Automatic evaluation method and device for intracranial collateral circulation, storage medium and computing equipment

Abstract

This invention provides an automated evaluation method, apparatus, storage medium, and computing device for intracranial collateral circulation. The method includes: obtaining cerebral blood flow images based on ASL images from multiple PLD times; segmenting the cerebral blood flow images to obtain regions of interest (ROIs) for arterial arrival artifacts; calculating quantitative parameters for evaluating intracranial collateral circulation based on the ROIs; and marking the ROIs and the quantitative parameters in the cerebral blood flow images. The method of this invention can more objectively evaluate the severity of arterial arrival artifacts, thereby evaluating collateral circulation and better assisting physicians in making clearer judgments and analyses of patients' conditions in clinical practice.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to an automatic evaluation method, device, storage medium, and computing equipment for intracranial collateral circulation. Background Technology

[0002] Collateral circulation is a network of blood vessels formed between proximal and distal branches of the main coronary artery. These networks are inherent and normally dormant. However, they become active when the main coronary artery is blocked, undertaking some of the circulatory work to supplement or even completely replace the insufficient blood flow from the main coronary artery. This ensures that the blood supply to the tissues is not interrupted. Collateral circulation refers to the interconnection between coronary arteries. When a coronary artery or a larger branch becomes severely narrowed or blocked, other coronary arteries supply blood to the diseased coronary artery through these interconnecting branches. This circulatory system is called collateral circulation. Coronary angiography shows that one coronary artery supplies blood to another severely narrowed or blocked coronary artery through these interconnecting branches.

[0003] Arterial spin labeling (ASL) is an MRI technique that uses water molecules in the blood as an endogenous, freely diffusing tracer for brain perfusion imaging. Compared to other drug-assisted perfusion methods, ASL is simpler, safer, and faster. However, 3DASL cannot accurately and clearly segment ischemic areas because it cannot acquire peak-time cerebral blood flow (CBF) images.

[0004] Clinicians can evaluate collateral circulation by identifying the arterial transit artifact (ATA) in the arterial arrival signal (ASL). ATA is commonly found in collateral compensating vessels or proximal to occluded vessels. Blood flow in these collateral circulations or occluded vessels is slower, and this slower flow remains within the vessel during imaging, hence the term "arterial arrival artifact." Clinically, this is also known as "intravascular hyperintensity," somewhat similar to the FLAIR hyperintensity sign. The presence of this sign indicates the opening of collateral circulation.

[0005] Currently, methods for evaluating collateral circulation based on ATA (Automatic Assessment of Tracheal Circulation) mainly rely on subjective human evaluation to identify high-signal and highly visible signal regions within the collateral circulation area of ​​the ASL-CBF sequence. High-signal regions are scored as 1, and highly visible signal regions as 2. Finally, all high-signal and highly visible signal regions are summed to obtain the ATA score.

[0006] Another method for evaluating collateral circulation patency is MRA sequence. The MRA criteria for evaluating the degree of cerebral vascular stenosis are as follows: ICA and MCA scores range from 0 to 3 (0 for normal, 3 for not visible); ACA and posterior cerebral artery (PCA) scores range from 0 to 2 (0 for normal, 2 for not visible). ICA and MCA stenosis is categorized as normal (0), mild stenosis (1), moderate stenosis (2), and severe stenosis or occlusion (3); ACA and PCA stenosis is categorized as normal (0), mild to moderate stenosis (1), and severe stenosis or occlusion (2). A score of 0 indicates no vessel involvement, and all other scores indicate vessel involvement. Two attending radiologists with over 5 years of experience independently score the cerebral vessels. If their opinions differ, they consult and reach a consensus. These four scores are then superimposed to obtain a 0-10 rating scale, which is used to evaluate collateral circulation.

[0007] The method of identifying high-signal and high-signal regions in the lateral recurrent region of the ASL-CBF sequence assigns a score of 1 to high-signal and a score of 2 to high-signal. Finally, all high-signal and high-signal regions are summed to obtain the final score. This scoring method has no upper limit and cannot obtain a relative quantitative parameter. In addition, the number of high-signal and high-signal regions is strongly correlated with the signal-to-noise ratio of the CBF itself. Therefore, the score obtained by this method is prone to being less objective due to image quality.

