Perfusion analysis method and system

Abstract

The present application provides a kind of perfusion analysis method, first in brain perfusion image quickly determine several venous feature points and several arterial feature points, then utilize the time density curve corresponding to the venous feature point and the time density curve corresponding to the arterial feature point are screened, obtain one of the venous feature points for representing the whole brain perfusion output condition, and one of the arterial feature points for representing the whole brain perfusion input condition, then utilize the screened venous feature point and the arterial feature point carry out perfusion analysis.The method provided by the present application is high in implementation efficiency and the selected point result can be reproduced, and since the venous feature point for representing the whole brain perfusion output condition and the arterial feature point for representing the whole brain perfusion input condition are used for perfusion analysis, the reliability of perfusion analysis is very high.Correspondingly, the present application also provides a kind of perfusion analysis system.

Description

Technical Field

[0001] This invention relates to the field of medical technology, specifically to a perfusion analysis method and system. Background Technology

[0002] Perfusion imaging is a technique that uses imaging instruments to obtain clear images reflecting organ function and metabolism, thereby diagnosing diseases. Different imaging devices, due to their different imaging principles, employ different methods for implementing perfusion imaging. A common thread is the need to capture the imaging changes of blood vessels in the imaging area in a timely manner before and after contrast agent injection. Taking CT perfusion imaging as an example, a contrast agent is rapidly injected intravenously, and continuous CT scans are performed on the region of interest. By plotting a time-dense curve (TDC), the change in contrast agent concentration in the tissue over time can be obtained. Mathematical modeling and analysis of the curve can yield a series of parameters related to tissue blood perfusion, which is of great significance for disease diagnosis and quantification.

[0003] A time density curve (TDC) is a graph used to characterize the change of CT values ​​in a region of interest (ROI) over time in a CT perfusion scan. Taking a brain perfusion CT scan as an example, the TDC plots the changes in CT values ​​for an arterial ROI and a venous ROI over time. The methods for obtaining the TDC curve typically involve: acquiring a specific arterial region in the tissue and obtaining the arterial input function (AIF) by monitoring the CT values ​​of that region over time; acquiring a specific venous region and obtaining the venous output function (VOF) by monitoring the CT values ​​of that region over time. These arterial / venous regions are usually obtained by expanding arterial / venous feature points into ROIs of a certain volume using a specific algorithm. Different arterial / venous feature points correspond to different ROIs; therefore, the determination of arterial / venous feature points is particularly important in perfusion analysis. However, traditional methods for determining arterial / venous feature points suffer from low efficiency and large errors. Summary of the Invention

[0004] The purpose of this invention is to provide a perfusion analysis method and system to solve the problems of low efficiency and large error in existing methods for determining arterial and venous feature points in perfusion analysis.

[0005] To achieve the above objectives, the present invention provides a perfusion analysis method, comprising:

[0006] Identify several venous and arterial feature points in brain perfusion images;

[0007] By filtering using the time density curves corresponding to the venous feature points and the time density curves corresponding to the arterial feature points, one of the venous feature points representing the whole brain perfusion output and one of the arterial feature points representing the whole brain perfusion input are obtained; and,

[0008] Perfusion analysis was performed using the selected venous and arterial feature points.

[0009] Optionally, the perfusion analysis method further includes:

[0010] Based on the received selection instructions, venous replacement feature points and / or arterial replacement feature points are determined in the brain perfusion image; and,

[0011] The selected vein feature points are replaced with the vein replacement feature points, and / or the selected artery feature points are replaced with the artery replacement feature points, and perfusion analysis is performed.

[0012] Optionally, the location markers of the selected vein feature points and the selected artery feature points are displayed on the temporal maximum density projection map of the brain perfusion image, and the vein replacement feature points and / or the artery replacement feature points are determined on the temporal maximum density projection map according to the received selection instruction.

[0013] Optionally, after determining a plurality of the venous feature points and a plurality of the arterial feature points in the brain perfusion image, the perfusion analysis method further includes:

[0014] Using the anatomical location of the vein feature point and / or its corresponding time density curve, determine whether the vein feature point meets a first predetermined condition; using the anatomical location of the artery feature point and / or its corresponding time density curve, determine whether the artery feature point meets a second predetermined condition; and,

[0015] Remove the vein feature points that do not meet the first predetermined condition and the artery feature points that do not meet the second predetermined condition.

[0016] Optionally, the step of screening using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points includes:

[0017] All vein feature points are scored using the time density curves corresponding to the vein feature points and a third predetermined condition; all artery feature points are scored using the time density curves corresponding to the artery feature points and a fourth predetermined condition; and...

[0018] The vein feature point with the highest score is used to represent the whole brain perfusion output, and the artery feature point with the highest score is used to represent the whole brain perfusion input.

