Automatic measurement method for net water intake rate of acute ischemic stroke based on CTP infarct core

By designing an automated algorithm based on the RAPID-CTP core infarction area, the standardization and automation of NWU measurement in patients with acute ischemic stroke is solved, and the accurate and rapid measurement of NWU is achieved, the clinical diagnosis and treatment process is simplified, and the treatment decisions are improved. efficiency.

CN119048474APending Publication Date: 2024-11-29JIANGSU PROVINCE HOSPITAL (THE FIRST AFFILIATED HOSPITAL OF NANJING MEDICAL UNIVERSITY)
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411170591.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

In the prior art, the measurement methods of water purification intake rate (NWU) in patients with ischemic lesions in acute ischemic stroke lack uniform standardization and automation, resulting in inconsistent measurement results and long time, affecting the timeliness of treatment decisions.

Method used

Through the Python programming language, an automated algorithm based on RAPID-CTP core infarction area and NCCT images is designed to realize the registration of NCCT and CTP images, extract the infarction core area, and calculate the NWU value using the 20-80HU threshold range to avoid manual intervention and subjective interference.

Benefits of technology

The automated quantitative measurement of NWU is realized, which improves the accuracy and consistency of measurements, simplifies the operation process, reduces the waiting time for treatment decisions, and provides objective imaging biomarker support.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048474A_ABST
    Figure CN119048474A_ABST
Patent Text Reader

Abstract

The invention relates to an acute ischemic stroke net water intake rate automatic measurement method based on a CTP infarction core, and the method comprises the following steps: (1) registration of an NCCT image and a CTP image, (2) extraction of an RAPID-CTP core infarction region, (3) mapping of the infarction core region, (4) searching of a brain midline on the NCCT, and (5) mirroring of the core infarction region on the NCCT. (6) threshold operation; and (7) NWU value calculation. And calculating according to an NWU calculation formula so as to quantify the NWU of the AIS ischemic infarction focus. The technical scheme has the characteristics of automation, standardization, simplicity and convenience in operation and the like, and manual intervention is not needed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an automated measurement method, and in particular to a method for realizing automated measurement of the net water uptake rate (NWU) of ischemic lesions in patients with acute ischemic stroke (AIS) in clinical diagnosis and treatment practice. Background Art

[0002] Stroke, also known as ischemic stroke, includes ischemic stroke and hemorrhagic stroke. It refers to a group of cerebrovascular diseases caused by stenosis or occlusion of intracranial blood vessels and vascular rupture. The main clinical features are sudden onset and rapid onset of neurological dysfunction.

[0003] After acute ischemic stroke (AIS), brain tissue edema is an important pathophysiological change. Studies have found that the net water uptake (NWU) of ischemic lesions is an emerging and potential quantitative imaging biomarker of brain edema, which can be quantified by relative computed tomography (CT) density values. The quantitative method is as follows: (1) The average CT density of ischemic brain tissue on the ischemic hemisphere side is recorded as HU ischemic ; (2) The average CT density of normal brain tissue on the normal hemisphere side of the mirror image is recorded as HU normal (3) Calculate the NWU per unit volume of ischemic lesion according to the quantitative formula: NWU% = [1-(HU ischemic ) / (HU normal )]*100%.

[0004] At present, in clinical research and diagnosis and treatment practice, the main methods for measuring NWU are as follows:

[0005] ① Based on the computed tomography perfusion (CT Perfusion, CTP) method: First, the maximum range of brain tissue ischemia and hypoperfusion is determined using the time to drain (TTD) perfusion parameter map; secondly, the final ischemic core area is determined by combining the cerebral blood volume (CBV) perfusion parameter map; finally, the determined ischemic core area is mapped to the non-contrast enhanced computed tomography (NCCT) image, and the average CT density value of the ischemic core area on the NCCT is measured, which is recorded as HU ischemic; The ischemic core area was mirrored to the corresponding normal brain tissue area in the contralateral hemisphere, and the average CT density value of the normal brain tissue area was measured and recorded as HU normal. According to the formula: CTP-NWU% = [1-(HUischemic ) / (HU normal )]*100%, measured CTP-NWU.

[0006] The advantages of this method are: due to the inherent characteristics of CTP imaging technology, this method has the advantage of high sensitivity in identifying ischemic infarction areas, so this method is the "gold standard" for NWU measurement methods. The disadvantages are: the differences in post-processing software from different equipment suppliers lead to differences in post-processing methods, resulting in different volumes of the assessed ischemic core areas, which in turn leads to a lack of standardization and unification in NWU measurement. In addition, since the post-processing of CTP source images involves time-consuming processes such as image segmentation and registration, the treatment of AIS patients is delayed to a certain extent.

