A method, device, and procedure for assessing vascular occlusion-reperfusion therapy based on thrombus permeability.

Abstract

This application relates to the field of intelligent healthcare, specifically to a method, device, and program product for assessing vascular occlusion reperfusion therapy based on thrombus permeability. It includes acquiring brain images of patients with acute ischemic stroke; classifying the responsible vessel in the brain images to determine whether it is large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion; when the determination result is medium vessel occlusion, thrombus region localization and measurement are performed on the brain images, thrombus permeability is calculated, and reperfusion therapy is assessed based on the thrombus permeability to obtain an assessment result for intravenous thrombolysis or endovascular therapy. This application has significant clinical application value.

Description

Technical Field

[0001] This application relates to the field of intelligent healthcare, specifically to a method, device, program product, and computer-readable storage medium for evaluating vascular occlusion reperfusion therapy based on thrombus permeability. Background Technology

[0002] Acute ischemic stroke is a neurological disorder caused by acute occlusion of intracranial arteries, leading to ischemic necrosis of brain tissue and posing a high risk of disability. Reperfusion therapy (including intravenous thrombolysis and endovascular treatment) is currently one of the main treatment methods to improve neurological outcomes in patients with acute ischemic stroke. However, the efficacy and safety response to reperfusion therapy vary significantly among patients; some patients benefit significantly, while others experience limited benefit and may experience adverse events such as bleeding. Therefore, how to achieve individualized selection of reperfusion therapy strategies in the acute phase remains a key issue in clinical practice.

[0003] With the development of imaging technology, the location and diameter of vascular occlusion have gradually become important bases for stroke classification and reperfusion therapy decisions. In recent years, medium vessel occlusion (MeVO), as a type of occlusion between large vessel occlusion and distal small vessel occlusion, has gradually attracted attention. Compared with large vessel occlusion, MeVO-related vessels have smaller diameters and more complex anatomical structures, making the trade-off between the potential benefits of reperfusion therapy and bleeding / procedure-related risks more sensitive. From a clinical pathway perspective, after imaging has completed the classification of the responsible vessel and identified it as MeVO, the key decision-making focus is on whether to use only pharmacological reperfusion or to further proceed to endovascular intervention. Currently, there is a lack of imaging indicators that can provide quantitative support at the above decision-making points, leading to differences among different physicians and centers in whether to upgrade endovascular intervention, making it difficult to form repeatable and standardized decision-making criteria. In stroke imaging assessment, the correlation between thrombotic imaging characteristics and the effectiveness of reperfusion therapy has become one of the important research directions. Therefore, how to perform reproducible quantitative assessment of thrombus structural features based on conventional imaging is one of the key issues that existing technologies need to address. Summary of the Invention

[0004] Differences in thrombus composition, density, and internal structure can affect the penetration / diffusion of thrombolytic drugs within the thrombus and the recanalization effect of endovascular thrombectomy. However, existing thrombus permeability studies are mostly focused on large vessel occlusion patients, primarily used to describe thrombus imaging characteristics or assess the efficacy of single treatment modalities. Systematic application protocols for MeVO patients remain insufficient. Especially in the selection of MeVO reperfusion therapy pathways, current technologies lack an operational application process that combines quantified thrombus permeability results with pathway selection such as "drug-only reperfusion / intravascular intervention." Furthermore, MeVO vessels have small diameters and limited thrombus volume, making thrombus attenuation measurements more susceptible to variations in scanning phase, ROI placement, and operator differences, thus increasing the uncertainty in permeability measurement. Therefore, how to embed thrombus permeability assessment into the MeVO reperfusion therapy pathway selection process while ensuring measurement feasibility and consistency remains a challenge. To address these issues, this invention provides a method for assessing vascular occlusion reperfusion therapy based on thrombus permeability, specifically including:

[0005] Obtain brain images of patients with acute ischemic stroke;

[0006] The brain images are graded to determine the responsible vessel, resulting in large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion. When the determination result is medium vessel occlusion, the brain images are used to locate and measure the thrombus region, calculate thrombus permeability, and evaluate reperfusion therapy based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy.

[0007] Optionally, the thrombus permeability is obtained by calculating the attenuation difference of thrombi at the same anatomical location before and after enhancement and the attenuation difference of normal blood in the unoccluded vessel on the opposite side before and after enhancement.

[0008] Optionally, the thrombus permeability is calculated by the ratio of the attenuation difference of the thrombus at the same anatomical location before and after enhancement to the attenuation difference of normal blood in the unoccluded vessel on the opposite side before and after enhancement.

[0009] Optionally, the brain images include non-contrast CT of the head and CT angiography, and the thrombus region is located and measured by non-contrast CT of the head and CT angiography to obtain measurement results;

[0010] The thrombus permeability is obtained by first calculating the attenuation difference between non-contrast CT and CT angiography of the thrombus at the same anatomical location in the head and the attenuation difference between normal blood in the unoccluded blood vessel on the contralateral side in the head and CT angiography, and then calculating the ratio of the two.

[0011] Optionally, the thrombus permeability (TPI) is calculated as follows:

[0012] TPI = (CTA) thrombus NCCT thrombus ) / (CTA contralateral NCCT contralateral )

[0013] Among them, NCCT thrombus Non-contrast CT or CTA of the head indicating thrombosis thrombus CT angiography indicating thrombosis, NCCT contralateral This indicates a normal non-contrast CT scan of the head, CTA. contralateral This indicates a normal CT angiography.

[0014] Optionally, the method further includes image registration, thrombus localization and delineation on non-contrast CT of the head to obtain a thrombus mask, mapping the thrombus mask to CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT attenuation value of the head at the same anatomical location, and calculating the difference between the two to obtain the attenuation difference of the thrombus at the same anatomical location in non-contrast CT of the head and CT angiography.

[0015] Optionally, after thrombus localization on non-contrast CT of the head, a representative slice is selected from the proximal, middle and distal segments along the long axis of the thrombus. The representative slices are delineated and the thrombus mask is mapped onto the CT angiography of the corresponding anatomical location. The average attenuation value of the CT angiography of the three representative slices and the average attenuation value of the non-contrast CT of the head are calculated. The attenuation difference between the average attenuation value of the CT angiography and the average attenuation value of the non-contrast CT of the head is then calculated to obtain the average attenuation difference. Thrombus permeability is calculated using the average attenuation difference.