[0008] The ASL-CBF used in the ATA scoring above is mainly obtained through single PLD-PWI. However, for single PLD-PWI calculations of CBF, it's impossible to determine whether the perfusion reached the designated area or whether perfusion was completed under that PLD, so the CBF results obtained from a single PLD may be inaccurate.

[0009] Furthermore, in clinical practice, physicians may encounter cases where unilateral or bilateral posterior circulation shows hypoperfusion images in the early stages (PLD = 1.5s), but the hypoperfused areas recover to normal perfusion in the later stages (PLD = 2.5s). Based on this, physicians diagnose the patient as having early unilateral or bilateral posterior circulation hypoperfusion followed by later compensation. MRA examination suggests that such cases, or most cases, do not involve significant stenosis of the large vessels in the posterior circulation. This condition may be due to embryonic posterior cerebral artery. Such posterior cerebral arteries often exhibit developmental variations on both sides. ASL images, based on arterial transit time, can provide a relatively accurate CBF (circulatory flow factor). In normal individuals, the hemodynamics of the anterior and posterior circulations are not consistent. Embryonic posterior cerebral arteries act like a "bridge," connecting the anterior and posterior circulations, delivering blood from the anterior circulation to the posterior circulation in a shorter time, causing early hypoperfusion areas to appear as normal perfusion in the later stages. In this case, MRA may not indicate any abnormality in the posterior circulation vessels. Summary of the Invention

[0010] In view of the above problems, the present invention proposes an automatic evaluation method, apparatus, storage medium and computing device for intracranial collateral circulation that overcomes or at least partially solves the above problems.

[0011] According to a first aspect of the present invention, an automatic evaluation method for intracranial collateral circulation is provided, the method comprising:

[0012] Cerebral blood flow images were obtained from ASL images with multiple PLD times;

[0013] The cerebral blood flow image is segmented to obtain the region of interest for artery arrival artifacts;

[0014] Quantitative parameters for evaluating intracranial collateral circulation are calculated based on the region of interest of the artery reaching the artifact.

[0015] The region of interest for the artery arrival artifact and the quantization parameters are marked in the cerebral blood flow image.

[0016] Optionally, obtaining cerebral blood flow images based on ASL images from multiple PLD times includes:

[0017] Acquire ASL images for multiple PLDs at different times, and register the ASL images of each PLD;

[0018] Cerebral blood flow images were obtained from ASL images at each PLD time after registration; the registered ASL images included ASL proton density maps and ASL perfusion maps.

[0019] Optionally, obtaining cerebral blood flow images based on ASL images from each registered PLD time includes:

[0020] For each registered ASL perfusion map, the weighted average of PLD time is calculated pixel by pixel, using PLD time as the sample and the pixel value of the ASL perfusion map as the weight.

[0021] Calculate the average of the weighted average values ​​corresponding to the time of each PLD, and use it as a time reference value;

[0022] Find the target PLD time that is greater than the time reference value and has the smallest difference from the time reference value, and calculate the cerebral blood flow image based on the ASL proton density map and ASL perfusion map corresponding to the target PLD time.

[0023] Optionally, segmenting the cerebral blood flow image to obtain the region of interest for artery arrival artifacts includes:

[0024] Obtain a brain structure image corresponding to the location of the ASL image, and register the ASL proton density map as a fixed image and the brain structure image as a floating image to obtain the registered target brain structure image.

[0025] A standard brain structure image is acquired, and the standard brain structure image is registered with the target brain structure image to map the occipital lobe region, frontal lobe region, and temporal lobe region in the standard brain structure image to the cerebral blood flow image.

[0026] The region of interest for arterial arrival artifacts is determined based on the pixel values ​​of the occipital lobe, frontal lobe, and temporal lobe regions in the cerebral blood flow image; the region of interest for arterial arrival artifacts includes the high signal region of arterial arrival artifacts and the high signal region of arterial arrival artifacts.