[0007] ② Based on the Alberta Stroke Program Early CTScore (ASPECTS) method: ASPECTS is a semi-quantitative scoring tool for identifying the scope and degree of early ischemic changes in the middle cerebral artery (MCA) supply area of ​​AIS patients. This scoring tool divides the supply area of ​​each middle cerebral artery into 10 areas, including the cerebral cortex M1-M6, caudate nucleus, lenticular nucleus, internal capsule, and insular cortex. Each area is scored 1 point, and when ischemia accumulates in one area, 1 point is subtracted from the total score of 10. The final score = 10-area affected by ischemia. The higher the final score, the smaller the ischemic range; a final score of zero indicates that there are extensive ischemic lesions in the supply area of ​​the middle cerebral artery. The automated RAPID (iSchemaView) software can automatically calculate the average CT value of each ASPECTS area in each cerebral hemisphere. At the same time, the density difference between the corresponding ASPECTS areas of the two cerebral hemispheres is measured. If the density difference exceeds a certain preset threshold, the ASPECTS area on the ischemic hemisphere side is judged as an ischemic area or a normal area. The average CT density of all ischemic ASPECTS areas on the ischemic hemisphere side is recorded as HU ischemic , the average density of the mirrored contralateral hemisphere ASPECTS area is recorded as HU normal Using the formula CT-ASPECTS-NWU%=[1-(HU ischemic ) / (HU normal )]*100%, calculate CT-ASPECTS-NWU.

[0008] The advantages of this measurement method are: based on the ASPECTS scoring tool, using the automated ASPECTS software for calculation, it is easy and fast to operate, and compared with the CTP-NWU method, this method is widely used in clinical diagnosis and treatment practice; Disadvantages: This method is based on the automated ASPECTS software, which is based on artificial intelligence technology. In some patients with displacement of the brain midline or irregular scanning posture, the automated ASPECTS software will have ASPECTS area segmentation errors on NCCT images. At this time, experienced neuroimaging physicians are required to manually verify all affected ASPECTS areas. This limits the clinical application of this method in some special cases to a certain extent.

[0009] ③ Based on manual experience, the MCA region of interest (ROI) is delineated: First, an experienced neuroradiologist delineates the MCA region at the basal ganglia level of the NCCT image. The MCA region is defined as follows: the extension line of the anterior horn of the lateral ventricle is used as the anterior boundary of the MCA, the extension line of the posterior horn of the lateral ventricle is used as the posterior boundary of the MCA, the outer edge of the cerebral cortex is used as the outer boundary of the MCA, and the inner edge of the caudate nucleus head and the posterior limb of the internal capsule is used as the inner boundary of the MCA. Secondly, the average density of the MCA region is measured and calculated, recorded as HU ischemic ; The density of the homologous region on the contralateral side of MCA is recorded as HU normal Finally, according to the NWU quantification formula Blinded-NWU% = [1-(HU ischemic ) / (HU normal )]*100%, calculate NWU.

[0010] The advantages of this method are: only NCCT images are required, it does not rely on CTP imaging, it does not rely on automated software, and it only relies on experienced neuroradiologists to outline the ROI. In some primary hospitals where CTP cannot be implemented, Blinded-NWU can be used as an alternative NWU measurement method; Disadvantages: The outlined ROI does not represent the true infarct area, and compared with CTP-NWU, it underestimates the true NWU value. In addition, we can see that this method relies on manual experience, is subjective, and lacks automation and standardization.

[0011] ④ Image-patch based NWU measurement method: First, after reading the patient's NCCT images, two experienced neuroradiologists selected four slice levels as standard reference images and marked a 30*30 pixel mirror image block in the left and right MCA regions of each slice; then, the average density of each pair of image blocks was measured, among which the image block with lower density was considered as the ischemic area, and its density was recorded as HU ischemicThe image block with higher density on the opposite side is considered normal, and its density is recorded as HU normal Finally, based on the average density of the ischemic image patch and the average density of the normal image patch, according to the formula: Image-patch-NWU% = [1-(HU ischemic ) / (HU normal )]*100%, and calculate NWU.

[0012] The advantages of this method are: easy operation, and it is a practical tool for NWU measurement in some hospitals that lack CTP technology; disadvantages: the method of NWU quantification based on the image block method also relies on experienced neuroradiologists, the consistency between and within operators is questionable, and it lacks automation and standardization.

