Disclosure of Invention
In view of the above, the present invention proposes an intracranial collateral circulation automatic evaluation method, apparatus, storage medium, and computing device that overcome or at least partially solve the above problems.
According to a first aspect of the present invention, there is provided an intracranial collateral circulation automatic evaluation method, the method comprising:
obtaining a cerebral blood flow image based on the ASL image of the multiple PLD times;
segmenting the cerebral blood flow image to obtain an artery arrival artifact region of interest;
calculating a quantitative parameter for performing intracranial collateral circulation evaluation based on the artery arrival artifact region of interest;
marking the arterial arrival artifact region of interest and the quantification parameter in the cerebral blood flow image.
Optionally, the obtaining a brain blood flow image based on the multiple PLD time ASL image includes:
acquiring ASL images of multiple PLD times, and registering the ASL images of the PLDs;
obtaining a cerebral blood flow image based on the ASL images of the PLD times after registration; the registered ASL images include an ASL proton density map and an ASL perfusion map.
Optionally, the obtaining a cerebral blood flow image based on the registered ASL images at the PLD times includes:
for each ASL perfusion image after registration, taking the PLD time as a sample and the pixel value of the ASL perfusion image as a weight, and calculating the weighted average value of the PLD time pixel by pixel;
calculating the average value of the weighted average values corresponding to each PLD time as a time reference value;
and searching for a target PLD time which is larger than the time reference value and has the minimum difference with the time reference value, and calculating according to the ASL proton density map and the ASL perfusion map corresponding to the target PLD time to obtain a cerebral blood flow image.
Optionally, the segmenting the cerebral blood flow image to obtain an arterial arrival artifact region of interest includes:
acquiring a brain structure image corresponding to the ASL image position, and registering by taking an ASL proton density image as a fixed image and the brain structure image as a floating image to obtain a registered target brain structure image;
acquiring a standard brain structure image, and registering the standard brain structure image and the target brain structure image so as to map an occipital lobe area, a frontal lobe area and a temporal lobe area in the standard brain structure image to the cerebral blood flow image;
determining an artery arrival artifact region of interest based on pixel values of an occipital lobe region, a frontal lobe region and a temporal lobe region in the cerebral blood flow image; the artery arrival artifact region of interest comprises an artery arrival artifact high signal area and an artery arrival artifact high signal area.
Optionally, the determining an artery arrival artifact region of interest based on pixel values of an occipital lobe region, a frontal lobe region and a temporal lobe region in the cerebral blood flow image comprises:
calculating the median value of pixels in the occipital lobe area in the cerebral blood flow image;
and comparing pixel values and the pixel median values in the frontal lobe area and the temporal lobe area in the cerebral blood flow image so as to mark an artery arrival artifact high signal area and an artery arrival artifact high signal area according to a comparison result.
Optionally, the quantization parameter includes one or more of an abnormal region volume, an abnormal side, a mismatch volume, and a mismatch ratio.
Optionally, the calculating a quantification parameter for performing an intracranial collateral circulation evaluation based on the artery arrival artifact region of interest includes:
counting a first abnormal area volume corresponding to the high signal area and a second abnormal area volume corresponding to the high signal area;
calculating a symmetry axis by adopting a symmetry axis detection algorithm, and marking a hemisphere where the high signal or the apparent high signal is positioned as an abnormal side;
calculating a sum of abnormal lateral frontal and temporal lobe regional volumes, calculating a mismatch volume based on the sum of abnormal lateral frontal and temporal lobe regional volumes, a first abnormal regional volume, and a second abnormal regional volume;
and calculating the ratio of the first abnormal region volume, the second abnormal region volume and the sum of the abnormal lateral frontal lobe and temporal lobe region volumes as a mismatch ratio.
According to a second aspect of the present invention, there is provided an intracranial sidebranch circulation automatic evaluation apparatus comprising:
the cerebral blood flow image acquisition module is used for acquiring a cerebral blood flow image based on the ASL image of the multi-PLD time;
an ATA region segmentation module used for segmenting the cerebral blood flow image to obtain an artery arrival artifact region of interest;
a quantitative parameter calculation module for calculating quantitative parameters for intracranial collateral circulation evaluation based on the artery arrival artifact region of interest;
a result display module for marking the artery arrival artifact region of interest and the quantification parameter in the cerebral blood flow image.
According to a third aspect of the present invention, there is provided a computer-readable storage medium for storing program code for executing the intracranial collateral circulation automatic evaluation method according to any one of the first aspects.
According to a fourth aspect of the invention, there is provided a computing device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the intracranial collateral loop automatic evaluation method of any one of the first aspects according to instructions in the program code.
The invention provides an intracranial collateral circulation automatic evaluation method, a device, a storage medium and a computing device, which are used for correcting a CBF image based on multiple PLD (focal resolved volumes) ASL (amplitude-dependent gradients), so that the CBF is more consistent with the real perfusion condition, and the defect of calculating the CBF by using the existing single PLD-ASL and multiple PLD-ASL is overcome.