[0019] Optionally, the step of screening using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points includes:

[0020] Combine any one of the vein feature points and any one of the artery feature points;

[0021] The combination is scored using the relative relationship of the time density curves corresponding to the vein feature points and the artery feature points in each combination, and a fifth predetermined condition; and,

[0022] The vein feature points and artery feature points in the highest-scoring combination are used to represent the whole brain perfusion output and input, respectively.

[0023] Optionally, before determining the venous feature points and the arterial feature points in the brain perfusion image, the non-infarct hemisphere region of the brain is determined using data on the change of contrast agent density over time in the left and right hemisphere regions of the brain in the brain perfusion image, and the arterial feature points are located in the non-infarct hemisphere region.

[0024] Optionally, the step of determining the non-infarcted hemisphere region of the brain includes:

[0025] Within the left hemisphere, several first regions to be determined are identified, and within the right hemisphere, second regions to be determined are identified that correspond to each of the first regions to be determined in a mirror image.

[0026] The first peak time required for the contrast agent density of each first region to be determined to reach its peak value is obtained, and the first region to be determined corresponding to the first peak time exceeding a threshold is determined to be an infarct region.

[0027] The second peak time required for the contrast agent density of each second region to be determined to reach its peak value is obtained, and the second region to be determined corresponding to the second peak time exceeding the threshold is determined to be an infarct region.

[0028] The number of infarcted regions in the left and right hemispheres is summed and compared, and the hemisphere with the smaller number of infarcted regions is determined to be the non-infarcted hemisphere.

[0029] The present invention also provides a perfusion analysis system, comprising:

[0030] The feature point determination module is used to determine several venous feature points and several arterial feature points in brain perfusion images;

[0031] A filtering module is used to filter the vein feature points using the time density curves corresponding to the vein feature points and the artery feature points, to obtain one of the vein feature points representing the whole brain perfusion output, and one of the artery feature points representing the whole brain perfusion input; and,

[0032] The perfusion analysis module is used to perform perfusion analysis using the selected venous feature points and arterial feature points.

[0033] Optional, also includes:

[0034] The user interaction module is used to display the vein feature points and the artery feature points, and in response to a selection command, to determine vein replacement feature points and / or artery replacement feature points in the brain perfusion image, and to replace the vein feature points selected by the filtering module with the vein replacement feature points, and / or to replace the selected artery feature points with the artery replacement feature points.

[0035] In the perfusion analysis method provided by this invention, several venous feature points and several arterial feature points are first rapidly identified in the brain perfusion image. Then, the time density curves corresponding to the venous feature points and the arterial feature points are used for screening to obtain one of the venous feature points representing the whole brain perfusion output and one of the arterial feature points representing the whole brain perfusion input. Perfusion analysis is then performed using the screened venous and arterial feature points. The method provided by this invention has high implementation efficiency and the point selection results are reproducible. Furthermore, because the perfusion analysis utilizes the venous feature points representing the whole brain perfusion output and the arterial feature points representing the whole brain perfusion input, the reliability of the perfusion analysis is very high. Correspondingly, this invention also provides a perfusion analysis system. Attached Figure Description

[0036] Figure 1 A flowchart of the perfusion analysis method provided in an embodiment of the present invention;

[0037] Figure 2 A detailed flowchart of the perfusion analysis method provided in this embodiment of the invention;

[0038] Figure 3 A flowchart for determining venous and arterial feature points in brain perfusion images provided in an embodiment of the present invention;

[0039] Figure 4 A detailed flowchart for determining venous and arterial feature points in brain perfusion images provided in this embodiment of the invention;

[0040] Figure 5A flowchart for removing obviously unreasonable venous or arterial feature points provided in an embodiment of the present invention;

[0041] Figure 6 A flowchart of a screening step provided in an embodiment of the present invention;

[0042] Figure 7 A flowchart of another screening step provided in an embodiment of the present invention;

[0043] Figure 8 This is a flowchart illustrating the method for determining the non-infarcted hemisphere region provided in an embodiment of the present invention.

[0044] Figure 9 A flowchart for re-perfusion analysis provided in an embodiment of the present invention;

[0045] Figure 10 This is a structural block diagram of the perfusion analysis system provided in an embodiment of the present invention;

[0046] Figure 11 A block diagram of the electronic device provided in an embodiment of the present invention;

[0047] The attached figures are labeled as follows:

[0048] 10-Feature point determination module; 20-Filtering module; 30-Infusion analysis module; 101-Processor; 102-Communication interface; 103-Memory; 104-Communication bus; 105-Display. Detailed Implementation

[0049] To make the objectives, advantages, and features of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to facilitate and clarify the explanation of the embodiments of this invention. Furthermore, the structures shown in the drawings are often part of the actual structures. In particular, different figures may emphasize different aspects and may sometimes use different scales.

