spatio-temporal features corresponding to at least one training vascular image sequence into the uncompleted lesion volume prediction model to obtain the predicted lesion volume corresponding to the training monitoring time, determining the third loss function according to the predicted lesion volume and the standard lesion volume, and adjusting the model parameters of the lesion volume prediction model according to the third loss function; until the third loss function converges, taking the lesion volume prediction model in the current iteration process as the completed lesion volume prediction model.

[0062] Exemplarily, the function types of the third loss function include, but are not limited to, square loss function, logarithmic loss function, exponential loss function, mean square error loss function, logistic regression loss function, Huber loss function, cross-entropy loss function, or Kullback-Leibler divergence loss function, etc. The function type of the third loss function is not limited here, and can be specifically customized according to actual needs.

[0063] In the technical solution of this embodiment, the baseline lesion volume is determined according to the first brain image to be detected corresponding to the baseline acquisition time. According to at least one sequence of images of the blood vessels of interest corresponding to the type of stroke to be detected, the spatio-temporal characteristics of the blood vessels corresponding to at least one blood vessel of interest are determined. The spatio-temporal characteristics of each blood vessel, the baseline lesion volume, and the volume monitoring duration are input into a pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time. Among them, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time. This embodiment utilizes the correlation between the spatio-temporal characteristics of blood vessels and the change in the lesion volume of stroke, solves the problem that the lesion volume of stroke needs to be monitored by medical imaging equipment, and reduces the medical cost and radiation dose of stroke condition monitoring.

[0064] Figure 4 FIG. 4 is a flowchart of another method for monitoring the lesion volume of stroke provided by an embodiment of the present invention. In this embodiment, the step of "determining the spatio-temporal characteristics of the blood vessels of interest according to the sequence of images of the blood vessels of interest corresponding to the blood vessels of interest" in the above embodiment is further refined. As Figure 4 shown, the method includes:

[0065] S210. Obtain a dataset of images to be detected corresponding to the type of stroke to be detected.

[0066] S220. Determine the baseline lesion volume corresponding to the baseline acquisition time according to the first brain image to be detected.

[0067] S210-S220 in this embodiment are the same or similar to Figure 1 S110-S120 shown above, and will not be described in detail herein.

[0068] S230. For each blood vessel of interest, perform a straightening operation on at least two images of the blood vessels of interest in the sequence of images of the blood vessels of interest corresponding to the blood vessel of interest to obtain at least two straightened blood vessel images.

[0069] In an optional embodiment, performing a straightening operation on at least two images of the blood vessels of interest in the sequence of images of the blood vessels of interest corresponding to the blood vessel of interest to obtain at least two straightened blood vessel images includes: for each image of the blood vessel of interest, obtaining the center point coordinates corresponding to at least two blood vessel center points on the blood vessel center line corresponding to the image of the blood vessel of interest; determining the tangent vectors corresponding to the blood vessel center points according to the center point coordinates; determining the normal plane coordinates corresponding to at least two blood vessel center points according to the center point coordinates and the tangent vectors; and determining the straightened blood vessel image corresponding to the image of the blood vessel of interest according to the normal plane coordinates.

[0070] Exemplarily, the acquisition tool for the center point coordinates can be The vascular modeling toolkit (VMTK). There is no limitation on the acquisition tool adopted herein, and it can be specifically selected according to actual requirements.

[0071] Figure 5 FIG. 4 is a schematic diagram of the process from the blood vessel image of interest to the straightened blood vessel image provided by an embodiment of the present invention. Specifically, Figure 5 FIG. 4(a) shows the blood vessel image of interest. Each normal plane is represented by {P 1 , P 2 ,..., P n}, and each blood vessel center point is represented by {a 1 , a 2 ,..., a L}. represents the tangent vector corresponding to the i-th blood vessel center point. Specifically, L represents the number of blood vessel center points on the blood vessel center line.

[0072] As shown in FIG. 4(b), for two adjacent normal planes on the blood vessel center line, such as the normal plane P Figure 5 and the normal plane P i-1 and the normal plane P i , the normal plane P i can be regarded as obtained by rotating the normal plane P i-1 to obtain the plane P′ i , and then translating the plane P′ i .