[0016] Optionally, the vascular region is located and delineated on the non-enhanced CT scan of the head on the side with normal blood, and the vascular region is mapped onto the CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-enhanced CT attenuation value of the head at the same normal anatomical location. The difference between the two is calculated to obtain the attenuation difference between the non-enhanced CT and CT angiography of normal blood in the unoccluded blood vessels on the contralateral side.

[0017] Optionally, the localization and delineation are obtained through any one or more of the following segmentation models: Mask R-CNN, YOLACT, SOLO, PolarMask, CondInst, BlendMask, HTC;

[0018] Optionally, the non-contrast CT scan of the head can be input into a segmentation model for identification and segmentation to obtain the thrombus region or vascular region;

[0019] Optionally, the training process of the segmentation model is as follows:

[0020] Obtain the non-contrast CT dataset and labels for the skull;

[0021] The non-enhanced CT dataset of the skull and its labels are input into the segmentation model to be trained until the loss function tends to stabilize, thus obtaining the segmentation model.

[0022] Optionally, the moderate vascular occlusion includes any one or more of the following occlusions: middle cerebral artery M2 segment occlusion, anterior cerebral artery A2 segment occlusion, and posterior cerebral artery P2 segment occlusion;

[0023] Optionally, the brain images are used to locate and measure the thrombus region to obtain measurement results. The type judgment result is that the middle cerebral artery M2 segment is occluded and / or the anterior cerebral artery A2 segment is occluded and / or the posterior cerebral artery P2 segment is occluded. The thrombus permeability is calculated for the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. The reperfusion treatment is evaluated based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis treatment or endovascular treatment.

[0024] Optionally, the method further includes calculating a second thrombus permeability index, and using the thrombus permeability and the second thrombus permeability index to evaluate reperfusion therapy and obtain the evaluation result of intravenous thrombolytic therapy or endovascular therapy; the second thrombus permeability index is obtained by calculating the attenuation difference between thrombi at the same anatomical location on non-contrast CT and CT angiography of the head.

[0025] Optionally, the second thrombus permeability index is calculated by the difference between the attenuation value of non-contrast CT of the head and the attenuation value of CT angiography.

[0026] The purpose of this invention is to provide a computer program product comprising a computer program or instructions, wherein the computer program or instructions are executed by a processor to implement the above-described method for evaluating vascular occlusion-reperfusion therapy based on thrombus permeability.

[0027] The purpose of this invention is to provide a computer device comprising a memory, a processor, and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-described method for evaluating vascular occlusion reperfusion therapy based on thrombus permeability.

[0028] The purpose of this invention is to provide a computer-readable storage medium having a computer program or instructions stored thereon, which is executed by a processor to implement the above-described method for evaluating vascular occlusion-reperfusion therapy based on thrombus permeability.

[0029] Advantages of this invention:

[0030] 1. This invention proposes a novel method for calculating thrombus permeability, which exhibits good robustness and repeatability. By introducing a contralateral unoccluded vessel as a reference, the influence of contrast agent injection parameters, scanning phase, and individual hemodynamic differences on the measurement results is reduced. Furthermore, this invention incorporates quantified thrombus permeability results, improving the radiographic stratification capability for selecting MeVO reperfusion therapy pathways.

[0031] 2. This invention uses thrombus permeability as a continuous variable for analysis, which helps to avoid information loss caused by simple threshold stratification and allows the relationship between imaging features and clinical outcomes to be presented in a more realistic continuous form.

[0032] 3. Compared to existing methods that primarily rely on vascular anatomy, occlusion location, or general clinical indicators for reperfusion therapy decisions, this invention introduces quantitative imaging assessment of thrombus permeability, providing additional imaging reference information for selecting treatment strategies in patients with moderate vascular occlusion (MeVO) acute ischemic stroke. Without increasing the number of examinations or imaging burden, this application can differentiate the potential efficacy differences among different reperfusion therapy modalities, thereby improving the precision of treatment decisions. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a schematic diagram of the process for evaluating vascular occlusion reperfusion therapy based on thrombus permeability, provided in an embodiment of the present invention.

[0035] Figure 2 This is a schematic diagram of a vascular occlusion reperfusion therapy assessment system based on thrombus permeability provided in an embodiment of the present invention;

[0036] Figure 3 A schematic diagram of a computer device provided in an embodiment of the present invention;

[0037] Figure 4 This is an overall technical roadmap provided for embodiments of the present invention;

[0038] Figure 5For the thrombus attenuation measurement on NCCT and CTA provided in the embodiments of the present invention, A represents NCCT thrombus attenuation measurement, and the corresponding thrombus ROI on NCCT; B represents CTA thrombus attenuation measurement, and the corresponding thrombus ROI on CTA; C represents NCCT attenuation measurement on the normal contralateral side, and the placement of the contralateral reference ROI on NCCT in an anatomically matched non-occluded artery; D represents CTA attenuation measurement on the normal contralateral side, and the corresponding contralateral reference ROI on CTA.

[0039] Figure 6 This invention provides a treatment-specific association between thrombus permeability and 90-day functional independence. Detailed Implementation

[0040] To enable those skilled in the art to better understand 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.

[0041] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as S101, S102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0042] Figure 1 A schematic diagram of the vascular occlusion reperfusion therapy assessment method based on thrombus permeability provided in this embodiment of the invention is shown, specifically including:

[0043] S1: Obtain brain images of patients with acute ischemic stroke;

[0044] In one embodiment, the brain imaging includes non-contrast CT of the head and CT angiography, and the thrombus region is located and measured by non-contrast CT of the head and CT angiography to obtain measurement results.

[0045] In one specific embodiment, the method described in this invention is applicable to patients with acute ischemic stroke whose onset time to hospital admission is no more than 24 hours and whose intracranial moderate vascular occlusion is confirmed by imaging. However, this method is not applicable to situations where the image quality cannot meet basic measurement requirements or the occluded segment of the responsible vessel cannot be clearly identified. All patients underwent routine NCCT and CTA examinations upon admission, and the image data came from standard clinical examination procedures, requiring no additional scans or special parameter settings. During the image preprocessing stage, the original NCCT and CTA images were uniformly spatially registered and resampled to isovoxel resolution to ensure the consistency of spatial correspondence between different modalities and the stability of subsequent density measurements.