Furthermore, through high signal and high signal area segmentation and quantization, new quantization parameters including abnormal area volume, abnormal side, mismatch volume and MismatchRatio are calculated, and the opening condition of the collateral circulation is reflected through the parameters, so that the severity of the arterial arrival artifact is evaluated more objectively, and the collateral circulation is evaluated; the problem caused by subjective evaluation of threshold judgment when the artery reaches the artifact ATA is solved, and the problem caused by subjective evaluation such as low consistency of repeated evaluation is solved.
The above description is only an overview of the technical solutions of the present invention, and the present invention can be implemented in accordance with the content of the description so as to make the technical means of the present invention more clearly understood, and the above and other objects, features, and advantages of the present invention will be more clearly understood.
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 thereof, taken in conjunction with the accompanying drawings.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The input to the algorithm is a T1, ASL sequence for the same patient, where ASL includes two parts, proton density map (ASL-M0) and perfusion map (ASL-PWI). The detailed implementation steps of the algorithm are as follows:
an embodiment of the present invention provides an intracranial collateral circulation automatic evaluation method, and as shown in fig. 1, the intracranial collateral circulation automatic evaluation method provided by the embodiment of the present invention may at least include the following steps S101 to S104.
S1, acquiring a cerebral blood flow image based on the ASL image of the multi-PLD time.
S2, segmenting the cerebral blood flow image to obtain an area of interest of an artery arrival artifact ATA;
s3, calculating quantitative parameters for intracranial collateral circulation evaluation based on the region of interest of the artery arrival artifact;
and S4, marking the artery arrival artifact interested region and the quantification parameter in the cerebral blood flow image.
After a period of time, the labeled blood reaches the capillary, and image acquisition can be performed at this time, wherein the time interval from the labeling to the acquisition is the PLD (Post Label Delay) time. Different PLD times may reflect different perfusion results and perfusion behavior. In the embodiment, a cerebral blood flow image (CBF image) is obtained based on the ASL image of multiple PLDs, so that the defect of CBF calculation by the existing single PLD-ASL is overcome, the cerebral blood flow image is more consistent with the real perfusion condition, and accurate intracranial collateral circulation evaluation parameters are obtained.
In some embodiments, the obtaining a brain blood flow image based on the multi PLD time ASL image in step S1 may include:
s1-1, acquiring ASL images of multiple PLD times, and registering the ASL images of the PLDs; optionally, the ASL images of the PLDs may be rigidly registered to prevent head misalignment during image acquisition by the patient from causing resultant deviations.
And S1-2, acquiring a cerebral blood flow image based on the ASL image of each PLD time after registration. After the ASL images of the PLD times after the registration are obtained, a cerebral blood flow image, which is referred to as a CBF image in this embodiment, can be obtained. Wherein, the ASL images of each PLD time after registration can comprise two parts of an ASL proton density map (ASL-M0) and an ASL perfusion map (ASL-PWI).
In an alternative embodiment of the present invention, the obtaining of the cerebral blood flow image based on the ASL image of each PLD time after the registration in step S1-2 may further include:
s1-2-1, calculating the weighted average value of PLD time pixel by taking PLD time as a sample and pixel values of ASL perfusion maps as weights for each registered ASL perfusion map; calculating the average value of the weighted average values corresponding to the PLD time as a time reference value;
that is, for each ASL-PWI sequence after registration, taking the PLD time corresponding to each ASL-PWI sequence as a sample, taking the pixel value of the ASL-PWI as a weight value, calculating the weighted average value delta i of the PLD time pixel by pixel, and then calculating the average value of all delta i to obtain delta, which is used as a time reference value.
S1-2-2, searching for a target PLD time which is larger than the time reference value delta and has the minimum difference with the time reference value delta, and calculating according to an ASL proton density map and an ASL perfusion map corresponding to the target PLD time to obtain a cerebral blood flow image.
And finding the PLD time which is larger than the delta and is closest to the time reference value delta according to the delta to serve as a target PLD time, and calculating according to the ASL-PWI and the ASL-M0 of the target PLD time to obtain a CBF image according to the following formula.
Wherein, ω is i For PLD time, R 1a The longitudinal relaxation rate of blood, tau the duration of the marking pulse, lambda the blood brain partition coefficient, alpha the PCASL marking efficiency, Δ M the perfusion weighted image ASL-PWI, M0 the proton density image ASL-M0. The signals of each pixel point in the ASL sequence are changed from noise to signals and then to noise along with the time lapse of PLD, and in this embodiment, the PLD time with the highest signal contribution can be found by a weighted average method, and the ASL sequence with the highest signal-to-noise ratio is found as a reference for calculating the cerebral blood flow graph.
After the cerebral blood flow image is obtained, it can be segmented to obtain the region of interest of the arterial arrival artifact. In some embodiments, the segmenting the cerebral blood flow image to obtain the artery arrival artifact region of interest in step S2 includes:
s2-1, acquiring a brain structure image corresponding to the ASL image position, and registering by taking an ASL proton density map as a fixed image and the brain structure image as a floating image to obtain a registered target brain structure image; the brain structure image of the present embodiment can be understood as a T1 image, the T1 image is easily obtained, and the brain structure is clear.