[0027] Optionally, determining the region of interest for artery arrival artifacts based on pixel values ​​in the occipital, frontal, and temporal lobe regions of the cerebral blood flow image includes:

[0028] Calculate the median pixel value in the occipital lobe region of the cerebral blood flow image;

[0029] The pixel values ​​and the median values ​​in the frontal and temporal lobe regions of the cerebral blood flow image are compared to mark the high signal regions of arterial arrival artifacts and the high signal regions of arterial arrival artifacts based on the comparison results.

[0030] Optionally, the quantization parameters include one or more of the following: abnormal region volume, abnormal side, mismatch volume, and mismatch rate.

[0031] Optionally, the quantitative parameters calculated based on the region of interest of the artery reaching the artifact for evaluating intracranial collateral circulation include:

[0032] The volumes of the first anomalous region corresponding to the high signal region and the second anomalous region corresponding to the high signal region are statistically analyzed.

[0033] The axis of symmetry is calculated using a symmetry axis detection algorithm, and the hemisphere containing the high signal or the prominent high signal is marked as the abnormal side.

[0034] Calculate the sum of the volumes of the abnormal frontal lobe and temporal lobe regions, and calculate the mismatch volume based on the sum of the volumes of the abnormal frontal lobe and temporal lobe regions, the volume of the first abnormal region, and the volume of the second abnormal region;

[0035] The ratio of the volume of the first abnormal region, the volume of the second abnormal region, to the sum of the volumes of the abnormal frontal and temporal lobes is calculated as the mismatch rate.

[0036] According to a second aspect of the present invention, an automated evaluation device for intracranial collateral circulation is provided, the device comprising:

[0037] The cerebral blood flow image acquisition module is used to obtain cerebral blood flow images based on ASL images with multiple PLD times;

[0038] The ATA region segmentation module is used to segment the cerebral blood flow image to obtain the region of interest for artery arrival artifacts;

[0039] The quantitative parameter calculation module is used to calculate quantitative parameters for evaluating intracranial collateral circulation based on the region of interest of the artery reaching the artifact.

[0040] The results display module is used to mark the region of interest for artery arrival artifacts and the quantization parameters in the cerebral blood flow image.

[0041] According to a third aspect of the present invention, a computer-readable storage medium is provided for storing program code for performing the automatic evaluation method for intracranial collateral circulation as described in the first aspect.

[0042] According to a fourth aspect of the present invention, a computing device is provided, the computing device comprising a processor and a memory:

[0043] The memory is used to store program code and transmit the program code to the processor;

[0044] The processor is configured to execute the automatic evaluation method for intracranial collateral circulation as described in any one of the first aspects according to the instructions in the program code.

[0045] This invention provides an automatic evaluation method, device, storage medium, and computing device for intracranial collateral circulation. This invention corrects CBF images based on multi-PLD ASL, making CBF more consistent with the actual perfusion situation, and solving the shortcomings of existing single-PLD-ASL and multi-PLD-ASL CBF calculations.

[0046] Furthermore, through high-signal and high-signal region segmentation and quantization, new quantification parameters are calculated, including abnormal region volume, abnormal side, mismatch volume, and mismatch ratio. These parameters reflect the opening of collateral circulation, thereby more objectively evaluating the severity of arterial arrival artifacts and thus evaluating collateral circulation. This addresses the problems caused by subjective evaluation of arterial arrival artifact ATA (Arterial Arrival Artery Awareness) due to inaccurate threshold judgment and low consistency in repeated evaluations. By obtaining the relative quantification value of ATA, the degree of ATA can be evaluated more objectively, thereby evaluating collateral circulation.

[0047] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below.

[0048] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0049] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0050] Figure 1 A schematic flowchart of an automatic evaluation method for intracranial collateral circulation according to an embodiment of the present invention is shown;

[0051] Figure 2 A schematic diagram of an automatic evaluation device for intracranial collateral circulation according to an embodiment of the present invention is shown. Detailed Implementation

[0052] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0053] The algorithm takes T1 and ASL sequences from the same patient as input, where ASL consists of two parts: a proton density map (ASL-M0) and a perfusion map (ASL-PWI). The detailed implementation steps of the algorithm are as follows:

[0054] This invention provides an automated evaluation method for intracranial collateral circulation, such as... Figure 1 As shown, the automatic evaluation method for intracranial collateral circulation provided in this embodiment of the invention may include at least the following steps S101 to S104.