[0046] Compared to large vessel occlusion, intermediate-sized vessel occlusion involves smaller causative vessel diameters and shorter thrombus lengths, resulting in more pronounced radiographic heterogeneity of the thrombus in both the axial and radial directions. In this context, directly applying the single-slice or single-point measurement methods commonly used in large vessel occlusion studies makes it easier for thrombus attenuation characteristics to be affected by partial volume effects, local anatomical variations, and differences in measurement location, making it difficult to stably reflect the overall structural characteristics of the thrombus. Furthermore, the potential benefit-risk boundary of reperfusion therapy for intermediate-sized vessel occlusion is more sensitive, placing higher demands on the stability and reproducibility of radiographic indicators. Based on these radiographic characteristics of intermediate-sized vessel occlusion, this invention specifically designs thrombus localization, sampling methods, and quantitative strategies in the measurement and application of thrombus permeability to reduce local measurement errors and improve the usability of this indicator in treatment decision-making scenarios.

[0047] S2: The brain images are graded to determine the responsible vessel, resulting in large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion. When the determination result is medium vessel occlusion, the brain images are used to locate and measure the thrombus region, calculate the thrombus permeability, and evaluate reperfusion therapy based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy.

[0048] In one embodiment, the method further includes image registration, performing thrombus localization and delineation on non-contrast CT of the head to obtain a thrombus mask, mapping the thrombus mask to CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT attenuation value of the head at the same anatomical location, and calculating the difference between the two to obtain the attenuation difference of the thrombus at the same anatomical location in non-contrast CT of the head and CT angiography.

[0049] In one embodiment, after thrombus localization using non-contrast CT of the head, a representative slice is selected from the proximal, mid-segment, and distal ends along the long axis of the thrombus. The representative slices are delineated, and the thrombus mask is mapped onto the CT angiography of the corresponding anatomical location. The average attenuation value of the CT angiography of the three representative slices and the average attenuation value of the non-contrast CT of the head are calculated. The attenuation difference between the average attenuation value of the CT angiography and the average attenuation value of the non-contrast CT of the head are then calculated to obtain the average attenuation difference. Thrombus permeability is calculated using the average attenuation difference.

[0050] Thrombus permeability is calculated by comparing the average attenuation difference on the thrombus side with the average attenuation difference on the normal side.

[0051] In one embodiment, the vascular region is located and delineated on the non-enhanced CT scan of the head on the side with normal blood, and the vascular region is mapped onto the CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-enhanced CT attenuation value of the head at the same normal anatomical location. The difference between the two is calculated to obtain the attenuation difference between the non-enhanced CT and CT angiography of normal blood in the unoccluded blood vessels on the contralateral side.

[0052] In one embodiment, after locating the corresponding blood vessel on the other side of the thrombus using a non-contrast CT scan of the head, a representative slice is selected from the proximal, mid-segment, and distal ends of the blood vessel along its long axis. The representative slices are delineated, and the blood vessel mask is mapped onto the CT angiography of the corresponding anatomical location. The average attenuation value of the CT angiography of the three representative slices and the average attenuation value of the non-contrast CT scan of the head are calculated. Then, the attenuation difference between the average attenuation value of the CT angiography and the average attenuation value of the non-contrast CT scan of the head is calculated to obtain the average attenuation difference of the blood vessel.

[0053] In one embodiment, the localization and delineation are obtained through any one or more of the following segmentation models: Mask R-CNN, YOLACT, SOLO, PolarMask, CondInst, BlendMask, HTC;

[0054] Optionally, the non-contrast CT scan of the head can be input into a segmentation model for identification and segmentation to obtain the thrombus region or vascular region.

[0055] In one embodiment, the training process of the segmentation model is as follows:

[0056] Obtain the non-contrast CT dataset and labels for the skull;

[0057] The non-enhanced CT dataset of the skull and its labels are input into the segmentation model to be trained until the loss function tends to stabilize, thus obtaining the segmentation model.

[0058] The labels include category labels and pixel-level masks.

[0059] In one embodiment, the image registration process further includes image verification, which involves segmenting the thrombus or blood vessel in the mapped CT angiography and back-projecting it onto a non-enhanced CT scan of the head. The spatial overlap between the back-projection and the original non-enhanced CT scan of the head is compared. If the overlap is low, the registration is re-performed; otherwise, the attenuation value is calculated.

[0060] In one embodiment, a segmentation model is used to identify and segment high-density thrombus regions and / or vascular pathways from non-enhanced CT scans of the head. A second segmentation model is used to identify and segment vascular images from CT angiography to obtain vascular lumens and / or permeability regions. In the image registration process, the mapping first matches the vessel wall or thrombus boundary to avoid interference from surrounding brain tissue changes, calculating vascular structure similarity. Then, high-density thrombus regions and permeability region registration is performed, calculating thrombus feature similarity. Simultaneously, the grayscale similarity of the segmented masked image is calculated. The final similarity is obtained by fusing grayscale similarity, vascular structure similarity, and thrombus feature similarity, and the final similarity is used to complete accurate mapping. Non-enhanced CT scans of the head are deformed into CT angiography, forcing the alignment of vascular structures and thrombus feature regions while maintaining overall grayscale similarity. The CT value change at the same location is calculated to obtain accurate permeability values.

[0061] The registration process becomes more interpretable by incorporating medical features, making it easier for doctors to understand and trust the algorithm's results. Furthermore, by emphasizing the matching of key medical features, it effectively suppresses registration noise from irrelevant tissues (such as brain tissue that moves slightly with breathing). This registration method provides an important foundation for accurately extracting thrombus permeability features.

[0062] The second segmentation model is the same as the previous segmentation model, and it is trained using CTA images and labels (CTA pixel mask and category label) during training.

[0063] In another embodiment, the method further includes prediction, inputting non-enhanced CT and CT angiography of the head in brain images into a two-branch neural network for feature extraction to obtain a first NCCT feature map and a first CTA feature map, inputting the first NCCT feature map and the first CTA feature map into a differentiable spatial transformation network for feature alignment, outputting a deformation field constrained by smoothness, the deformation field being used to transform the first CTA feature map to obtain a second CTA feature map, fusing the first NCCT feature map and the second CTA feature map and inputting it along with thrombus permeability into a prediction and evaluation module to obtain an evaluation result for intravenous thrombolysis or endovascular treatment.