[0013] From the above introduction, we can know that NWU of ischemic lesions can be used as a quantitative imaging biomarker of brain edema in patients with acute ischemic stroke (AIS). The key to measuring NWU is to determine the extent of ischemic brain tissue.

[0014] Method ① CTP-NWU. This method relies on multimodal advanced (CTP) neuroimaging technology to complete the measurement, but it is not feasible for some primary stroke centers that cannot achieve or lack CTP technology, which limits the universality of this method to a certain extent; in addition, there are differences in the core infarct volume derived from different CTP post-processing software, which leads to differences in NWU value measurement;

[0015] Method ②ASPECTS-NWU. This method is a semi-automatic measurement method that relies on the automated ASPCETS software. The automated software divides the MCA blood supply area into 10 sub-areas and calculates the average CT density value of each sub-area. When the range of ischemic brain tissue only involves part of a sub-area, the average density of the entire sub-area is used to reflect the density of the ischemic brain tissue, which will overestimate the density of the ischemic brain tissue and thus underestimate the NWU value of the ischemic brain tissue, thereby underestimating the severity of cerebral edema. Therefore, compared with the CTP-NWU method, this method underestimates the true NWU value of the ischemic brain tissue. In addition, the automated ASPECTS software based on artificial intelligence technology performs poorly in some special cases and needs to be further verified by experienced neuroimaging physicians, resulting in poor universality of the method and lack of true automation.

[0016] Method ③ Blinded-NWU and method ④ Image-Patch-NWU both rely on experienced neuroradiologists to manually determine the ischemic brain tissue area. The NWU values ​​calculated by methods ③ and ④ are lower than the actual NWU values, and both lack automation.

[0017] In summary, there is currently a lack of a unified and standardized automatic measurement method for NWU. Based on this background, we intend to develop a set of algorithms to achieve automated quantification of NWU: based on the core infarct area (RAPID-CTP core infarct area) derived from CTP raw images post-processed by RAPID (iSchemaView) software and NCCT images, and use the Python programming language to design an automatic algorithm to achieve automated quantification of NWU. Summary of the invention

[0018] The present invention aims at the technical problems existing in the prior art and provides an automated measurement method for the net water uptake rate of acute ischemic stroke based on CTP infarct core. The technical solution has the characteristics of automation, standardization, and easy operation, and does not require human intervention.

[0019] In order to achieve the above object, the technical solution of the present invention is as follows: an automated measurement method for net water intake rate of acute ischemic stroke based on CTP infarct core, the method comprising the following steps:

[0020] Step (1) NCCT image and CTP image registration. In the real world clinical diagnosis and treatment practice, the head of some AIS patients may move during the period between the collection of NCCT image and CTP image of AIS patients at admission, resulting in the mismatch of the spatial position of NCCT image and CTP image. Therefore, image registration is required. The specific method is: take the NCCT image as the reference image, take the first frame of CTP image as the floating image, use FLIRT (FMRIB's Linear Image Registration Tool) for registration, and save the registered deformation matrix at the same time.

[0021] Step (2) Extraction of the core infarct area of ​​RAPID-CTP: Based on the cerebral blood flow (CBF) parameter map generated by post-processing the CTP source image using RAPID (iSchemaView) post-processing software, the core infarct area is extracted according to the CBF < 30% threshold.

[0022] Step (3) mapping of the infarct core area: the deformation matrix obtained in step (1) is multiplied by the infarct core mask image obtained in step (2) to map the RAPID-CTP core infarct area to the NCCT image space, and the infarct core area on NCCT is obtained, which is denoted as ischemic.

[0023] Step (4) Find the brain midline on the NCCT, use FLIRT to register the NCCT to the MNI-152 standard template space, and save the registered deformation matrix. Use the brain midline map in the MNI-152 standard template space and the inverse matrix of the deformation matrix to perform multiplication to obtain the brain midline displayed on the NCCT image.

[0024] Step (5) Mirroring of the infarct core area on NCCT: The infarct core area on NCCT is flipped and mirrored along the midline of the brain to the corresponding area of ​​the contralateral cerebral hemisphere, that is, the normal brain tissue area of ​​the ischemic area on the ischemic hemisphere is mirrored on the contralateral normal hemisphere, and this area is recorded as normal.

[0025] Step (6) Threshold operation: According to previous studies, in order to exclude the influence of non-brain tissue density, NWU quantification needs to set a threshold range of 20-80HU. First, based on the CT density value threshold range of 20-80HU, voxels representing brain tissue are selected from the ischemic and normal regions; secondly, the average CT density values ​​of the voxels selected from the ischemic and normal regions are calculated and recorded as HU ischemic and HU normal .