[0055] S1, cerebral blood flow images are obtained based on ASL images from multiple PLD times.

[0056] S2, the cerebral blood flow image is segmented to obtain the region of interest for arterial arrival artifact (ATA);

[0057] S3, Calculate quantitative parameters for evaluating intracranial collateral circulation based on the region of interest of the artifact reached by the artery;

[0058] S4, mark the region of interest for the artery arrival artifact and the quantization parameters in the cerebral blood flow image.

[0059] After being labeled, the blood reaches the capillaries after a certain period of time, at which point image acquisition can be performed. The time interval from labeling to acquisition is called the PLD (Post Label Delay) time. Different PLD times can reflect different perfusion results and perfusion behaviors. This embodiment obtains cerebral blood flow images (CBF images) based on ASL images with multiple PLDs, overcoming the shortcomings of existing single PLD-ASL methods for calculating CBF, making the cerebral blood flow images more consistent with the actual perfusion situation, and thus obtaining accurate evaluation parameters for intracranial collateral circulation.

[0060] In some embodiments, obtaining cerebral blood flow images based on ASL images with multiple PLD times in step S1 above may include:

[0061] S1-1, acquire ASL images of multiple PLDs at different times, and register the ASL images of each PLD; optionally, the ASL images of each PLD can be rigidly registered to prevent head displacement during patient image acquisition from causing result deviation.

[0062] S1-2, Obtain cerebral blood flow images based on the ASL images at each registered PLD time. After obtaining the ASL images at each registered PLD time, cerebral blood flow images can be obtained, which are denoted as CBF images in this embodiment. Each registered ASL image at each PLD time can include two parts: an ASL proton density map (ASL-M0) and an ASL perfusion map (ASL-PWI).

[0063] In an optional embodiment of the present invention, step S1-2 above, which obtains cerebral blood flow images based on ASL images of each registered PLD time, may further include:

[0064] S1-2-1, For each registered ASL perfusion map, using PLD time as the sample and the pixel value of the ASL perfusion map as the weight, calculate the weighted average of PLD time for each pixel; calculate the average of the weighted averages corresponding to each PLD time as the time reference value;

[0065] In other words, for each registered ASL-PWI sequence, the PLD time corresponding to each ASL-PWI sequence is used as a sample, and the pixel value of ASL-PWI is used as a weight. The weighted average value δi of the PLD time is calculated for each pixel, and then the average value of all δi is calculated to obtain δ, which is used as a time reference value.

[0066] S1-2-2, find the target PLD time that is greater than the time reference value δ and has the smallest difference from the time reference value δ, and calculate the cerebral blood flow image based on the ASL proton density map and ASL perfusion map corresponding to the target PLD time.

[0067] The target PLD time is determined by finding the PLD time that is greater than δ and closest to the time reference value δ. The CBF image is then calculated based on the ASL-PWI and ASL-M0 of the target PLD time using the following formula.

[0068]

[0069] Where, ω i For PLD time, R 1a Here, τ represents the longitudinal relaxation rate of blood, λ represents the duration of the labeled pulse, λ represents the blood-brain partitioning coefficient, α represents the PCASL labeling efficiency, ΔM represents the perfusion-weighted image ASL-PWI, and M0 represents the proton density image ASL-M0. The signal of each pixel in the ASL sequence changes from noise to signal and then back to noise as the PLD time progresses. In this embodiment, a weighted averaging method is used to find the PLD time with the highest signal contribution, which is then used as a reference to find the ASL sequence with the highest signal-to-noise ratio for calculating the cerebral blood flow map.

[0070] After obtaining the cerebral blood flow image, it can be segmented to obtain the region of interest for arterial arrival artifacts. In some embodiments, the segmentation of the cerebral blood flow image to obtain the region of interest for arterial arrival artifacts in step S2 above includes:

[0071] S2-1, Obtain the brain structure image corresponding to the position of the ASL image, and register the ASL proton density map as a fixed image and the brain structure image as a floating image to obtain the registered target brain structure image; The brain structure image in this embodiment can be understood as a T1 image, which is easy to obtain and has a clear brain structure.