[0064] Among them, the dual-branch neural network is a dual-branch 3D convolutional neural network (including any of the following variants of 3DResNet and ViT). The differentiable spatial transformation network DSTN learns a spatial transformation parameter (such as affine or thin-plate spline transformation) from the first feature map of NCCT to the feature space of the first feature map of CTA.

[0065] The DSTN consists of a localization network, a mesh generator, and a sampler. The localization network is a fully connected layer. It receives the first feature map of NCCT and the first feature map of CTA and performs affine calculations to obtain transformation parameters, generating a deformation field. The mesh generator uses the transformation parameters in the deformation field and the coordinates of each pixel in the first feature map of CTA to calculate the sampling coordinates of the pixel on the input feature map. It then extracts pixel values ​​from the input feature map using a differentiable sampling method based on the sampling coordinates calculated by the mesh generator, generating the transformed second feature map of CTA.

[0066] How it works: DSTN automatically calculates the optimal transformation to align the two feature maps based on their contents. Then, it transforms the first feature map of CTA to obtain a new feature map spatially aligned with the first feature map of NCCT. The entire process is differentiable, meaning that gradients can be backpropagated, and the network can learn "how to align features to optimize the final prediction target."

[0067] Training data and annotations:

[0068] Input: A large number of paired plain CT-CTA images; task label: the clinical outcome corresponding to each case; loss function is the predicted cross-entropy loss function.

[0069] In one specific embodiment, during thrombus identification and quantification, the occluded segment of the responsible vessel clearly shown in the CTA image is used as the basis for thrombus localization. At least two physicians with experience in interpreting neuroimaging images manually delineate the thrombus region on the multiplanar reconstructed image, and the consistent region is selected as the final thrombus mask. (Alternatively, a segmentation model is trained based on the final thrombus mask to form a segmentation model). This thrombus mask is mapped to the registered NCCT space to obtain the attenuation values ​​of the same anatomical location under different modalities.

[0070] To reduce the partial volume effect and improve measurement repeatability, three representative slices—proximal, mid-segment, and distal—were selected along the long axis of the thrombus. A region of interest (ROI, 1 mm²) was placed in the central region of the thrombus, and the average attenuation value of the thrombus in NCCT and CTA images was measured simultaneously. Simultaneously, a corresponding ROI was placed in the lumen of the contralateral, non-occluded artery, and the attenuation value of normal blood under NCCT and CTA was measured as a reference.

[0071] In one embodiment, the thrombus permeability is obtained by calculating the attenuation change of thrombus at the same anatomical location before and after enhancement and the attenuation change of normal blood in the contralateral unoccluded vessel before and after enhancement.

[0072] Optionally, the thrombus permeability is calculated by the ratio of the attenuation change of the thrombus at the same anatomical location before and after enhancement to the attenuation change of normal blood in the unoccluded vessel on the opposite side before and after enhancement.

[0073] In one embodiment, the thrombus permeability is obtained by first calculating the attenuation difference between non-contrast CT and CT angiography of the thrombus at the same anatomical location and the attenuation difference between non-contrast CT and CT angiography of normal blood in the contralateral unoccluded vessel, and then calculating the ratio of the two.

[0074] In one embodiment, the thrombus permeability (TPI) is calculated as follows:

[0075] TPI = (CTAthrombus) NCCTthrombus) / (CTAcontralateral NCCTcontralateral)

[0076] Among them, NCCTthrombus means non-contrast CT of the head with thrombosis, CTAthrombus means CT angiography with thrombosis, NCCTcontralateral means normal non-contrast CT of the head, and CTAcontralateral means normal CT angiography.

[0077] In one embodiment, the moderate vascular occlusion includes any one or more of the following occlusions: middle cerebral artery M2 segment occlusion, anterior cerebral artery A2 segment occlusion, and posterior cerebral artery P2 segment occlusion.

[0078] Optionally, the brain images are used to locate and measure the thrombus region to obtain measurement results. The type judgment result is that the middle cerebral artery M2 segment is occluded and / or the anterior cerebral artery A2 segment is occluded and / or the posterior cerebral artery P2 segment is occluded. The thrombus permeability is calculated for the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. The reperfusion treatment is evaluated based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis treatment or endovascular treatment.

[0079] In one embodiment, the method further includes calculating a second thrombus permeability index, and using the thrombus permeability and the second thrombus permeability index to evaluate reperfusion therapy and obtain the evaluation result of intravenous thrombolytic therapy or endovascular therapy; the second thrombus permeability index is obtained by calculating the attenuation difference between thrombi at the same anatomical location on non-contrast CT and CT angiography of the head.

[0080] Optionally, the second thrombus permeability index is calculated by the difference between the attenuation value of non-contrast CT of the head and the attenuation value of CT angiography.

[0081] In one specific embodiment, based on the above measurement results, the present invention calculates the thrombus perviousness index (TPI), which is defined as: TPI = (CTA_thrombus NCCT_thrombus) / (CTA_contralateral NCCT_contralateral), where all differences are expressed in Hounsfield units (HU). The numerator of TPI (CTA_thrombus) is... NCCT_thrombus represents the attenuation change of a thrombus at the same anatomical location before and after contrast enhancement, reflecting the absolute degree of contrast agent penetration into the thrombus. This indicator is mainly affected by factors such as the microstructure, density, and fibrous network arrangement within the thrombus, and can characterize the thrombus's ability to restrict the penetration of contrast agents, drugs, or blood flow, serving as a direct imaging indicator for characterizing thrombus permeability. The denominator of TPI (CTA_contralateral) NCCT_contralateral indicates the attenuation of normal blood in the contralateral non-occluded vessel before and after enhancement, reflecting the overall vascular enhancement level of the patient under the current scanning conditions. This index comprehensively reflects non-thrombosis-related factors such as contrast agent injection protocol, scanning phase, and individual hemodynamic status. The enhancement level that the contrast agent can achieve in normal vessels can be regarded as the "reference enhancement level" under the current imaging conditions. TPI is used to reflect the degree of contrast agent penetration into the thrombus and the relative looseness of the thrombus microstructure; a higher value indicates higher thrombus permeability.