[0073] Specifically, the normal plane P i = {v ij | 1 ≤ j ≤ m}, where m represents the number of points included in the normal plane P i . The included angle between the tangent vector i-1 corresponding to the normal plane P and the tangent vector i corresponding to the normal plane P satisfies the formula:

[0074]

[0075] Specifically, the rotation axis between the tangent vector and the tangent vector is a unit vector, satisfying the formula:

[0076]

[0077] Specifically, the plane P′ obtained by rotating the normal plane P i-1 ​i The process can be regarded as all points v on the normal plane P i-1 , along a rotation axis u passing through the center point a of the blood vessel on the normal plane P (i-1)j , rotate by θ i-1 to obtain all the corresponding points v' on the plane P'. i-1 (i-1)i (i-1)i i ij .

[0078] Specifically, the rotation of the corresponding points is represented by the rotation of vectors in three-dimensional space. As shown in Figure c of Figure 5 , rotate by θ along a rotation axis passing through point a to obtain During this rotation process, the rotation axis is a quantity with both magnitude and direction, but its magnitude (length) is not important here. To eliminate the redundant degree of freedom of the rotation axis modulus length and for the convenience of calculation, the rotation axis modulus length is defined as: That is is a unit vector.

[0079] Specifically, decompose and into two component vectors parallel to the rotation axis and perpendicular (orthogonal) to the rotation axis ( and ), that is where is the orthogonal projection of on the rotation axis , so we can get:

[0080]

[0081] Since and are both perpendicular (orthogonal) to the rotation axis so we can regard rotating around the rotation axis to obtain as a rotation in a plane. Because rotation does not change the modulus length, the rotation trajectory can be represented by a circle. As shown in Figure d of Figure 5 , a vector and that is perpendicular (orthogonal) to the rotation axis can be constructed, that is​​​​

[0082]

[0083]

[0084] Among them, is equal in modulus to , indicating that is also on the circle. Decompose into two component vectors perpendicular (orthogonal) to and perpendicular (orthogonal) to , namely That is

[0085]

[0086]

[0087] Specifically, the process of obtaining the plane P' i-1 by rotating the normal plane P i can be regarded as all the vectors i-1 on the normal plane P rotating along a rotation axis i-1 passing through the blood vessel center point a i-1 of the normal plane P (unit vector) by θ (i-1)i to obtain all the corresponding vectors on the plane P' i , that is

[0088]

[0089]

[0090]

[0091]

[0092] Specifically, the process of obtaining the normal plane P i by translating the plane P' i can be regarded as all the vectors i on the plane P' rotating and translating along the direction of the blood vessel center point a i-1 on the blood vessel center line and the blood vessel center point a i to obtain all the corresponding vectors on the normal plane P i , that is That is

[0093]

[0094] Through the above derivation process, the normal plane P can be obtained.i-1 and the rotation and translation relationship with the normal plane P i Therefore, only need to arbitrarily construct a known plane P 0 , then all the normal planes can be obtained.

[0095] Specifically, set the known plane P 0 as the plane passing through the origin of the coordinate system and perpendicular to the z-axis, the center point a 0 = [0, 0, 0], the normal vector and the coordinates of each point {v 01 , v 02 ,..., v 0m} on it are known. According to the above formula, all the normal plane coordinates P = {P 1 , P 2 ,..., P n} on the blood vessel center line can be obtained. By using the interpolation algorithm, introduce P into the voxel values in the four-dimensional blood vessel image of interest, and get all the normal plane coordinates V = {V 1 , V 2 ,..., V n} in the straightened blood vessel image.

[0096] As shown in the e figure of Figure 5 , introduce V into a new rectangular coordinate system and arrange them in sequence on the z-axis, then the straightened blood vessel image can be obtained. At this time, the straightened blood vessel image is in a four-dimensional matrix with a size of [T, L, 16, 16], where T represents the number of blood vessel images of interest in the blood vessel image sequence of interest.

[0097] S240. Input each straightened blood vessel image into the pre-trained feature extraction model to obtain the spatio-temporal features of the blood vessel of interest output.

[0098] The feature extraction model in this embodiment is the same or similar to the feature extraction model used for feature extraction of the blood vessel image sequence of interest in the above embodiment, and will not be elaborated here in this embodiment.

[0099] S250. Input each spatio-temporal feature of the blood vessel, the baseline lesion volume and the volume monitoring duration into the pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time output.

[0100] S250 in this embodiment is the same or similar to S140 shown in the above embodiment Figure 1 , and will not be elaborated here in this embodiment.

[0101] In the technical solution of this embodiment, for each blood vessel of interest, at least two blood vessel images of interest in the image sequence of the blood vessel of interest are respectively subjected to a straightening operation to obtain at least two straightened blood vessel images. The straightened blood vessel images are input into a pre-trained feature extraction model to obtain the spatio-temporal features of the blood vessel of interest as the output, eliminating the interference of the blood vessel morphological features in the traditional blood vessel images on the prediction result of the lesion volume prediction model and improving the accuracy of monitoring the lesion volume of stroke.