[0050] Figure 1 This is a flowchart of the perfusion analysis method provided in this embodiment. Figure 1 As shown, the perfusion analysis method includes steps S1, S2 and S3.

[0051] Step S1: Identify several venous feature points and several arterial feature points in the brain perfusion image.

[0052] Figure 2 This is a flowchart illustrating the perfusion analysis method provided in this embodiment. Figure 2As shown, a large number of brain perfusion images labeled with arterial and venous feature points are first provided as training samples. The deep learning network is then trained using these training samples to obtain a feature point generation model.

[0053] Next, a brain perfusion image is provided. This image can be a CT brain perfusion image or a feature point image obtained by inference from brain perfusion data. In this embodiment, the brain perfusion image is input into a pre-trained feature point generation model. This model can automatically predict several vein feature points and several artery feature points (specifically, predict the three-dimensional coordinates of the vein and artery feature points). The method of obtaining several vein and artery feature points using the feature point generation model requires no human intervention, is highly efficient, and the point selection results are reproducible.

[0054] As an optional embodiment, before inputting the brain perfusion image into the feature point generation model, the brain perfusion image can be preprocessed in a series of steps, including but not limited to registration, skull removal, and midline extraction, to improve the accuracy of the venous and arterial feature points. The preprocessed data is then cached for later use.

[0055] In some embodiments, arterial and venous vascular information (such as vascular segmentation results or brain image partitioning) can be used to initially locate the distribution range of the venous feature points and the arterial feature points, and then the venous feature points and the arterial feature points can be obtained through the feature point generation model.

[0056] Figure 3 This is a flowchart illustrating the determination of venous and arterial feature points in the cerebral perfusion image provided in this embodiment. Figure 4 A detailed flowchart illustrating the determination of venous and arterial feature points in the cerebral perfusion image provided in this embodiment is shown below. Figure 3 and Figure 4 As shown, in some embodiments, the steps of obtaining the vein feature points and the artery feature points using the feature point generation model include steps S11, S12 and S13.

[0057] Step S11: Identify multiple regions of interest in the brain perfusion image.

[0058] For example, multiple regions of interest (ROIs) for veins and arteries in the brain perfusion image can be determined using deep learning models, thresholding, and feature extraction. For instance, batch vascular annotation can be performed on the brain perfusion image to manually delineate cerebral blood vessels, and the generated images can be used as the gold standard input to a deep learning model for training. The goal is to output vascular segmentation results from the input brain perfusion image. Then, based on information such as the brain region where the blood vessels are located, post-processing can be used to obtain the ROIs for veins and arteries. Alternatively, candidate points for arteries and veins can be directly delineated, and the generated images can be used as the gold standard input to a deep learning model for training. The goal is to output the ROIs for veins and arteries from the input brain perfusion image. Furthermore, due to the enhancement effect of contrast agents, the CT values ​​of the vascular portion in the brain perfusion image are significantly higher than those of the tissue portion, making a clear boundary between blood vessels and tissue readily distinguishable. Traditional thresholding segmentation methods can be used, such as calculating the global average CT value of the brain perfusion image after skull removal as a threshold to divide the image into background and vascular portions.

[0059] Step S12: Determine candidate vein feature points or candidate artery feature points in each region of interest.

[0060] Specifically, candidate vein feature points are determined in the region of interest for veins, and candidate artery feature points are determined in the region of interest for arteries. It should be understood that the candidate vein and artery feature points are preliminary feature points, which may include artery or vein feature points, and may also include noise points or other interfering elements.

[0061] Step S13: Select several vein feature points from the candidate vein feature points, and select several artery feature points from the candidate artery feature points.

[0062] For example, candidate vein feature points and candidate artery feature points can be preliminarily screened using candidate regions and surrounding information to exclude feature points that do not meet the feature point selection requirements (such as points outside blood vessels), thereby obtaining a number of artery feature points and a number of vein feature points.

[0063] After all the vein and artery regions of interest have been screened, if the required number of vein and artery feature points is not met, the acquisition rules for the regions of interest can be readjusted to acquire some new regions of interest.

[0064] It should be understood that when determining the region of interest in the brain perfusion image, it may not be possible to determine whether the region of interest is an arterial or venous region of interest. In this case, even after filtering feature points from the candidate point list, these feature points may not be definitively identified as arterial or venous feature points. Therefore, after filtering feature points, the distribution location of the feature points can be used to determine whether they are arterial or venous feature points. For example, due to the enhancement effect of contrast agents, the CT value of the vascular portion is significantly higher than that of the tissue portion in the brain perfusion image, making it possible to clearly distinguish the boundary between blood vessels and tissue. Furthermore, blood vessels can be distinguished from other highlighted areas (e.g., artifacts, skull) based on their location corresponding to the anatomical structure of the blood vessels (e.g., in axial views, the corresponding perpendicular line can be obtained from the brain midline at each level; vessels above the perpendicular line are arteries, and vessels below the perpendicular line are veins), morphology, and other characteristics.