[0082] In one specific embodiment, this invention provides an application method based on thrombus permeability imaging characteristics to assist in the selection of reperfusion therapy pathways for patients with medium vessel occlusion (MeVO) acute ischemic stroke. This method, based on conventional emergency non-contrast CT (NCCT) and CT angiography (CTA) images, quantitatively calculates thrombus permeability without requiring additional examinations. The quantitative results are used to output relative indication information under different reperfusion therapy pathways, thereby providing imaging references for individualized treatment of MeVO patients. In terms of the overall process, this method incorporates the "vascular classification—treatment pathway selection" decision framework for stroke diagnosis and treatment. Clinically, the anatomical level of the responsible vessel is first determined based on emergency imaging. After classifying it as large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion, the corresponding level of reperfusion therapy assessment process is initiated. When the classification is determined to be MeVO, the thrombus permeability assessment process is executed to assist in determining whether only pharmacological reperfusion is used or whether endovascular intervention is required. Thrombus permeability quantification results, serving as auxiliary decision-making information within the MeVO level, are preferably used as input along with basic clinical information and routine imaging findings to generate auxiliary indicators for reperfusion therapy pathway selection. The treatment pathway characterizes whether to proceed with endovascular intervention; specific drug types and interventional procedures are not limitations of this method.

[0083] This method is applicable to meVO occlusions including, but not limited to, the M2 segment of the middle cerebral artery, the A2 segment of the anterior cerebral artery, and the P2 segment of the posterior cerebral artery, as well as other medium-diameter intracranial artery occlusions. The overall technical approach is illustrated below. Figure 4 As shown.

[0084] In one specific embodiment, to verify the robustness of the results, the present invention also calculates TAI as a substitute indicator, TAI = CTA_thrombus. NCCT_thrombus. It should be noted that in moderate vascular occlusion scenarios, due to the smaller diameter of the responsible vessel and the limited thrombus volume, the thrombus attenuation value in CTA images is more easily affected by fluctuations in overall vascular enhancement. The proportion of non-thrombus factors in TAI is relatively increased. Therefore, in this invention, simultaneous analysis using TAI and TPI helps to distinguish the impact of differences in thrombus permeability and overall enhancement on the measurement results. Thrombus permeability measurement procedure and ROI placement illustration. Figure 5 As shown.

[0085] In one specific embodiment, from a technical implementation perspective, the application method of this invention has a clear input, processing, and output structure, enabling standardized implementation in clinical settings. The input information includes routine NCCT and CTA imaging data from patients with acute ischemic stroke, as well as information on the responsible vessel for moderate vascular occlusion identified by imaging. The imaging data originates from current clinical examination procedures, requiring no additional imaging sequences or special scanning parameters. During processing, based on uniformly spatially registered NCCT and CTA images, the thrombus region within the occluded segment of the responsible vessel is standardized for localization and quantitative measurement. Thrombus permeability-related indicators are calculated, and robustness assessments can be performed using normalization or alternative indicators as needed. The processing flow revolves around the imaging data itself, without relying on complex modeling or external parameter settings, and features clear operational steps and repeatability. At the output level, this invention does not use a single numerical value as the endpoint, but rather uses the thrombus permeability assessment results as auxiliary decision-making information within the moderate vascular occlusion level, indicating the relative benefit trend under different reperfusion therapy pathways. The output results can be presented through an imaging workstation, decision support interface, or report, providing clinicians with imaging references at the critical juncture of whether to upgrade to endovascular intervention, thereby elevating the invention from an imaging discovery to an application technology solution that can be implemented in actual diagnosis and treatment processes.

[0086] Compared to large vessel occlusion, MeVO patients lack unified quantitative imaging evidence for reperfusion therapy selection, resulting in insufficient consistency in treatment decisions. Previous studies on thrombus permeability primarily focused on large vessel occlusion patients, with conclusions often used to explain differences in the efficacy of single treatment modalities or as descriptive predictors of recanalization rates and prognostic outcomes. Due to differences in vessel diameter, thrombus burden, and measurement stability between MeVO and large vessel occlusion, previous conclusions are difficult to directly guide treatment pathway selection for MeVO, and the application of thrombus permeability in this scenario remains unclear. This invention addresses this critical gap in clinical practice by proposing the introduction of thrombus permeability imaging assessment into the reperfusion therapy decision-making process for moderate vessel occlusion, to characterize the relative indicative significance of thrombus permeability under different reperfusion therapy pathways. By systematically analyzing the interaction between thrombus permeability and treatment modality in a MeVO population, this invention demonstrates that in the MeVO scenario, relying solely on treatment modality or a single thrombus permeability index is insufficient to form a stable decision. However, using thrombus permeability as a modifier for treatment pathway selection helps differentiate the relative indicative direction of different reperfusion pathways. Therefore, this method, through standardized thrombus permeability quantification and output rules, uses permeability indices as auxiliary prompts for MeVO treatment pathway selection, forming an application scheme that can be embedded into existing processes. This transforms thrombus permeability from a research-based observational indicator into an auxiliary decision-making tool supporting reperfusion treatment pathway selection, thereby addressing the practical problem of insufficient imaging stratification in MeVO reperfusion treatment decision-making.

[0087] At the level of application in treatment decision-making, this invention does not merely use thrombus permeability as a single outcome predictor, but rather as a modifier of treatment effect, evaluating its differentiated impact on functional outcomes under different reperfusion strategies. Based on data from 255 patients with radiographically confirmed MeVO acute ischemic stroke, of whom 154 received intravenous thrombolysis (IVT) and 101 received endovascular treatment (EVT, including bridging therapy), the analysis showed that in the overall population, there was no significant difference between IVT and EVT in terms of 90-day functional independence rate (modified Rankin Scale mRS≤2) (51.9% vs 51.5%, P=0.96). Thrombus permeability itself also did not show a stable association with 90-day functional outcomes under unstratified conditions.