[0102] Figure 6 FIG. 4 is a flowchart of another method for monitoring the lesion volume of stroke provided by an embodiment of the present invention. This embodiment further refines the "obtaining the to-be-detected image dataset corresponding to the to-be-detected stroke type" in the above embodiment. As Figure 6 shown, the method includes:

[0103] S310. Obtain the blood vessel type of at least one blood vessel of interest corresponding to the to-be-detected stroke type.

[0104] In this embodiment, when the to-be-detected stroke type is ischemic stroke, the blood vessel type is arterial blood vessel, and when the to-be-detected stroke type is hemorrhagic stroke, the blood vessel type is venous blood vessel.

[0105] S320. Perform a segmentation operation on the second to-be-detected brain image sequence according to the time decay curves respectively corresponding to the brain voxels in the second to-be-detected brain image sequence to obtain a cerebrovascular image sequence corresponding to the blood vessel type.

[0106] Specifically, the second to-be-detected brain image sequence includes at least two second to-be-detected brain images based on a time series, and the second to-be-detected brain images are used to represent brain images including at least one blood vessel of interest.

[0107] In an optional embodiment, the method further includes: obtaining the skull coordinates in the first frame of the original brain image sequence of the original brain image sequence, and performing a rigid registration operation on the original brain images in the original brain image sequence except the first frame of the original brain image according to the skull coordinates to obtain a registered brain image sequence; performing a segmentation operation on the first frame of the registered brain image sequence in the registered brain image sequence according to a preset skull threshold to obtain a skull mask image; and performing a deletion operation on each registered brain image in the registered brain image sequence according to the skull mask image to obtain the second to-be-detected brain image sequence.

[0108] Specifically, the rigid registration operation includes registration operations in six degrees of freedom, namely translation along the x-axis, translation along the y-axis, translation along the z-axis, rotation around the x-axis, rotation around the y-axis, and rotation around the z-axis.

[0109] Exemplarily, when the image type of the original brain image is a CT image, the preset skull threshold can be 155 HU. Specifically, an image composed of image voxels with image voxel values less than 155 HU in the first-frame registered brain image is used as the skull mask image. Here, the preset skull threshold is not limited and can be custom-set according to actual needs.

[0110] The advantage of such a setting is that it can reduce the alignment error of the brain image sequence and the influence of the skull on the segmentation effect of the subsequent blood vessel images of interest, improve the image quality of the second brain image to be detected, and thus improve the accuracy of monitoring the lesion volume of stroke.

[0111] Exemplarily, the time decay curve can be generated using a Gaussian function. In an alternative embodiment, according to the time decay curves respectively corresponding to each brain voxel in the second brain image sequence to be detected, a segmentation operation is performed on the second brain image sequence to be detected to obtain a cerebrovascular image sequence corresponding to the blood vessel type, including: determining a curve sum sequence respectively corresponding to each brain voxel according to at least one time decay high-order curve respectively corresponding to each brain voxel in the second brain image sequence to be detected; using an image sequence composed of at least one brain voxel in the second brain image sequence to be detected whose curve sum sequence meets the preset sequence condition as the cerebrovascular image sequence; determining the time to peak respectively corresponding to each cerebrovascular voxel according to the time decay zero-order curve respectively corresponding to each cerebrovascular voxel in the cerebrovascular image sequence; using an image sequence composed of at least one cerebrovascular voxel in the cerebrovascular image sequence whose time to peak meets the preset time range corresponding to the blood vessel type as the cerebrovascular image sequence corresponding to the blood vessel type.

[0112] Exemplarily, each time decay high-order curve can be a time decay first-order curve and a time decay second-order curve. Specifically, the curve sum sequence includes the absolute value sum of curves respectively corresponding to at least one time decay high-order curve.

[0113] Exemplarily, the absolute value sum FG of the curve corresponding to the time decay first-order curve satisfies the formula: The absolute value sum SG of the curve corresponding to the time decay second-order curve satisfies the formula: where T represents the number of images in the second brain image sequence to be detected, v′ i represents the first-order curve parameter value of the brain voxel in the i-th frame of the second brain image to be detected, and v″ i represents the second-order curve parameter value of the brain voxel in the i-th frame of the second brain image to be detected.

[0114] Specifically, the preset sequence condition includes a curve parameter range corresponding to at least one time-decaying high-order curve. Exemplarily, the curve parameter range corresponding to the first-order time-decaying curve is [0 21], and the curve parameter range corresponding to the second-order time-decaying curve is [0 2.15]. Here, the preset sequence condition is not limited, and it can be specifically customized according to actual needs.