[0065] Next, after determining several vein feature points and several artery feature points, some obviously unreasonable vein feature points or artery feature points can be eliminated. Figure 5 This is a flowchart illustrating the process of removing obviously unreasonable venous or arterial feature points provided in this embodiment. For example... Figure 5 As shown, removing obviously unreasonable vein or artery feature points includes steps S14 and S15.

[0066] Step S14: Using the anatomical location of the vein feature point and / or the corresponding time density curve, determine whether the vein feature point meets the first predetermined condition; using the anatomical location of the artery feature point and / or the corresponding time density curve, determine whether the artery feature point meets the second predetermined condition.

[0067] In some embodiments, the anatomical location of the venous feature point can be used to determine whether the venous feature point meets the first predetermined condition, and the anatomical location of the arterial feature point can be used to determine whether the arterial feature point meets the second predetermined condition. For example, the first or second predetermined condition can be designed such that the venous feature point is located in the superior sagittal sinus region, and the arterial feature point is located in the middle cerebral artery region.

[0068] In some embodiments, the time density curve corresponding to the venous feature point can be used to determine whether the venous feature point meets the first predetermined condition, and the time density curve corresponding to the arterial feature point can be used to determine whether the arterial feature point meets the second predetermined condition. For example, the first or second predetermined condition can be designed such that the time density curve corresponding to the venous feature point includes a plateau phase, curve rise, curve peak, curve fall, and a slow contrast agent outflow phase; the blood flow direction is from artery to vein, and the venous peak is later than the arterial peak; at the same flow rate, the venous velocity is lower than the arterial velocity, and the CT value of the venous peak is higher than that of the arterial peak (because the contrast agent will accumulate in the vein to a certain extent).

[0069] In some embodiments, a combination of the above methods can be used to determine whether the vein feature point meets the first predetermined condition and whether the artery feature point meets the second predetermined condition. Alternatively, corresponding weights can be assigned to the various methods.

[0070] Of course, in other embodiments, the first predetermined conditions and the second predetermined conditions can be designed in combination with other methods. For example, whether the middle cerebral artery is within the scanning range, or whether the preferred selection site is affected by image quality and cannot reflect the true perfusion situation, and whether the scanning cycle is complete (whether the time density curve includes the complete plateau period, curve rise, curve peak, curve fall, and contrast agent slow outflow period), etc.

[0071] Step S15: Remove the vein feature points that do not meet the first predetermined condition and the artery feature points that do not meet the second predetermined condition.

[0072] By removing obviously unreasonable vein and artery feature points, the number of vein and artery feature points can be reduced, which facilitates subsequent screening steps.

[0073] Step S2: Filter the vein feature points and the artery feature points using the time density curves corresponding to the vein feature points and the artery feature points to obtain one of the vein feature points that represents the whole brain perfusion output and one of the artery feature points that represents the whole brain perfusion input.

[0074] Figure 6 This is a flowchart illustrating a screening step provided in this embodiment. Figure 6 As shown, the screening process using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points includes steps S21 and S22.

[0075] Execution step S21: Score all the vein feature points using the time density curve corresponding to the vein feature points and the third predetermined condition, and score all the artery feature points using the time density curve corresponding to the artery feature points and the fourth predetermined condition.

[0076] It should be noted that the third predetermined condition can be designed based on the relevant parameters of the time density curve of the vein feature point. For example, it can be scored based on the peak CT value, peak time, and curve shape of the time density curve of the vein feature point. For example, a higher peak CT value results in a higher score, but this is not a limitation. Similarly, the fourth predetermined condition can be designed based on the relevant parameters of the time density curve of the artery candidate point. For example, it can be scored based on the peak CT value, peak time, and curve shape of the time density curve of the artery candidate point. For example, a higher peak CT value results in a higher score, but this is not a limitation.

[0077] Step S22: The vein feature point with the highest score is used to represent the whole brain perfusion output, and the artery feature point with the highest score is used to represent the whole brain perfusion input.

[0078] Specifically, the higher the score of the vein feature point and the artery feature point, the better the contrast agent perfusion situation in the whole brain can be restored. Therefore, the vein candidate point with the highest score is considered to represent the whole brain perfusion output situation, and the artery candidate point with the highest score is considered to represent the whole brain perfusion input situation.

[0079] In summary, the vein candidate points and artery candidate points with the highest scores are the selected vein candidate points and artery candidate points.

[0080] Figure 7 A flowchart of another screening step provided in this embodiment. For example... Figure 7 As shown, the screening using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points may also include steps S23, S24 and S25.

[0081] Step S23: Combine any one of the vein feature points and any one of the artery feature points.

[0082] In other words, the vein feature points and the artery feature points are paired up to form several combinations, and each combination contains one vein feature point and one artery feature point.