[0088] The above results suggest that in patients with moderate vascular occlusion, neither comparing reperfusion therapy methods alone nor relying solely on thrombus permeability as an imaging indicator is sufficient to form a stable judgment that can be directly used for clinical stratification. Therefore, it is necessary to further examine the relative significance of thrombus permeability under different reperfusion therapy strategies. It should be noted that previous studies on large vessel occlusion have extensively discussed the relationship between thrombus permeability and the efficacy of intravenous thrombolysis and endovascular therapy. The general understanding is that intravenous thrombolysis is more beneficial when permeability is high, while endovascular therapy may be more advantageous when permeability is low. However, the formation of the above conclusions depends on the specific anatomical structure, thrombus burden, and treatment context of large vessel occlusion. For moderate vascular occlusion, the vessel diameter is smaller, the thrombus length is shorter, and imaging measurements are more easily affected by partial volume effects and local anatomical variations. At the same time, the risk-benefit boundary of reperfusion therapy is more sensitive, making it difficult to directly translate past experience into actionable decision-making criteria. Based on this, the present invention employs a standardized method for quantifying thrombus permeability in moderate vascular occlusion scenarios and combines it with specific reperfusion therapy methods for analysis, in order to characterize the relative indicative significance of thrombus permeability in different treatment strategies, thereby providing an imaging-level reference for the selection of reperfusion therapy strategies.

[0089] After incorporating TPI as a continuous variable into the treatment modality interaction model, we found that thrombotic permeability significantly modified the efficacy of different reperfusion therapies. In a multivariate logistic regression model based on propensity score inverse probability weighting (IPTW), the interaction term between TPI and treatment modality was statistically significant (OR 0.37, 95% CI 0.24–0.56, P < 0.001). Further stratified analysis showed that in the IVT group, for every 0.05 increase in TPI, the probability of functional independence at 90 days increased (OR 1.73, 95% CI 1.19–2.53, P = 0.005); while in the EVT group, increased TPI was associated with a decreased probability of a good prognosis at 90 days (OR 0.65, 95% CI 0.49–0.87, P = 0.003). Sensitivity analyses of different weighting methods and alternative permeability indicators (TAI) were consistent with these conclusions, suggesting that this effect has good robustness (as shown in Table 1).

[0090] To visually demonstrate the role of thrombus permeability under different treatment strategies, this invention further employs a restricted cubic spline model to describe the continuous relationship between TPI and the 90-day functionally independent probability. The results show that in IVT patients, the probability of a good prognosis continuously increases with increasing TPI, and tends to plateau in the high permeability range; while in EVT patients, the opposite trend is observed. Figure 6As shown. Based on the above results, this invention proposes a new clinical application pathway: in patients with MeVO acute ischemic stroke, the thrombus permeability index is calculated using emergency CT / CTA images to help differentiate between patients more likely to benefit from intravenous thrombolysis or endovascular treatment. For patients with high thrombus permeability, intravenous thrombolysis may achieve better functional outcomes; while in patients with low thrombus permeability, endovascular treatment may have a relative advantage. This application method requires no additional invasive procedures, does not rely on complex equipment or time-consuming analysis, and can be directly embedded into existing emergency imaging workflows, providing objective and repeatable imaging decision support for reperfusion treatment strategies in MeVO patients. In practical applications, the thrombus permeability analysis results can be presented through imaging workstations, decision support interfaces, or reports, providing clinicians with relative benefit indications or risk stratification information for different reperfusion treatment strategies.

[0091] Table 1 shows the treatment-specific association between thrombus permeability index and 90-day functional independence (mRS≤2) using the IPTW method.

[0092]

[0093] In one specific embodiment, one of the technical effects of this invention lies in improving the imaging stratification capability for MeVO reperfusion therapy selection by introducing quantified thrombus permeability results. Without considering thrombus permeability, intravenous thrombolysis and endovascular therapy did not show a significant difference in 90-day functional independence rates, suggesting that relying solely on the treatment method itself is insufficient for effective stratification. However, when the thrombus permeability index was introduced into the analytical model and its interaction with the treatment method was examined, it was observed that the direction of the influence of thrombus permeability on functional outcomes changed significantly with the treatment method. This result indicates that thrombus permeability can serve as a modifier of treatment effects, used to identify patient groups that may have different benefit characteristics under different reperfusion strategies.

[0094] Secondly, this invention employs an analysis approach that treats thrombus permeability as a continuous variable, which helps avoid information loss caused by simple threshold stratification and allows the relationship between imaging features and clinical outcomes to be presented in a more realistic, continuous form. Restricted cubic spline analysis showed that in patients receiving intravenous thrombolysis, increased thrombus permeability was associated with an increased probability of a good functional outcome, while the opposite trend was observed in patients receiving endovascular treatment. These results suggest that the differences in thrombus microstructure reflected by thrombus permeability may play different roles in different reperfusion therapy pathways, and the application scheme of this invention can capture and reflect this difference.

[0095] At the methodological level, the thrombus permeability calculation method used in this invention exhibits good robustness and repeatability. By introducing the contralateral unoccluded vessel as a reference for normalization, the influence of contrast agent injection parameters, scanning phase, and individual hemodynamic differences on the measurement results can be reduced to some extent. Furthermore, when using the uncorrected increase in thrombus attenuation as a surrogate indicator for sensitivity analysis, the interaction effect between thrombus permeability and treatment modality remains consistent, further supporting the stability of this application. Combined with the sensitivity analysis results obtained by simultaneously using TAI as a surrogate indicator, it can be seen that normalization does not change the basic correlation between thrombus permeability and reperfusion treatment outcomes, but rather improves the stability and repeatability of this indicator in treatment decision-making applications, especially in the application scenario of moderate vessel occlusion, which is more sensitive to imaging measurement errors.

[0096] From a clinical implementation perspective, this invention demonstrates good feasibility. The required imaging data are all derived from routine NCCT and CTA examinations of patients with acute ischemic stroke, requiring no additional imaging sequences or complex post-processing procedures. Thrombus permeability measurement is based on a limited number of standardized ROI settings, making the operation relatively simple and suitable for application in stroke emergency settings or retrospective decision support. Therefore, this application provides a valuable supplementary imaging decision-making tool for MeVO patients without increasing patient burden or medical costs.

[0097] In summary, this invention addresses the lack of imaging stratification in existing decision-making models by incorporating thrombus permeability imaging parameters into the decision-making process for reperfusion therapy in patients with moderate vascular occlusion. This allows for the differentiation of potential benefits from different treatment approaches and provides feasible and clinically significant technical support for individualized reperfusion therapy in MeVO patients.

[0098] The present invention also discloses a computer program product or system, including a computer program that, when executed by a processor, implements the above-described method steps.