[0115] Specifically, the time to peak (TTP) is used to characterize the time when the zero-order curve parameter value of the cerebrovascular voxel in the zero-order time-decaying curve reaches the peak.

[0116] In an alternative embodiment, the method further includes: generating a cerebrovascular voxel-time to peak histogram according to the time to peak corresponding to each cerebrovascular voxel; taking the minimum time to peak between the two peak times to peak in the cerebrovascular voxel-time to peak histogram as the preset time threshold; constructing a preset time range according to the blood vessel type and the preset time threshold; wherein, when the blood vessel type is a venous blood vessel, the minimum boundary value in the preset time range is the preset time threshold, and when the blood vessel type is an arterial blood vessel, the maximum boundary value in the preset time range is the preset time threshold.

[0117] Specifically, the cerebrovascular image sequence is a cerebral venous image sequence or a cerebral arterial image sequence. Since the time for the contrast agent to reach the artery is earlier than the time to reach the vein, the cerebral venous image and the cerebral arterial image in the cerebrovascular image can be distinguished by the preset time threshold.

[0118] S330. Perform a segmentation operation on the cerebrovascular image sequence according to the zero-order time-decaying curve corresponding to each cerebrovascular voxel in the cerebrovascular image sequence to obtain an interested blood vessel image sequence corresponding to each interested blood vessel.

[0119] In an alternative embodiment, performing a segmentation operation on the cerebrovascular image sequence according to the zero-order time-decaying curve corresponding to each cerebrovascular voxel in the cerebrovascular image sequence to obtain an interested blood vessel image sequence corresponding to each interested blood vessel includes: determining the time to peak corresponding to each cerebrovascular voxel according to the zero-order time-decaying curve corresponding to each cerebrovascular voxel in the cerebrovascular image sequence; obtaining the difference time corresponding to each time to peak and the preset time threshold; for each interested blood vessel, taking the image sequence composed of at least one cerebrovascular voxel whose difference time satisfies the difference time range corresponding to the interested blood vessel as the interested blood vessel image sequence corresponding to the interested blood vessel.

[0120] Specifically, the method for determining the preset time threshold is the same as or similar to the method for determining the preset time threshold in the process of determining the preset time range in the above embodiment, and this embodiment will not be elaborated here.

[0121] Specifically, the difference time ranges corresponding to each blood vessel of interest can be the same or different. Exemplarily, the difference time range can be [0 5], which is not limited in this embodiment and can be specifically set according to actual needs.

[0122] S340. Add the first brain image to be detected and the image sequences of each blood vessel of interest to the dataset of images to be detected corresponding to the type of stroke to be detected.

[0123] S350. Determine the baseline lesion volume corresponding to the baseline acquisition time according to the first brain image to be detected.

[0124] Specifically, the first brain image to be detected can be any second brain image in the second sequence of brain images to be detected, or can be other brain images. The image type of the first brain image to be detected can be the same as or different from the image type of the second brain image to be detected.

[0125] S360. For each blood vessel of interest, determine the spatio-temporal characteristics of the blood vessel of interest according to the image sequence of the blood vessel of interest corresponding thereto.

[0126] S370. Input the spatio-temporal characteristics of each blood vessel, the baseline lesion volume, and the volume monitoring duration into the pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time as the output.

[0127] S350 - S370 in this embodiment are the same as or similar to Figure 1 S120 - S140 shown in the above embodiment, or are the same as or similar to Figure 4 S220 - S250 shown in the above embodiment, and will not be elaborated herein.

[0128] Figure 7 It is a flowchart of a specific example of a method for monitoring the lesion volume of stroke provided by an embodiment of the present invention. Specifically, taking the first brain image to be detected as an NCCT image (Non - comparative Computed Tomography), the second sequence of brain images to be detected as a TRCT image sequence (Temporal resolution Computed Tomography), and the type of stroke to be detected as hemorrhagic stroke as an example, where the baseline NCCT image represents the NCCT image collected at the baseline acquisition time, and the baseline TRCT image sequence represents the TRCT image sequence collected at the baseline acquisition time.

[0129] Specifically, based on the first-frame brain image in the baseline TRCT image sequence, a rigid registration operation is performed on the baseline TRCT image sequence to obtain a registered TRCT image sequence. An image composed of image voxels with voxel values less than 155 HU in the first-frame brain image is used as a skull mask image, and based on the skull mask image, a deletion operation is performed on the registered TRCT image sequence to obtain a TRCT image sequence to be detected.