[0072] The T1 image and ASL-M0 are rigidly registered in three dimensions, where the T1 sequence is a moving image and the ASL-M0 is a fixed image, resulting in the registered T1 sequence.

[0073] S2-2, Obtain a standard brain structure image, and register the standard brain structure image with the target brain structure image to map the occipital lobe region, frontal lobe region, and temporal lobe region in the standard brain structure image to the cerebral blood flow image;

[0074] In this embodiment, the standard brain structure image can serve as a template for the brain structure image. The standard brain structure image is non-rigidly registered with the registered T1 image, mapping the occipital, frontal, and temporal lobe regions of the standard brain structure to the CBF image. In this embodiment, the T1 sequence is used to assist in locating the occipital, frontal, and temporal lobe brain regions. This sequence can be replaced with any MR image that can easily distinguish brain tissue structural information.

[0075] S2-3, Based on the pixel values ​​of the occipital lobe region, frontal lobe region, and temporal lobe region in the cerebral blood flow image, determine the region of interest for arterial arrival artifacts; the region of interest for arterial arrival artifacts includes the high signal region of arterial arrival artifacts and the high signal region of arterial arrival artifacts.

[0076] In some embodiments, steps S2-3 above may include:

[0077] S2-3-1, Calculate the median pixel value med_occipital in the occipital region of the cerebral blood flow image;

[0078] S2-3-2, compare the pixel values ​​and the median values ​​of the pixels in the frontal and temporal lobe regions of the cerebral blood flow image to mark the high signal regions of arterial arrival artifacts and the high signal regions of arterial arrival artifacts based on the comparison results.

[0079] In this embodiment, a high-signal region is defined as a region with a CBF value higher than twice that of the occipital lobe, and a highly visible signal region is defined as a region with a CBF value higher than 1.5 times but less than twice that of the occipital lobe. Therefore, after calculating the median pixel value med_occipital in the occipital lobe region, two thresholds can be determined: 2*med_occipital and 1.5*med_occipital. Regions in the frontal and temporal lobes with pixel values ​​greater than 2*med_occipital are marked as high-signal regions of arterial arrival artifact ATAhigh1, and regions in the frontal and temporal lobes with pixel values ​​greater than 1.5*med_occipital but less than 2*med_occipital are marked as highly visible signal regions of arterial arrival artifact ATAhigh2. Morphological processing is then performed on the aforementioned high-signal regions ATAhigh1 and highly visible signal regions ATAhigh2.

[0080] In practical applications, for the segmentation of high signal regions and regions with significant high signal, deep learning-based segmentation network models can be used for image segmentation. For example, high signal regions and regions with significant high signal can be labeled in advance, and these data can be used to train a deep learning network to achieve intelligent segmentation of high signal regions and regions with significant high signal.

[0081] Furthermore, after obtaining the high-signal region of artery arrival artifact and the high-signal region of artery arrival artifact, quantitative parameters for evaluating intracranial collateral circulation can be calculated. In this embodiment, the quantitative parameters for evaluating intracranial collateral circulation may include one or more of the following: abnormal region volume, abnormal side, mismatch volume, and mismatch rate. The specific calculation methods for each quantitative parameter are as follows:

[0082] Abnormal region volume: Calculate the first abnormal region volume Vol_ATAhigh1 corresponding to the high signal region and the second abnormal region volume Vol_ATAhigh2 corresponding to the high signal region;

[0083] Abnormal side: The symmetry axis is calculated using a symmetry axis detection algorithm, and the hemisphere where the high signal or the high signal is located is marked as the abnormal side;

[0084] Mismatch volume: Calculate the sum of the volumes of the abnormal frontal and temporal lobe regions, Vol_roi. The mismatch volume is calculated based on the sum of the volumes of the abnormal frontal and temporal lobe regions, the volume of the first abnormal region, and the volume of the second abnormal region. MismatchVol = Vol_roi – (Vol_ATAhigh1 + Vol_ATAhigh2)

[0085] Mismatch Ratio: The ratio of the volume of the first abnormal region, the volume of the second abnormal region, and the sum of the volumes of the abnormal frontal and temporal lobes is used as the mismatch ratio; MismatchRatio=(Vol_ATAhigh1+Vol_ATAhigh2) / Vol_roi.