[0026] Step (7) Calculation of NWU value. Based on the density HU of the infarct core area ischemic Normal brain tissue density HU mirrored with the contralateral hemisphere normal , calculate the NWU value according to the following equation:

[0027] NWU%=(1-HU ischemic / HU normal )*100%.

[0028] Compared with the prior art, the present invention has the following advantages. The present invention creates a set of algorithms for realizing the automated quantification of NWU of ischemic brain tissue by using Python programming language: based on RAPID-CTP infarct core and NCCT images, the automated measurement method is used to realize the automated quantification process of NWU of ischemic lesions in AIS patients. CTP imaging technology has the advantage of high sensitivity in identifying the infarct core area, and the accuracy of determining the range of ischemic brain tissue by CTP is high; in addition, among many CTP post-processing software, the consistency between the infarct core volume obtained by post-processing CTP source data with RAPID (iSchemaView) software and the final infarct volume (FIV) is the highest. Therefore, the present invention creates the calculation of the NWU value of ischemic brain tissue based on the RAPID-CTP core infarct area, and the obtained NWU value is the true NWU value of the ischemic lesion. The automated measurement method created by the present invention can realize the automatic quantification of NWU of ischemic brain tissue, and no further human intervention is required in the quantification process, thus avoiding the interference of human subjectivity.

[0029] (2) Advantages:

[0030] Compared with method ①, the automated method designed by the present invention has the advantages of automation and standardization. It only needs to input the CBF parameter map and NCCT image derived from the RAPID post-processed CTP image of the AIS patient into the automated measurement method, and the NWU value of the AIS patient will be output, thus realizing the automated calculation of NWU. In addition, the quantification of NWU based on the core infarct area of ​​RAPID-CTP, in addition to its high accuracy and sensitivity, also has the advantages of being fast and simple, which can save time for AIS patients in real-world clinical diagnosis and treatment, and provide assistance for clinical decision-making for AIS patients;

[0031] Compared with method ②, the NWU value calculated by the method designed by the present invention can truly reflect the NWU value of ischemic lesions in AIS patients, and the intermediate process does not require manual verification, avoiding the interference of human subjective factors, and is objective. However, method ② has the disadvantages of underestimating the true NWU value of ischemic brain tissue and requiring manual verification, which has a certain degree of subjectivity.

[0032] Compared with methods ③ and ④, the method designed by the present invention has the advantages of automation, standardization, and objectivity. Methods ③ and ④ all rely on experienced neuroimaging physicians to manually outline the ischemic ROI, lack automation, and have subjective disadvantages. In summary, the automated method designed by the present invention has the characteristics of automation, standardization, and easy operation, and does not require manual intervention. In the clinical diagnosis and treatment process of AIS patients where "time is brain", it is helpful to shorten the time AIS patients wait for test results, save time for AIS patients, and provide help and guidance for the treatment decision-making of AIS patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the overall process of the present invention. DETAILED DESCRIPTION

[0034] In order to deepen the understanding of the present invention, the present embodiment is described in detail below with reference to the accompanying drawings.

[0035] Example 1: See Figure 1 , an automated method for measuring net water intake rate in acute ischemic stroke based on CTP infarct core, the method comprising the following steps:

[0036] Step (1) NCCT image and CTP image registration. In the real world clinical diagnosis and treatment practice, the head of some AIS patients may move during the period between the collection of NCCT image and CTP image of AIS patients at admission, resulting in the mismatch of the spatial position of NCCT image and CTP image. Therefore, an image registration step is required. The specific method is: take the NCCT image as the reference image, take the first frame of CTP image as the floating image, and use the flirt function in FLIRT (FMRIB's Linear Image Registration Tool) (where the degrees of freedom are 6 and the interpolation method is trilinear) for registration, and save the registered deformation matrix at the same time.

[0037] Step (2) Extraction of the core infarct area of ​​RAPID-CTP: Based on the cerebral blood flow (CBF) parameter map generated by post-processing the CTP source image using RAPID (iSchemaView) post-processing software, the core infarct area is extracted according to the CBF < 30% threshold.

[0038] Step (3) Mapping of the infarct core area: The deformation matrix obtained in step (1) is multiplied by the infarct core mask image obtained in step (2) to map the RAPID-CTP core infarct area to the NCCT image space, and the infarct core area on NCCT is obtained, which is denoted as ischemic.

[0039] Step (4) Find the brain midline on NCCT, use FLIRT to register NCCT to the MNI-152 standard template space, and save the registered deformation matrix. Use the brain midline map in the MNI-152 standard template space and the inverse matrix of the deformation matrix to perform multiplication to obtain the brain midline displayed on the NCCT image.