[0130] Specifically, based on the TRCT image sequence to be detected, a time decay first-order curve and a time decay second-order curve are determined. The total absolute value FG of the curve corresponding to the time decay first-order curve and the total absolute value SG of the curve corresponding to the time decay second-order curve are obtained. According to the preset sequence condition constructed by the total absolute value FG of the curve being less than 21 and the total absolute value SG of the curve, a segmentation operation is performed on the TRCT image sequence to be detected to obtain a cerebral blood vessel image sequence.

[0131] Specifically, based on the time decay zero-order curve corresponding to the cerebral blood vessel image sequence, a cerebral blood vessel voxel-time to peak histogram is determined. The minimum time to peak between the two peak times to peak in the cerebral blood vessel voxel-time to peak histogram is used as a preset time threshold Vttp. An image sequence composed of cerebral blood vessel voxels with a time to peak TTP greater than the preset time threshold Vttp in the cerebral blood vessel image sequence is used as a cerebral vein image sequence, and an image sequence composed of cerebral vein voxels with a difference time between the time to peak TTP and the preset time threshold Vttp less than 5 in the cerebral vein image sequence is used as an image sequence of blood vessels of interest.

[0132] Specifically, using the VMTK modeling tool, the centerlines of blood vessels are respectively obtained from each image of blood vessels of interest, and based on the centerlines of each blood vessel, the straightened blood vessel images respectively corresponding to each image of blood vessels of interest are determined. The straightened blood vessel images are input into a pre-trained encoder to obtain the output spatio-temporal features of blood vessels. Using the MITK segmentation tool, a segmentation operation is performed on the baseline NCCT image to obtain a stroke lesion image, and the baseline lesion volume corresponding to the stroke lesion image is obtained. The fusion features obtained based on the spatio-temporal features of blood vessels, the baseline lesion volume, and the volume monitoring duration are input into a pre-trained lesion volume prediction model to obtain the output monitored lesion volume corresponding to the target monitoring time.

[0133] The technical solution of this embodiment solves the problem of high acquisition difficulty caused by directly collecting the image sequence of the blood vessels of interest through an image imaging device. By obtaining the blood vessel types of at least one blood vessel of interest corresponding to the type of stroke to be detected, performing a segmentation operation on each second brain image to be detected according to the time decay curve corresponding to each brain voxel in the second brain image sequence to obtain a cerebrovascular image sequence corresponding to the blood vessel type, performing a segmentation operation on at least two cerebrovascular images according to the time decay zero-order curve corresponding to each cerebrovascular voxel in the cerebrovascular image sequence to obtain an image sequence of the blood vessels of interest corresponding to each blood vessel of interest, and adding the first brain image to be detected and the image sequences of the blood vessels of interest to the dataset of the image to be detected corresponding to the type of stroke to be detected, the image quality of the image sequence of the blood vessels of interest is improved, and further the accuracy of monitoring the lesion volume of the stroke is improved.

[0134] The following is an embodiment of a device for monitoring the lesion volume of a stroke provided by an embodiment of the present invention. This device and the method for monitoring the lesion volume of a stroke in the above embodiment belong to the same inventive concept. For the details not described in detail in the embodiment of the device for monitoring the lesion volume of a stroke, reference may be made to the content of the method for monitoring the lesion volume of a stroke in the above embodiment.

[0135] Figure 8 The following is a schematic structural diagram of a device for monitoring the lesion volume of a stroke provided by an embodiment of the present invention. As Figure 8 shown, the device includes: a to-be-detected image dataset acquisition module 410, a baseline lesion volume determination module 420, a blood vessel spatio-temporal feature determination module 430, and a monitored lesion volume output module 440.

[0136] Among them, the to-be-detected image dataset acquisition module 410 is configured to acquire a to-be-detected image dataset corresponding to the type of stroke to be detected; wherein, the to-be-detected image dataset includes a first to-be-detected brain image corresponding to the baseline acquisition time and an image sequence of at least one blood vessel of interest.

[0137] The baseline lesion volume determination module 420 is configured to determine a baseline lesion volume corresponding to the baseline acquisition time according to the first to-be-detected brain image.

[0138] The blood vessel spatio-temporal feature determination module 430 is configured to, for each blood vessel of interest, determine the spatio-temporal feature of the blood vessel of interest according to the image sequence of the blood vessel of interest corresponding to the blood vessel of interest.

[0139] The monitored lesion volume output module 440 is configured to input each blood vessel spatio-temporal feature, the baseline lesion volume, and the volume monitoring duration into a pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time as the output.