[0086] Finally, the high signal regions can be marked with different colors in the CBF result graph, and the calculated quantization parameters can also be displayed in the result graph.

[0087] The automatic evaluation method for intracranial collateral circulation provided in this invention uses ASL based on multiple PLDs to correct CBF images, making CBF more consistent with the actual perfusion situation. Through high-signal and high-signal region segmentation and quantization, new quantitative parameters are calculated, including abnormal region volume, abnormal side, mismatch volume, and mismatch ratio. These parameters reflect the opening of collateral circulation, thus providing a more objective evaluation of the severity of arterial arrival artifacts and further evaluating collateral circulation. Compared to other collateral circulation scoring algorithms, such as MRA scoring, the ATA assessment mechanism can quantify some diseases that MRA cannot detect, better assisting doctors in making clearer judgments and analyses of patients' conditions in clinical work. The ATA method for assessing collateral circulation has a wider range of applications and stronger robustness.

[0088] Based on a unified inventive concept, embodiments of the present invention also provide an automatic evaluation device for intracranial collateral circulation, such as... Figure 2 As shown, the automated assessment device for intracranial collateral circulation includes:

[0089] The cerebral blood flow image acquisition module 210 is used to acquire cerebral blood flow images based on ASL images with multiple PLD times;

[0090] ATA region segmentation module 220 is used to segment the cerebral blood flow image to obtain the region of interest for artery arrival artifacts;

[0091] The quantitative parameter calculation module 230 is used to calculate quantitative parameters for evaluating intracranial collateral circulation based on the region of interest of the artery reaching the artifact.

[0092] Results display module 240 is used to mark the region of interest of the artery arrival artifact and the quantization parameters in the cerebral blood flow image.

[0093] In an optional embodiment of the present invention, the cerebral blood flow image acquisition module 210 can also be used for:

[0094] Acquire ASL images for multiple PLDs at different times, and register the ASL images of each PLD;

[0095] Cerebral blood flow images were obtained from ASL images at each PLD time after registration; the registered ASL images included ASL proton density maps and ASL perfusion maps.

[0096] In an optional embodiment of the present invention, the cerebral blood flow image acquisition module 210 can also be used for:

[0097] For each registered ASL perfusion map, the weighted average of PLD time is calculated pixel by pixel, using PLD time as the sample and the pixel value of the ASL perfusion map as the weight.

[0098] Calculate the average of the weighted average values ​​corresponding to the time of each PLD, and use it as a time reference value;

[0099] Find the target PLD time that is greater than the time reference value and has the smallest difference from the time reference value, and calculate the cerebral blood flow image based on the ASL proton density map and ASL perfusion map corresponding to the target PLD time.

[0100] In an optional embodiment of the present invention, the ATA region segmentation module 220 can also be used for:

[0101] Obtain a brain structure image corresponding to the location of the ASL image, and register the ASL proton density map as a fixed image and the brain structure image as a floating image to obtain the registered target brain structure image.

[0102] A standard brain structure image is acquired, and the standard brain structure image is registered with the target brain structure image to map the occipital lobe region, frontal lobe region, and temporal lobe region in the standard brain structure image to the cerebral blood flow image.

[0103] The region of interest for arterial arrival artifacts is determined based on the pixel values ​​of the occipital lobe, frontal lobe, and temporal lobe regions in the cerebral blood flow image; the region of interest for arterial arrival artifacts includes the high signal region of arterial arrival artifacts and the high signal region of arterial arrival artifacts.

[0104] In an optional embodiment of the present invention, the ATA region segmentation module 220 can also be used for:

[0105] Calculate the median pixel value in the occipital lobe region of the cerebral blood flow image;

[0106] The pixel values ​​and the median values ​​in the frontal and temporal lobe regions of the cerebral blood flow image are compared to mark the high signal regions of arterial arrival artifacts and the high signal regions of arterial arrival artifacts based on the comparison results.