[0083] Step S24: Using the relative relationship of the time density curves corresponding to the vein feature points and the artery feature points in each combination and the fifth predetermined condition, score the combination.

[0084] For example, the fifth predetermined condition can be designed based on the relative relationship of relevant parameters of the time density curves corresponding to the venous feature points and the arterial feature points in each combination. These relevant parameters can include peak CT value, peak time, curve shape, etc. For instance, the larger the time interval between the peak times of the time density curves corresponding to the venous feature points and the arterial feature points, the higher the score, but this is not a limitation.

[0085] Step S25: The venous feature points and arterial feature points in the highest-scoring combination are used to represent the whole brain perfusion output and input, respectively.

[0086] Specifically, the higher the score of the combination, the better the vein feature points and artery feature points can reproduce the contrast agent perfusion situation in the whole brain. Therefore, it is believed that the vein feature points and artery feature points in the highest score combination are used to represent the whole brain perfusion output and input situation, respectively.

[0087] Compared to scoring the vein feature points and artery feature points one by one, the method of combining the vein feature points and artery feature points and scoring the combination may have a lower accuracy, but it can reduce the amount of computation.

[0088] It should be understood that the feature points used to represent whole-brain perfusion output and input are not necessarily the vein feature points and artery feature points with the highest scores, nor are they necessarily the vein feature points and artery feature points from the highest-scoring combinations. In some embodiments, some of the vein feature points and artery feature points with the highest scores (or some combinations of the highest scores) may be obtained first, and then the vein feature points used to represent whole-brain perfusion output and the artery feature points used to represent whole-brain perfusion input (or a combination thereof) may be selected from them.

[0089] Furthermore, due to the influence of vascular lesions, the site of ischemic stroke and the affected blood vessels may experience a certain degree of stenosis or even blockage. Selecting feature points on the infarct side is not conducive to reflecting the true cerebral perfusion status. Therefore, in this embodiment, when determining the arterial feature points in the cerebral perfusion image using the feature point generation model, the arterial feature points are located in the non-infarct hemisphere region of the cerebral perfusion image (since venous feature points are usually near the midline of the brain, they do not need to be considered). In this way, the arterial feature points can more objectively and accurately reflect the patient's true perfusion status, making the analysis results more valuable for clinical diagnosis.

[0090] In this embodiment, the data of the contrast agent density varying with time in the left cerebral hemisphere region and the right cerebral hemisphere region of the brain in the cerebral perfusion image can be used to determine the non-infarcted hemisphere region of the brain.

[0091] Figure 8 is a schematic flow chart of the method for determining a non-infarcted hemisphere region provided by this embodiment. As Figure 8 shows, determining the non-infarcted hemisphere region of the brain includes step S01, step S02, step S03 and step S04.

[0092] Perform step S01: determining a plurality of first regions to be determined in the left cerebral hemisphere region, and determining a second region to be determined that is mirror-symmetric corresponding to each of the first regions to be determined in the right cerebral hemisphere region.

[0093] That is to say, the first regions to be determined correspond to the second regions to be determined one by one, and the first regions to be determined are mirror-symmetric with the corresponding second regions to be determined, which facilitates subsequent comparison.

[0094] Perform step S02: acquiring a first time-to-peak required for the contrast agent density to reach a peak in each first region to be determined, and determining the first region to be determined corresponding to the first time-to-peak exceeding a threshold as an infarcted region.

[0095] Perform step S03: acquiring a second time-to-peak required for the contrast agent density to reach a peak in each second region to be determined, and determining the second region to be determined corresponding to the second time-to-peak exceeding the threshold as an infarcted region.

[0096] After performing step S02 and step S03, it can be determined which of the first regions to be determined and the second regions to be determined are infarcted regions (and it can also be determined which are non-infarcted regions).

[0097] Perform step S04: accumulating and comparing the number of infarcted regions in the left cerebral hemisphere region and the right cerebral hemisphere region respectively, and determining the one with a smaller number of infarcted regions as the non-infarcted side hemisphere region.

[0098] For example, the total number of infarcted regions in the left cerebral hemisphere region is N, and the total number of infarcted regions in the right cerebral hemisphere region is M. If N < M, the left cerebral hemisphere region is determined as the non-infarcted side hemisphere region; if N > M, the right cerebral hemisphere region is determined as the non-infarcted side hemisphere region.