[0140] Among them, the volume monitoring duration characterizes the time length between the baseline acquisition time and the target monitoring time.

[0141] The technical solution of this embodiment utilizes the correlation relationship between the spatio-temporal characteristics of blood vessels and the volume change of the lesion in stroke, solves the problem that the volume of the lesion in stroke needs to be monitored by medical imaging equipment, and reduces the medical cost and radiation dose of stroke condition monitoring.

[0142] In an alternative embodiment, the blood vessel spatio-temporal feature determination module 430 includes:

[0143] A straightened blood vessel image determination unit, configured to perform a straightening operation on at least two interested blood vessel images in an interested blood vessel image sequence corresponding to an interested blood vessel, to obtain at least two straightened blood vessel images;

[0144] A blood vessel spatio-temporal feature output unit, configured to input each straightened blood vessel image into a pre-trained feature extraction model, to obtain the output spatio-temporal features of the interested blood vessel.

[0145] In an alternative embodiment, the straightened blood vessel image determination unit is specifically configured to:

[0146] For each interested blood vessel image, obtain the center point coordinates corresponding to at least two blood vessel center points on the blood vessel center line corresponding to the interested blood vessel image;

[0147] According to each center point coordinate, determine the tangent vector corresponding to each blood vessel center point;

[0148] According to each center point coordinate and each tangent vector, determine the normal plane coordinates corresponding to at least two blood vessel center points;

[0149] According to each normal plane coordinate, determine the straightened blood vessel image corresponding to the interested blood vessel image.

[0150] In an alternative embodiment, the to-be-detected image data set acquisition module 410 includes:

[0151] A blood vessel type acquisition unit, configured to acquire the blood vessel types of at least one interested blood vessel corresponding to the to-be-detected stroke type;

[0152] A cerebrovascular image sequence determination unit, configured to perform a segmentation operation on the second to-be-detected brain image sequence according to the time decay curves corresponding to each brain voxel in the second to-be-detected brain image sequence, to obtain a cerebrovascular image sequence corresponding to the blood vessel type;

[0153] An interested blood vessel image sequence determination unit, configured to perform a segmentation operation on the cerebrovascular image sequence according to the time decay zero-order curves respectively corresponding to each cerebrovascular voxel in the cerebrovascular image sequence, so as to obtain the interested blood vessel image sequences respectively corresponding to each interested blood vessel.

[0154] In an alternative embodiment, the cerebrovascular image sequence determination unit is specifically configured to:

[0155] Determine a curve sum sequence respectively corresponding to each brain voxel according to at least one time decay higher-order curve respectively corresponding to each brain voxel in the second brain image sequence to be detected; use the image sequence formed by at least one brain voxel in the second brain image sequence to be detected whose curve sum sequence meets a preset sequence condition as the cerebrovascular image sequence;

[0156] Determine the time to peak respectively corresponding to each cerebrovascular voxel according to the time decay zero-order curve respectively corresponding to each cerebrovascular voxel in the cerebrovascular image sequence;

[0157] Use the image sequence formed by at least one cerebrovascular voxel in the cerebrovascular image sequence whose time to peak meets a preset time range corresponding to the blood vessel type as the cerebrovascular image sequence corresponding to the blood vessel type.

[0158] In an alternative embodiment, the apparatus further includes:

[0159] A preset time range determination module, configured to generate a cerebrovascular voxel-time to peak histogram according to the time to peak respectively corresponding to each cerebrovascular voxel;

[0160] Use the minimum time to peak between the two peak times to peak in the cerebrovascular voxel-time to peak histogram as the preset time threshold;

[0161] Construct a preset time range according to the blood vessel type and the preset time threshold;

[0162] Wherein, when the blood vessel type is a venous blood vessel, the minimum boundary value in the preset time range is the preset time threshold, and when the blood vessel type is an arterial blood vessel, the maximum boundary value in the preset time range is the preset time threshold.

[0163] In an alternative embodiment, the interested blood vessel image sequence determination unit is specifically configured to:

[0164] Determine the time to peak respectively corresponding to each cerebrovascular voxel according to the time decay zero-order curve respectively corresponding to each cerebrovascular voxel in the cerebrovascular image sequence;

[0165] Obtain the difference time respectively corresponding to each time to peak and the preset time threshold;

[0166] For each blood vessel of interest, an image sequence composed of at least one cerebrovascular voxel whose difference time satisfies the difference time range corresponding to the blood vessel of interest is used as the image sequence of the blood vessel of interest corresponding to the blood vessel of interest.