[0107] In an optional embodiment of the present invention, the quantization parameter calculation module 230 can also be used for:

[0108] The volumes of the first anomalous region corresponding to the high signal region and the second anomalous region corresponding to the high signal region are statistically analyzed.

[0109] The axis of symmetry is calculated using a symmetry axis detection algorithm, and the hemisphere containing the high signal or the prominent high signal is marked as the abnormal side.

[0110] Calculate the sum of the volumes of the abnormal frontal lobe and temporal lobe regions, and calculate the mismatch volume based on the sum of the volumes of the abnormal frontal lobe and temporal lobe regions, the volume of the first abnormal region, and the volume of the second abnormal region;

[0111] The ratio of the volume of the first abnormal region, the volume of the second abnormal region, to the sum of the volumes of the abnormal frontal and temporal lobes is calculated as the mismatch rate.

[0112] This invention also provides a computer-readable storage medium for storing program code for executing the automatic evaluation method for intracranial collateral circulation described in the above embodiments.

[0113] This invention also provides a computing device, which includes a processor and a memory: the memory is used to store program code and transmit the program code to the processor; the processor is used to execute the automatic evaluation method for intracranial collateral circulation described in the above embodiments according to the instructions in the program code.

[0114] Those skilled in the art will clearly understand that the specific working process of the systems, devices, modules and units described above can be referred to the corresponding process in the foregoing method embodiments. For the sake of brevity, it will not be repeated here.

[0115] Furthermore, the functional units in the various embodiments of the present invention can be physically independent of each other, or two or more functional units can be integrated together, or all functional units can be integrated into one processing unit. The integrated functional units described above can be implemented in hardware, or in software or firmware.

[0116] Those skilled in the art will understand that if the integrated functional unit is implemented in software and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or all or part of it, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computing device (e.g., a personal computer, server, or network device) to execute all or part of the steps of the methods described in the embodiments of the present invention when running the instructions. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0117] Alternatively, all or part of the steps of the foregoing method embodiments can be implemented by hardware (such as a computing device, personal computer, server, or network device) related to program instructions. The program instructions can be stored in a computer-readable storage medium. When the program instructions are executed by the processor of the computing device, the computing device executes all or part of the steps of the methods described in the various embodiments of the present invention.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that within the spirit and principles of the present invention, modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the corresponding technical solutions to depart from the protection scope of the present invention.

Claims

1. An automated evaluation method for intracranial collateral circulation, characterized in that, The method includes: Cerebral blood flow images were obtained from ASL images with multiple PLD times; The cerebral blood flow image is segmented to obtain the region of interest for arterial arrival artifacts; the region of interest for arterial arrival artifacts includes a high-signal region for arterial arrival artifacts and a high-signal region for arterial arrival artifacts. The high-signal region is the region where the CBF value is more than twice that of the occipital lobe, and the high-signal region is the region where the CBF value is more than 1.5 times that of the occipital lobe but less than twice that of the occipital lobe. Quantitative parameters for evaluating intracranial collateral circulation are calculated based on the region of interest of the artery reaching the artifact; the quantitative parameters include one or more of the following: abnormal region volume, abnormal side, mismatch volume, and mismatch rate. In the cerebral blood flow image, the region of interest of the artery arrival artifact and the quantization parameters are marked. High signal regions and regions with significant high signal are marked in the cerebral blood flow image with different colors. The calculated quantization parameters are displayed in the cerebral blood flow image.

2. The method according to claim 1, characterized in that, The acquisition of cerebral blood flow images based on ASL images with multiple PLD times includes: Acquire ASL images for multiple PLDs at different times, and register the ASL images of each PLD; Cerebral blood flow images were obtained from ASL images at each PLD time after registration; the registered ASL images included ASL proton density maps and ASL perfusion maps.

3. The method according to claim 2, characterized in that, The cerebral blood flow images obtained based on the ASL images of each registered PLD time include: For each registered ASL perfusion map, the weighted average of PLD time is calculated pixel by pixel, using PLD time as the sample and the pixel value of the ASL perfusion map as the weight. Calculate the average of the weighted average values ​​corresponding to the time of each PLD, and use it as a time reference value; Find the target PLD time that is greater than the time reference value and has the smallest difference from the time reference value, and calculate the cerebral blood flow image based on the ASL proton density map and ASL perfusion map corresponding to the target PLD time.