[0099] Figure 2 A schematic diagram of the vascular occlusion reperfusion therapy assessment system based on thrombus permeability provided in this embodiment of the invention is shown, specifically including:

[0100] Acquisition Unit: Acquires brain images of patients with acute ischemic stroke;

[0101] Judgment and evaluation unit: The brain images are graded to determine the responsible vessel, resulting in large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion. When the type determination result is medium vessel occlusion, the brain images are used to locate and measure the thrombus region, calculate thrombus permeability, and evaluate reperfusion therapy based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy.

[0102] Figure 3 An embodiment of the present invention provides a schematic diagram of a computer device, specifically including:

[0103] A memory and a processor; the memory is used to store program instructions; the processor is used to invoke the program instructions, when any of the above-described methods for evaluating vascular occlusion-reperfusion therapy based on thrombus permeability are executed.

[0104] The present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, provides any of the above-described methods for evaluating vascular occlusion-reperfusion therapy based on thrombus permeability.

[0105] The verification results of this verification embodiment show that assigning inherent weights to indications can improve the performance of this method compared to the default settings. Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be electrical, mechanical, or other forms. The units described as separate components may or may not be physically separated; the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of this embodiment. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0106] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0107] The computer device provided by the present invention has been described in detail above. For those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A vascular occlusion reperfusion therapy assessment system based on thrombus permeability, characterized in that, include: Acquisition Unit: Acquires brain images of patients with acute ischemic stroke; Judgment and Assessment Unit: The brain images are graded to determine the responsible vessel, resulting in large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion. When the judgment result is medium vessel occlusion, the brain images are used to locate and measure the thrombus region, calculate thrombus permeability, and evaluate reperfusion therapy based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy. The thrombus permeability is calculated by the ratio of the attenuation difference of the thrombus at the same anatomical location before and after enhancement to the attenuation difference of normal blood in the contralateral unoccluded vessel before and after enhancement.

2. The vascular occlusion reperfusion therapy assessment system based on thrombus permeability according to claim 1, characterized in that, The brain images include non-contrast CT of the head and CT angiography. The thrombus region is located and measured by non-contrast CT of the head and CT angiography to obtain the measurement results. The thrombus permeability is obtained by first calculating the attenuation difference between non-contrast CT and CT angiography of the thrombus at the same anatomical location in the head, and the attenuation difference between normal blood in the unoccluded blood vessel on the contralateral side in non-contrast CT and CT angiography of the head, and then calculating the ratio of the two.

3. The vascular occlusion reperfusion therapy assessment system based on thrombus permeability according to claim 2, characterized in that, It also includes an image registration step, which involves spatially registering the non-contrast CT scan of the head with CT angiography; locating and delineating the thrombus on the registered non-contrast CT scan of the head to obtain a thrombus mask; mapping the thrombus mask onto the CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT scan attenuation value at the same anatomical location, and calculating the difference between the two to obtain the attenuation difference of the thrombus at the same anatomical location in the non-contrast CT scan of the head and the CT angiography.

4. The vascular occlusion reperfusion therapy assessment system based on thrombus permeability according to claim 3, characterized in that, After thrombus localization on non-contrast CT of the head, a representative slice is selected from the proximal, middle and distal ends of the thrombus along its long axis. Regions of interest are set on each slice, and the thrombus mask is mapped onto the CT angiography of the corresponding anatomical location. The average attenuation value of the CT angiography of the three representative slices and the average attenuation value of the non-contrast CT of the head are calculated. The attenuation difference between the average attenuation value of the CT angiography and the average attenuation value of the non-contrast CT of the head is then calculated to obtain the average attenuation difference. Thrombus permeability is calculated using the average attenuation difference. The vascular region was located and delineated on the non-contrast CT scan of the head on the side with normal blood. The vascular region was then mapped onto the CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT attenuation value of the head at the same normal anatomical location. The difference between the two was calculated to obtain the attenuation difference between the non-contrast CT and CT angiography of normal blood in the unoccluded blood vessels on the contralateral side.

5. The vascular occlusion reperfusion therapy assessment system based on thrombus permeability according to claim 1, characterized in that, The moderate vascular occlusion includes any one or more of the following occlusions: middle cerebral artery M2 segment occlusion, anterior cerebral artery A2 segment occlusion, and posterior cerebral artery P2 segment occlusion. The brain images were used to classify the responsible vessel, and the type of occlusion was determined to be the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. Thrombus permeability was calculated for the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. The reperfusion therapy was evaluated based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy.

6. The vascular occlusion reperfusion therapy assessment system based on thrombus permeability according to claim 1, characterized in that, It also includes the calculation of a second thrombus permeability index, and the evaluation results of intravenous thrombolysis or endovascular treatment are obtained by evaluating reperfusion therapy through the thrombus permeability and the second thrombus permeability index; the second thrombus permeability index is obtained by calculating the attenuation difference of thrombi at the same anatomical location on non-enhanced CT and CT angiography of the head. The second thrombus permeability index is calculated by the difference between the attenuation value of non-contrast CT of the head and the attenuation value of CT angiography.

7. A computer device comprising a memory, a processor, and a computer program or instructions stored in the memory, characterized in that, The computer program or instructions are executed by a processor to implement a method for evaluating vascular occlusion-reperfusion therapy based on thrombus permeability, including: Obtain brain images of patients with acute ischemic stroke; The brain images are graded to determine the responsible vessel, classifying it as large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion. When the determination is medium vessel occlusion, the brain images are used to locate and measure the thrombus region, and thrombus permeability is calculated. The thrombus permeability is used to evaluate reperfusion therapy, obtaining the evaluation result for intravenous thrombolysis or endovascular therapy. The thrombus permeability is calculated as the ratio of the attenuation difference of the thrombus at the same anatomical location before and after enhancement to the attenuation difference of normal blood in the contralateral unoccluded vessel before and after enhancement.

8. The computer device according to claim 7, characterized in that, The brain images include non-contrast CT of the head and CT angiography. The thrombus region is located and measured by non-contrast CT of the head and CT angiography to obtain the measurement results. The thrombus permeability is obtained by first calculating the attenuation difference between non-contrast CT and CT angiography of the thrombus at the same anatomical location in the head, and the attenuation difference between normal blood in the unoccluded blood vessel on the contralateral side in non-contrast CT and CT angiography of the head, and then calculating the ratio of the two.