[0167] In an optional embodiment, the device further includes:

[0168] A second brain image sequence to be detected determining module, configured to obtain the skull coordinates in the first original brain image in the original brain image sequence, and perform a rigid registration operation on the original brain images in the original brain image sequence except the first original brain image according to the skull coordinates to obtain a registered brain image sequence;

[0169] Perform a segmentation operation on the first registered brain image in the registered brain image sequence according to a preset skull threshold to obtain a skull mask image;

[0170] Perform a deletion operation on each registered brain image in the registered brain image sequence according to the skull mask image to obtain a second brain image sequence to be detected.

[0171] The monitoring device for the volume of a stroke lesion provided by the embodiments of the present invention can execute the monitoring method for the volume of a stroke lesion provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0172] Figure 9 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital assistant, a cellular phone, a smart phone, a wearable device (such as a helmet, glasses, a watch, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0173] Such as Figure 9As shown, the electronic device 10 includes at least one processor 11 and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. The memory stores a computer program executable by the at least one processor 11. The processor 11 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0174] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information or data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0175] The processor 11 can be various general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the method for monitoring the volume of a stroke lesion provided in the above embodiment.

[0176] In some embodiments, the method for monitoring the volume of a stroke lesion provided by the above embodiments can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for monitoring the volume of a stroke lesion described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the method for monitoring the volume of a stroke lesion by any other suitable means (e.g., by means of firmware).

[0177] The various embodiments of the systems and techniques described above in this document can be implemented in the following systems or combinations thereof: digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), system on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0178] The computer program for implementing the method for monitoring the volume of a stroke lesion of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer programs are executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0179] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable storage medium. Examples of machine-readable storage media would include electrical connections based on at least one wire, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0180] To provide for interaction with a user, the systems and techniques described herein can be implemented on a terminal device having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the terminal device. Other kinds of devices can also provide for interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0181] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0182] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The client-server relationship is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and virtual private servers (VPS).

[0183] It should be understood that various forms of processes shown above can be used, steps can be reordered, added or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0184] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for monitoring the volume of a stroke lesion, characterized in that, it includes: Obtaining a to-be-detected image dataset corresponding to the type of stroke to be detected; wherein, the to-be-detected image dataset includes a first to-be-detected brain image corresponding to the baseline acquisition time and an image sequence of interested blood vessels of at least one interested blood vessel; Determining the baseline lesion volume corresponding to the baseline acquisition time according to the first to-be-detected brain image; For each interested blood vessel, determining the vascular spatio-temporal feature of the interested blood vessel according to the image sequence of the interested blood vessel corresponding to the interested blood vessel; Inputting each of the vascular spatio-temporal features, the baseline lesion volume, and the volume monitoring duration into a pre-trained lesion volume prediction model to obtain the monitored lesion volume corresponding to the target monitoring time as output; wherein, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time.

2. The method according to claim 1, characterized in that, The determining the vascular spatio-temporal feature of the interested blood vessel according to the image sequence of the interested blood vessel corresponding to the interested blood vessel includes: Performing a straightening operation on at least two interested blood vessel images in the image sequence of the interested blood vessel corresponding to the interested blood vessel to obtain at least two straightened blood vessel images; Inputting each of the straightened blood vessel images into a pre-trained feature extraction model to obtain the vascular spatio-temporal feature of the interested blood vessel as output.

3. The method according to claim 2, characterized in that, The performing a straightening operation on at least two interested blood vessel images in the image sequence of the interested blood vessel corresponding to the interested blood vessel to obtain at least two straightened blood vessel images includes: For each interested blood vessel image, obtaining the center point coordinates corresponding to at least two blood vessel center points on the blood vessel center line corresponding to the interested blood vessel image; Determining the tangent vectors corresponding to the respective blood vessel center points according to the respective center point coordinates; Determining the normal plane coordinates corresponding to at least two blood vessel center points according to the respective center point coordinates and the respective tangent vectors; Determining the straightened blood vessel image corresponding to the interested blood vessel image according to the respective normal plane coordinates.

4. The method according to claim 1, characterized in that, The obtaining a to-be-detected image dataset corresponding to the type of stroke to be detected includes: Obtaining the blood vessel types of at least one interested blood vessel corresponding to the type of stroke to be detected; Performing a segmentation operation on the second to-be-detected brain image sequence according to the time decay curves corresponding to the respective brain voxels in the second to-be-detected brain image sequence to obtain a cerebrovascular image sequence corresponding to the blood vessel type; Performing a segmentation operation on the cerebrovascular image sequence according to the time decay zero-order curves corresponding to the respective cerebrovascular voxels in the cerebrovascular image sequence to obtain the image sequences of the interested blood vessels corresponding to the respective interested blood vessels; Adding the first to-be-detected brain image and the image sequences of the interested blood vessels to the to-be-detected image dataset corresponding to the type of stroke to be detected.