[0099] It should be noted that the analysis of the infarction status in patients with ischemic stroke can be obtained by analyzing the patient's imaging results and comparing parameters such as cerebral blood volume (CBV), cerebral blood flow (CBF), and time to peak contrast agent (TTP) in the left and right hemispheres. Accurate determination of the infarct side depends on generating the aforementioned parameter maps and making a comprehensive judgment; however, selecting appropriate arterial and venous feature points is a necessary prerequisite for generating these parameter maps. To resolve this contradiction, this embodiment directly obtains the data on the change of contrast agent density over time corresponding to the brain perfusion images. Contrast agent density information in the time dimension is extracted from the brain perfusion data (this type of data does not depend on the generation of parameter maps but can be obtained during the patient's examination), and the contrast agent density information at the corresponding mirror positions in both hemispheres is compared to adaptively extract the infarct area. Based on the quantitative analysis of the infarct areas in both hemispheres, the determination of the infarcted and non-infarcted hemisphere regions can be obtained.

[0100] Furthermore, after identifying the non-infarcted hemisphere and the infarcted hemisphere, in step S13, after selecting several arterial feature points, the arterial feature points can be screened again to exclude those within the infarcted hemisphere. This ensures that all obtained arterial feature points are located within the non-infarcted hemisphere.

[0101] Step S3: Perform perfusion analysis using the selected vein feature points and the selected artery feature points.

[0102] Subsequently, the perfusion parameter map can be calculated using the time density curves of the selected venous feature points and the selected arterial feature points (obtained in step S2). The perfusion parameter map is then analyzed to obtain the analysis results of the infarct area, displaying the mismatch ratio between the calculated infarct area volume and the penumbra volume. Furthermore, the perfusion analysis can obtain hemodynamic parameters and perfusion image representations such as cerebral blood volume (CBV), cerebral blood flow (CBF), mean transit time (MTT), and time to peak (TTP) of the contrast agent.

[0103] It should be noted that the method of obtaining the vein and artery feature points using the aforementioned feature point generation model may not yield correct feature points for uncommon data (such as artifact data, under-perfusion data, etc.), leading to perfusion analysis failure or inaccurate perfusion analysis. In such cases, the perfusion analysis can be performed again.

[0104] Figure 9This is a flowchart illustrating the re-perfusion analysis provided in this embodiment. (See attached flowchart.) Figure 9 As shown, the re-perfusion analysis includes steps S4 and S5.

[0105] Step S4: Based on the received selection instruction, determine the vein replacement feature points and / or artery replacement feature points in the brain perfusion image.

[0106] Users can use input devices such as a mouse to click and input selection commands on the interactive interface to determine vein replacement feature points and / or artery replacement feature points in the brain perfusion image. The vein replacement feature point is the location that the user considers to be more ideal for the vein feature point selected in step S2, and the artery replacement feature point is the location that the user considers to be more ideal for the artery feature point selected in step S2.

[0107] Preferably, the positions of all the vein feature points and artery feature points can be displayed on the Maximum Intensity Projection (MIP) map of the interactive interface for easy observation. In this way, the user can directly click on the MIP map of the interactive interface to input the selection command, thereby manually determining the vein replacement feature points and the artery replacement feature points. This setting enables efficient point selection using clearer images on the MIP map and obtains more reliable temporal density curves.

[0108] In some embodiments, the location markers of the selected vein feature points and the selected artery feature points may be displayed only on the temporal maximum density projection map of the brain perfusion image.

[0109] Step S5: Replace the selected vein feature points with the vein replacement feature points, and / or replace the selected artery feature points with the artery replacement feature points, and perform perfusion analysis.

[0110] In other words, the vein feature points determined by the feature point generation model are replaced with the manually determined vein replacement feature points, and / or the artery feature points determined by the feature point generation model are replaced with the manually determined artery replacement feature points, and then perfusion analysis is performed again, giving full play to the user's subjective initiative.

[0111] In summary, combining Figure 2As shown, in this embodiment, the brain perfusion image is first preprocessed and cached. Then, through an automatic point selection step, several venous feature points and several arterial feature points are obtained. Then, some venous feature points and / or arterial feature points are removed based on the first and second predetermined conditions. Next, the venous feature points are scored according to the third predetermined condition, and the arterial feature points are scored according to the fourth predetermined condition (or a combination of venous and arterial feature points is scored according to the fifth predetermined condition), resulting in a venous feature point representing the whole-brain perfusion output and an arterial feature point representing the whole-brain perfusion input. Perfusion analysis is then performed using the selected venous and arterial feature points. Before or after perfusion analysis, if the selection instruction is received, a manual point selection step can be performed, replacing the selected venous feature points with manually determined venous replacement feature points, and / or replacing the selected arterial feature points with manually determined arterial replacement feature points, before performing perfusion analysis.

[0112] This embodiment combines automatic and manual point selection modes, leveraging the advantages of both while overcoming their respective disadvantages. The manual point selection mode serves both as a verification and correction of the feature points obtained from the automatic selection mode, and as a review of those points, further enhancing the reliability of the point selection and significantly increasing user acceptance of the feature points and corresponding parameter map results. Figure 10 This is a structural block diagram of the perfusion analysis system provided in this embodiment.