9. The computer device according to claim 8, characterized in that, The method further includes an image registration step, which involves spatially registering the non-contrast CT scan of the head with CT angiography; locating and delineating the thrombus on the registered non-contrast CT scan of the head to obtain a thrombus mask; mapping the thrombus mask onto CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT scan attenuation value at the same anatomical location, and calculating the difference between the two to obtain the attenuation difference of the thrombus at the same anatomical location in the non-contrast CT scan of the head and the CT angiography.

10. The computer device according to claim 9, characterized in that, After thrombus localization on non-contrast CT of the head, a representative slice is selected from the proximal, middle and distal ends of the thrombus along its long axis. Regions of interest are set on each slice, and the thrombus mask is mapped onto the CT angiography of the corresponding anatomical location. The average attenuation value of the CT angiography of the three representative slices and the average attenuation value of the non-contrast CT of the head are calculated. The attenuation difference between the average attenuation value of the CT angiography and the average attenuation value of the non-contrast CT of the head is then calculated to obtain the average attenuation difference. Thrombus permeability is calculated using the average attenuation difference. The vascular region was located and delineated on the non-contrast CT scan of the head on the side with normal blood. The vascular region was then mapped onto the CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT attenuation value of the head at the same normal anatomical location. The difference between the two was calculated to obtain the attenuation difference between the non-contrast CT and CT angiography of normal blood in the unoccluded blood vessels on the contralateral side.

11. The computer device according to claim 7, characterized in that, The moderate vascular occlusion includes any one or more of the following occlusions: middle cerebral artery M2 segment occlusion, anterior cerebral artery A2 segment occlusion, and posterior cerebral artery P2 segment occlusion. The brain images were used to classify the responsible vessel, and the type of occlusion was determined to be the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. Thrombus permeability was calculated for the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. The reperfusion therapy was evaluated based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy.

12. The computer device according to claim 7, characterized in that, The method also includes the calculation of a second thrombus permeability index, and the evaluation results of intravenous thrombolysis or endovascular treatment are obtained by evaluating reperfusion therapy through the thrombus permeability and the second thrombus permeability index; the second thrombus permeability index is obtained by calculating the attenuation difference of thrombi at the same anatomical location on non-contrast CT and CT angiography of the head. The second thrombus permeability index is calculated by the difference between the attenuation value of non-contrast CT of the head and the attenuation value of CT angiography.

13. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, The computer program or instructions are executed by a processor to implement a method for evaluating vascular occlusion-reperfusion therapy based on thrombus permeability, the method comprising: Obtain brain images of patients with acute ischemic stroke; The brain images are graded to determine the responsible vessel, classifying it as large vessel occlusion, medium vessel occlusion, or distal small vessel occlusion. When the determination is medium vessel occlusion, the brain images are used to locate and measure the thrombus region, and thrombus permeability is calculated. The thrombus permeability is used to evaluate reperfusion therapy, obtaining the evaluation result for intravenous thrombolysis or endovascular therapy. The thrombus permeability is calculated as the ratio of the attenuation difference of the thrombus at the same anatomical location before and after enhancement to the attenuation difference of normal blood in the contralateral unoccluded vessel before and after enhancement.

14. The computer-readable storage medium according to claim 13, characterized in that, The brain images include non-contrast CT of the head and CT angiography. The thrombus region is located and measured by non-contrast CT of the head and CT angiography to obtain the measurement results. The thrombus permeability is obtained by first calculating the attenuation difference between non-contrast CT and CT angiography of the thrombus at the same anatomical location in the head, and the attenuation difference between normal blood in the unoccluded blood vessel on the contralateral side in non-contrast CT and CT angiography of the head, and then calculating the ratio of the two.

15. The computer-readable storage medium according to claim 14, characterized in that, The method further includes an image registration step, which involves spatially registering the non-contrast CT scan of the head with CT angiography; locating and delineating the thrombus on the registered non-contrast CT scan of the head to obtain a thrombus mask; mapping the thrombus mask onto CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT scan attenuation value at the same anatomical location, and calculating the difference between the two to obtain the attenuation difference of the thrombus at the same anatomical location in the non-contrast CT scan of the head and the CT angiography.

16. The computer-readable storage medium according to claim 15, characterized in that, After thrombus localization on non-contrast CT of the head, a representative slice is selected from the proximal, middle and distal ends of the thrombus along its long axis. Regions of interest are set on each slice, and the thrombus mask is mapped onto the CT angiography of the corresponding anatomical location. The average attenuation value of the CT angiography of the three representative slices and the average attenuation value of the non-contrast CT of the head are calculated. The attenuation difference between the average attenuation value of the CT angiography and the average attenuation value of the non-contrast CT of the head is then calculated to obtain the average attenuation difference. Thrombus permeability is calculated using the average attenuation difference. The vascular region was located and delineated on the non-contrast CT scan of the head on the side with normal blood. The vascular region was then mapped onto the CT angiography at the same anatomical location to obtain the CT angiography attenuation value and the non-contrast CT attenuation value of the head at the same normal anatomical location. The difference between the two was calculated to obtain the attenuation difference between the non-contrast CT and CT angiography of normal blood in the unoccluded blood vessels on the contralateral side.

17. The computer-readable storage medium according to claim 13, characterized in that, The moderate vascular occlusion includes any one or more of the following occlusions: middle cerebral artery M2 segment occlusion, anterior cerebral artery A2 segment occlusion, and posterior cerebral artery P2 segment occlusion. The brain images were used to classify the responsible vessel, and the type of occlusion was determined to be the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. Thrombus permeability was calculated for the middle cerebral artery M2 segment occlusion and / or the anterior cerebral artery A2 segment occlusion and / or the posterior cerebral artery P2 segment occlusion. The reperfusion therapy was evaluated based on the thrombus permeability to obtain the evaluation result of intravenous thrombolysis or endovascular therapy.

18. The computer-readable storage medium according to claim 13, characterized in that, The method also includes the calculation of a second thrombus permeability index, and the evaluation results of intravenous thrombolysis or endovascular treatment are obtained by evaluating reperfusion therapy through the thrombus permeability and the second thrombus permeability index; the second thrombus permeability index is obtained by calculating the attenuation difference of thrombi at the same anatomical location on non-contrast CT and CT angiography of the head. The second thrombus permeability index is calculated by the difference between the attenuation value of non-contrast CT of the head and the attenuation value of CT angiography.

Images