5. The method according to claim 4, It is characterized in that performing a segmentation operation on the second brain image sequence to be detected according to the time decay curves respectively corresponding to the brain voxels in the second brain image sequence to be detected to obtain a cerebrovascular image sequence corresponding to the blood vessel type, including: determining a curve sum sequence respectively corresponding to each of the brain voxels according to at least one time decay high-order curve respectively corresponding to the brain voxels in the second brain image sequence to be detected; taking an image sequence composed of at least one brain voxel in the second brain image sequence to be detected whose curve sum sequence meets a preset sequence condition as a cerebrovascular image sequence; determining the time to peak respectively corresponding to each of the cerebrovascular voxels according to the time decay zero-order curves respectively corresponding to the cerebrovascular voxels in the cerebrovascular image sequence; taking an image sequence composed of at least one cerebrovascular voxel in the cerebrovascular image sequence whose time to peak meets a preset time range corresponding to the blood vessel type as a cerebrovascular image sequence corresponding to the blood vessel type.

6. The method according to claim 5, It is characterized in that the method further includes: generating a cerebrovascular voxel-time to peak histogram according to the time to peak respectively corresponding to each of the cerebrovascular voxels; taking the minimum time to peak between the two peak times to peak in the cerebrovascular voxel-time to peak histogram as a preset time threshold; constructing a preset time range according to the blood vessel type and the preset time threshold; wherein, when the blood vessel type is a venous blood vessel, the minimum boundary value in the preset time range is the preset time threshold, and when the blood vessel type is an arterial blood vessel, the maximum boundary value in the preset time range is the preset time threshold.

7. The method according to claim 4, It is characterized in that performing a segmentation operation on the cerebrovascular image sequence to obtain an interested blood vessel image sequence respectively corresponding to each of the interested blood vessels according to the time decay zero-order curves respectively corresponding to the cerebrovascular voxels in the cerebrovascular image sequence, including: determining the time to peak respectively corresponding to each of the cerebrovascular voxels according to the time decay zero-order curves respectively corresponding to the cerebrovascular voxels in the cerebrovascular image sequence; acquiring a difference time respectively corresponding to each of the times to peak and the preset time threshold; for each interested blood vessel, taking an image sequence composed of at least one cerebrovascular voxel whose difference time meets a difference time range corresponding to the interested blood vessel as an interested blood vessel image sequence corresponding to the interested blood vessel.

8. The method according to claim 4, It is characterized in that the method further includes: acquiring the skull coordinates in the first frame of the original brain image in the original brain image sequence, and performing a rigid registration operation on the original brain images in the original brain image sequence except the first frame of the original brain image according to the skull coordinates to obtain a registered brain image sequence; performing a segmentation operation on the first frame of the registered brain image in the registered brain image sequence according to a preset skull threshold to obtain a skull mask image; Performing a deletion operation on each registered brain image in the registered brain image sequence according to the skull mask image to obtain a second brain image sequence to be detected.

9. A monitoring device for the volume of a stroke lesion Characterized in that It includes: A to-be-detected image dataset acquisition module, configured to acquire a to-be-detected image dataset corresponding to the to-be-detected stroke type; wherein, the to-be-detected image dataset includes a first to-be-detected brain image corresponding to the baseline acquisition time and an image sequence of an interested blood vessel of at least one interested blood vessel; A baseline lesion volume determination module, configured to determine a baseline lesion volume corresponding to the baseline acquisition time according to the first to-be-detected brain image; A blood vessel spatio-temporal feature determination module, configured to determine the spatio-temporal feature of the interested blood vessel for each interested blood vessel according to the image sequence of the interested blood vessel corresponding to the interested blood vessel; A monitoring lesion volume output module, configured to input each of the blood vessel spatio-temporal features, the baseline lesion volume, and the volume monitoring duration into a pre-trained lesion volume prediction model to obtain an output monitoring lesion volume corresponding to the target monitoring time; wherein, the volume monitoring duration represents the time length between the baseline acquisition time and the target monitoring time.

10. An electronic device Characterized in that The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for monitoring the volume of a stroke lesion according to any one of claims 1-8.

11. A computer-readable storage medium Characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the method for monitoring the volume of a stroke lesion according to any one of claims 1-8 when executed by a processor.

Images