[0113] like Figure 10 As shown, the perfusion analysis system includes:

[0114] Feature point determination module 10 is used to determine several venous feature points and several arterial feature points in brain perfusion images;

[0115] The filtering module 20 is used to filter the vein feature points using the time density curves corresponding to the vein feature points and the artery feature points, to obtain one of the vein feature points representing the whole brain perfusion output, and one of the artery feature points representing the whole brain perfusion input; and,

[0116] The perfusion analysis module 30 is used to perform perfusion analysis using the selected venous feature points and arterial feature points.

[0117] Furthermore, the perfusion analysis system also includes a user interaction module with a user interface for displaying the venous feature points and the arterial feature points. The user can also click on the user interface to input selection commands. In response to the selection command, the user interaction module determines venous replacement feature points and / or arterial replacement feature points in the brain perfusion image, and replaces the venous feature points selected by the filtering module with the venous replacement feature points, and / or replaces the selected arterial feature points with the arterial replacement feature points. Afterward, the perfusion analysis module 30 can perform perfusion analysis.

[0118] The present invention also provides an electronic device, please refer to... Figure 11 The diagram illustrates a block structure of an electronic device according to an embodiment of the present invention. Figure 11 As shown, the electronic device includes a memory 103 storing a computer program; a processor 101, which is communicatively connected to the memory 103 and executes any of the brain perfusion image feature point determination methods described above when the computer program is invoked; and a display 105, which is communicatively connected to the processor and the memory and is used to display a GUI interactive interface related to the determination of brain perfusion image feature points.

[0119] Since the electronic device provided by this invention and the brain perfusion image feature point determination method described above belong to the same inventive concept, the electronic device provided by this invention has all the advantages of the brain perfusion image feature point determination method described above. Therefore, the beneficial effects of the electronic device provided by this invention will not be described in detail here.

[0120] like Figure 11 As shown, the electronic device also includes a communication interface 102 and a communication bus 104, wherein the processor 101, the communication interface 102, and the memory 103 communicate with each other via the communication bus 104. The communication bus 104 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 104 can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not indicate that there is only one bus or one type of bus. The communication interface 102 is used for communication between the aforementioned electronic device and other devices.

[0121] The processor 101 referred to in this invention can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor 101 is the control center of the electronic device, connecting various parts of the entire electronic device through various interfaces and lines.

[0122] The memory 103 can be used to store the computer program. The processor 101 implements various functions of the electronic device by running or executing the computer program stored in the memory 103 and calling the data stored in the memory 103.

[0123] The memory 103 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0124] This invention also provides a readable storage medium storing a computer program that, when executed, implements any of the perfusion analysis methods described above. Since the readable storage medium provided by this invention belongs to the same inventive concept as the perfusion analysis methods described above, it possesses all the advantages of the perfusion analysis methods described above. Therefore, the beneficial effects of the readable storage medium provided by this invention will not be elaborated upon here.

[0125] The readable storage medium of embodiments of the present invention can be any combination of one or more computer-readable media. The readable medium can be a computer-readable signal medium or a computer-readable storage medium. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections having one or more wires, portable computer hard disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device.

[0126] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber, RF, etc., or any suitable combination thereof.

[0127] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0128] It should also be noted that although the present invention has been disclosed above with reference to preferred embodiments, these embodiments are not intended to limit the present invention. For any person skilled in the art, many possible variations and modifications can be made to the technical solutions of the present invention based on the disclosed technical content, or equivalent embodiments can be modified accordingly, without departing from the scope of the present invention. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the present invention shall still fall within the scope of protection of the present invention.

[0129] It should also be understood that, unless otherwise specified or indicated, the terms “first,” “second,” “third,” etc., in the specification are used only to distinguish the various components, elements, and steps in the specification, and not to indicate the logical or sequential relationships between the various components, elements, and steps.

[0130] Furthermore, it should be recognized that the terminology described herein is used only to describe particular embodiments and not to limit the scope of the invention. It must be noted that the singular forms “a” and “an” used herein and in the appended claims include plural bases unless the context clearly indicates otherwise. For example, a reference to “a step” or “an apparatus” means a reference to one or more steps or apparatuses, and may include secondary steps and secondary apparatuses. All conjunctions used should be understood in the broadest sense. And the word “or” should be understood to have the definition of logical “or” rather than logical “exclusive OR”, unless the context clearly indicates otherwise. Furthermore, implementation of embodiments of the invention may include performing selected tasks manually, automatically, or in combination.