[0040] Step (5) Mirroring the core infarct area on NCCT: flip the core infarct area on NCCT along the midline of the brain and mirror it to the corresponding area of ​​the contralateral cerebral hemisphere, that is, the normal brain tissue area on the normal hemisphere side of the ischemic area on the ischemic hemisphere side, and record this area as normal.

[0041] Step (6) Threshold operation: According to previous studies, in order to exclude the influence of non-brain tissue density, NWU quantification needs to set a threshold range of 20-80HU. First, based on the CT density value threshold range of 20-80HU, voxels representing brain tissue are selected from the ischemic and normal regions; secondly, the average CT density values ​​of the voxels selected from the ischemic and normal regions are calculated and recorded as HU ischemic and HU normal .

[0042] Step (7) Calculation of NWU value. Based on the density HU of the infarct core area ischemic Normal brain tissue density HU mirrored with the contralateral hemisphere normal , calculate the NWU value according to the following equation:

[0043] NWU%=(1-HU ischemic / HU normal )*100%.

[0044] It should be noted that the above embodiments are not intended to limit the protection scope of the present invention, and equivalent changes or substitutions made on the basis of the above technical solutions all fall within the protection scope of the claims of the present invention.

Claims

1. An automated method for measuring net water intake rate in acute ischemic stroke based on CTP infarct core, characterized in that: The method comprises the following steps: Step (1): NCCT image and CTP image registration, Step (2): Extraction of the core infarct area of ​​RAPID-CTP, Step (3): Mapping of the infarct core area, Step (4): Find the brain midline on NCCT. Step (5): Mirror image of the core infarct area on NCCT, Step (6): Threshold operation, Step (7): Calculation of NWU value.

2. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 1, characterized in that: In step (1), the NCCT image is registered with the CTP image. Specifically, the NCCT image is used as the reference image, the first frame of the CTP image is taken as the floating image, and FLIRT (FMRIB's Linear Image Registration Tool) is used for registration, and the registered deformation matrix is ​​saved.

3. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 2, characterized in that: In step (2), the extraction of the RAPID-CTP core infarct area is based on the cerebral blood flow (CBF) parameter map generated by post-processing the CTP source image with RAPID (iSchemaView) post-processing software, and the core infarct area is extracted from it according to the CBF < 30% threshold.

4. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 3, characterized in that: In step (3), the mapping of the infarct core area is achieved by multiplying the deformation matrix obtained in step (1) with the infarct core mask image obtained in step (2) to map the RAPID-CTP core infarct area to the NCCT image space, and obtain the infarct core area on NCCT, which is recorded as ischemic.

5. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 4, characterized in that: Step (4) Find the brain midline on the NCCT, use FLIRT to register the NCCT to the MNI-152 standard template space, save the registered deformation matrix, and use the brain midline map in the MNI-152 standard space template and the inverse matrix of the deformation matrix to multiply to obtain the brain midline displayed on the NCCT image.

6. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 5, characterized in that: Step (5) The core infarct area on NCCT is mirrored to the contralateral hemisphere. The core infarct area on NCCT is flipped along the midline of the brain and mirrored to the corresponding area of ​​the contralateral hemisphere, that is, the normal brain tissue area where the ischemic area on the ischemic hemisphere side is mirrored on the normal hemisphere side, and this area is recorded as normal.

7. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 2, characterized in that: Step (6) Threshold operation: NWU quantification requires setting a threshold range of 20-80HU. First, voxels representing brain tissue are selected from the ischemic and normal regions according to the CT density value threshold range of 20-80HU. Second, the average CT density values ​​of the voxels selected from the ischemic and normal regions are calculated and recorded as HU ischemic and HU normal。 8. The method for automated measurement of net water intake rate in acute ischemic stroke based on CTP infarct core according to claim 2, characterized in that: Step (7) Calculation of NWU value based on the density HU of the infarct core area ischemic The normal brain tissue density of the contralateral hemisphere is HU normal , calculate the NWU value according to the following equation: NWU%=(1-HU ischemic / HU normal )*100%.

Citation Information

Cited By

  • Cerebral infarction severity scoring method based on brain slice TTC staining of mouse cerebral infarction model

    CN119941702A

  • A cerebral infarction severity scoring method based on TTC staining of brain sections in a mouse cerebral infarction model

    CN119941702B

  • Skull CT (Computed Tomography) plain-scan image processing method and platform and storage medium

    CN121837288A