4. The method according to claim 2, characterized in that, The segmentation of the cerebral blood flow image to obtain the region of interest for artery arrival artifacts includes: Obtain a brain structure image corresponding to the location of the ASL image, and register the ASL proton density map as a fixed image and the brain structure image as a floating image to obtain the registered target brain structure image. A standard brain structure image is acquired, and the standard brain structure image is registered with the target brain structure image to map the occipital lobe region, frontal lobe region, and temporal lobe region in the standard brain structure image to the cerebral blood flow image. The region of interest for artery arrival artifacts is determined based on the pixel values ​​of the occipital lobe, frontal lobe, and temporal lobe regions in the cerebral blood flow image.

5. The method according to claim 4, characterized in that, The determination of the region of interest for artery arrival artifacts based on pixel values ​​in the occipital, frontal, and temporal lobe regions of the cerebral blood flow image includes: Calculate the median pixel value in the occipital lobe region of the cerebral blood flow image; The pixel values ​​and the median values ​​in the frontal and temporal lobe regions of the cerebral blood flow image are compared to mark the high signal regions of arterial arrival artifacts and the high signal regions of arterial arrival artifacts based on the comparison results.

6. The method according to claim 1, characterized in that, The quantitative parameters used for evaluating intracranial collateral circulation, calculated based on the region of interest of the artery reaching the artifact, include: The volumes of the first anomalous region corresponding to the high signal region and the second anomalous region corresponding to the high signal region are statistically analyzed. The axis of symmetry is calculated using a symmetry axis detection algorithm, and the hemisphere containing the high signal or the prominent high signal is marked as the abnormal side. Calculate the sum of the volumes of the abnormal frontal lobe and temporal lobe regions, and calculate the mismatch volume based on the sum of the volumes of the abnormal frontal lobe and temporal lobe regions, the volume of the first abnormal region, and the volume of the second abnormal region; The ratio of the volume of the first abnormal region, the volume of the second abnormal region, to the sum of the volumes of the abnormal frontal and temporal lobes is used as the mismatch rate. Mismatch volume MismatchVol = Vol_roi – (Vol_ATAhigh1+ Vol_ATAhigh2); Mismatch ratio MismatchRatio = (Vol_ATAhigh1+ Vol_ATAhigh2) / Vol_roi; Where Vol_roi represents the sum of the volumes of the frontal and temporal lobes on the abnormal side, Vol_ATAhigh1 represents the volume of the first abnormal region corresponding to the high signal region, and Vol_ATAhigh2 represents the volume of the second abnormal region corresponding to the high signal region.

7. An automatic evaluation device for intracranial collateral circulation, characterized in that, The device includes: The cerebral blood flow image acquisition module is used to obtain cerebral blood flow images based on ASL images with multiple PLD times; The ATA region segmentation module is used to segment the cerebral blood flow image to obtain the region of interest for arterial arrival artifacts. The region of interest for arterial arrival artifacts includes a high-signal region for arterial arrival artifacts and a high-signal region for arterial arrival artifacts. The high-signal region is the region where the CBF value is more than twice that of the occipital lobe, and the high-signal region is the region where the CBF value is more than 1.5 times that of the occipital lobe but less than twice that of the occipital lobe. The quantitative parameter calculation module is used to calculate quantitative parameters for evaluating intracranial collateral circulation based on the region of interest of the artery reaching the artifact; the quantitative parameters include one or more of the following: abnormal region volume, abnormal side, mismatch volume, and mismatch rate. The results display module is used to mark the region of interest of the artery arrival artifact and the quantization parameters in the cerebral blood flow image, mark the high signal region and the high signal region with different colors in the cerebral blood flow image, and display the calculated quantization parameters in the cerebral blood flow image.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code for executing the automatic evaluation method for intracranial collateral circulation according to any one of claims 1-6.

9. A computing device, characterized in that, The computing device includes a processor and memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the automatic evaluation method for intracranial collateral circulation according to any one of claims 1-6 according to the instructions in the program code.

Images