Claims

1. A perfusion analysis method, characterized in that, include: Using data on the change of contrast agent density over time in the left and right hemispheres of the brain from brain perfusion images, the non-infarcted hemispheres of the brain are determined. Several venous feature points and several arterial feature points were identified in the brain perfusion images, wherein the arterial feature points were located in the non-infarcted hemisphere region. By using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points, one of the vein feature points that represents the whole brain perfusion output is obtained, and one of the artery feature points that represents the whole brain perfusion input is obtained. as well as, Perfusion analysis was performed using the selected venous and arterial feature points. The steps for determining the non-infarcted hemisphere of the brain include: Within the left hemisphere, several first regions to be determined are identified, and within the right hemisphere, second regions to be determined are identified that correspond to each of the first regions to be determined in a mirror image. The first peak time required for the contrast agent density of each first region to be determined to reach its peak value is obtained, and the first region to be determined corresponding to the first peak time exceeding a threshold is determined to be an infarct region. The second peak time required for the contrast agent density of each second region to be determined to reach its peak value is obtained, and the second region to be determined corresponding to the second peak time exceeding the threshold is determined to be an infarct region. The number of infarcted regions in the left and right hemispheres is summed and compared, and the hemisphere with the smaller number of infarcted regions is determined to be the non-infarcted hemisphere.

2. The perfusion analysis method as described in claim 1, characterized in that, The perfusion analysis method further includes: Based on the received selection instructions, venous replacement feature points and / or arterial replacement feature points are determined in the brain perfusion image; and, The selected vein feature points are replaced with the vein replacement feature points, and / or the selected artery feature points are replaced with the artery replacement feature points, and perfusion analysis is performed.

3. The perfusion analysis method as described in claim 2, characterized in that, The selected vein feature points and the selected artery feature points are displayed on the temporal maximum density projection map of the brain perfusion image at least by marking their positions. Based on the received selection instruction, the vein replacement feature points and / or the artery replacement feature points are determined on the temporal maximum density projection map.

4. The perfusion analysis method according to any one of claims 1 to 3, characterized in that, After determining a number of the venous feature points and a number of the arterial feature points in the brain perfusion image, the perfusion analysis method further includes: Using the anatomical location of the vein feature point and / or its corresponding time density curve, determine whether the vein feature point meets a first predetermined condition; using the anatomical location of the artery feature point and / or its corresponding time density curve, determine whether the artery feature point meets a second predetermined condition; and, Remove the vein feature points that do not meet the first predetermined condition and the artery feature points that do not meet the second predetermined condition.

5. The perfusion analysis method as described in claim 1, characterized in that, The steps for screening using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points include: All vein feature points are scored using the time density curves corresponding to the vein feature points and a third predetermined condition; all artery feature points are scored using the time density curves corresponding to the artery feature points and a fourth predetermined condition; and... The vein feature point with the highest score is used to represent the whole brain perfusion output, and the artery feature point with the highest score is used to represent the whole brain perfusion input.

6. The perfusion analysis method as described in claim 1, characterized in that, The steps for screening using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points include: Combine any one of the vein feature points and any one of the artery feature points; The combination is scored using the relative relationship of the time density curves corresponding to the vein feature points and the artery feature points in each combination, and a fifth predetermined condition; and, The vein feature points and artery feature points in the highest-scoring combination are used to represent the whole brain perfusion output and input, respectively.

7. A perfusion analysis system, characterized in that, include: The feature point determination module is used to determine several venous feature points and several arterial feature points in the brain perfusion image after determining the non-infarcted hemisphere region of the brain, wherein the arterial feature points are located in the non-infarcted hemisphere region. The filtering module is used to filter the vein feature points using the time density curves corresponding to the vein feature points and the time density curves corresponding to the artery feature points, to obtain one of the vein feature points that represents the whole brain perfusion output, and to obtain one of the artery feature points that represents the whole brain perfusion input. as well as, The perfusion analysis module is used to perform perfusion analysis using the selected venous feature points and arterial feature points; The steps for determining the non-infarcted hemisphere of the brain include: In the brain perfusion image, several first regions to be determined are identified in the left hemisphere region of the brain, and second regions to be determined are identified in the right hemisphere region of the brain, which are mirror images of each of the first regions to be determined. The first peak time required for the contrast agent density of each first region to be determined to reach its peak value is obtained, and the first region to be determined corresponding to the first peak time exceeding a threshold is determined to be an infarct region. The second peak time required for the contrast agent density of each second region to be determined to reach its peak value is obtained, and the second region to be determined corresponding to the second peak time exceeding the threshold is determined to be an infarct region. The number of infarcted regions in the left and right hemispheres is summed and compared, and the hemisphere with the smaller number of infarcted regions is determined to be the non-infarcted hemisphere.

8. The perfusion analysis system as described in claim 7, characterized in that, Also includes: The user interaction module is used to display the vein feature points and the artery feature points, and in response to a selection command, to determine vein replacement feature points and / or artery replacement feature points in the brain perfusion image, and to replace the vein feature points selected by the filtering module with the vein replacement feature points, and / or to replace the selected artery feature points with the artery replacement feature points.

Images