And carrying out three-dimensional rigid registration on the T1 image and the ASL-M0, wherein the T1 sequence is a moving image, and the ASL-M0 is a fixed image, so as to obtain a registered T1 sequence.
S2-2, acquiring a standard brain structure image, and registering the standard brain structure image and the target brain structure image to map an occipital lobe area, a frontal lobe area and a temporal lobe area in the standard brain structure image to the cerebral blood flow image;
the standard brain structure image of the embodiment may be a template of a brain structure image, the standard brain structure image and the T1 image after registration are subjected to non-rigid registration, and occipital lobe, frontal lobe and temporal lobe areas of the standard brain structure are mapped to the CBF image. In the embodiment of the invention, the T1 sequence is an auxiliary positioning occipital lobe, frontal lobe and temporal lobe brain partition, and the sequence can be replaced by any MR image which can easily distinguish brain tissue structure information.
S2-3, determining an artery arrival artifact region of interest based on pixel values of an occipital lobe region, a frontal lobe region and a temporal lobe region in the cerebral blood flow image; the artery arrival artifact region of interest comprises an artery arrival artifact high signal area and an artery arrival artifact high signal area.
In some embodiments, the step S2-3 may include:
s2-3-1, calculating a pixel median med _ occipital of the occipital lobe area in the cerebral blood flow image;
s2-3-2, comparing pixel values and pixel median values in frontal lobe areas and temporal lobe areas in the cerebral blood flow images to mark arterial arrival artifact high signal areas and arterial arrival artifact high signal areas according to comparison results.
In this embodiment, the high signal region is defined as a region with a CBF value 2 times higher than the occipital lobe signal, and the high signal-appearing region is defined as a region with a CBF value 1.5 times higher than the occipital lobe and less than 2 times higher than the occipital lobe, so that after the pixel median med _ occipital of the occipital lobe region is calculated, two thresholds, i.e., 2 med _occipitaland 1.5 med _occipital, can be determined, and the regions with pixel values greater than 2 med u occipital in the frontal lobe region and the temporal lobe region are marked as artery arrival artifact high signal regions ATAhigh 1 Marking the area with the pixel value more than 1.5-med _occipitaland less than 2-med _occipitalin the frontal area and the temporal area as an artery arrival artifact high signal area ATAhigh 2 . For the high signal region ATAhigh 1 And an area ATAhigh showing a high signal 2 Morphological treatment is carried out.
In practical applications, for the segmentation of the high signal region and the high signal area, a segmentation network model based on deep learning may be used to perform image segmentation, for example, the high signal region and the high signal area are labeled in advance, and these data are used to perform training in the deep learning network, so as to realize intelligent segmentation of the high signal region and the high signal area.
Further, after the arterial arrival artifact high signal region and the arterial arrival artifact high signal region are obtained, a quantitative parameter for performing intracranial collateral circulation evaluation can be calculated. The quantitative parameters for performing intracranial and collateral branch cycle evaluation in this embodiment may include one or more of an abnormal region volume, an abnormal side, a mismatched volume, and a mismatch ratio, and the specific calculation manner of each quantitative parameter is as follows:
volume of abnormal region: counting a first abnormal area volume Vol _ ATAhigh1 corresponding to the high signal area and a second abnormal area volume Vol _ ATAhigh2 corresponding to the high signal area;
an abnormal side: calculating a symmetry axis by adopting a symmetry axis detection algorithm, and marking a hemisphere where the high signal or the height-displaying signal is positioned as an abnormal side;
mismatch volume (Mismatch volume): calculating a sum Vol _ roi of abnormal lateral frontal lobe and temporal lobe regional volumes, calculating a mismatch volume based on the sum of abnormal lateral frontal lobe and temporal lobe regional volumes, a first abnormal regional volume and a second abnormal regional volume; misatcchvol = Vol _ roi- (Vol _ ATAhigh1+ Vol _ ATAhigh 2)
Mismatch Ratio: calculating the ratio of the first abnormal region volume to the second abnormal region volume to the sum of the abnormal lateral frontal lobe and temporal lobe region volumes as a mismatch ratio; misatchratio = (Vol _ ATAhigh1+ Vol _ ATAhigh 2)/Vol _ roi.
Finally, the high signal area and the high signal area can be marked in the CBF result map with different colors, and meanwhile, the calculated quantization parameters can also be displayed in the result map.
According to the intracranial collateral circulation automatic evaluation method provided by the embodiment of the invention, the CBF image is corrected based on the ASL of multiple PLDs, so that the CBF is more consistent with the real perfusion condition; through high signal and high signal area segmentation and quantization, new quantization parameters including abnormal area volume, abnormal side, mismatch volume and MismatchRatio are calculated, and the side branch circulation opening condition is reflected through the parameters, so that the severity of the arterial arrival artifact is evaluated more objectively, and then the side branch circulation is evaluated; compared with other collateral circulation scoring algorithms, such as MRA scoring, the ATA assessment mechanism can quantitatively display some diseases which cannot be detected by MRA, can better assist doctors in judging and analyzing the disease conditions of patients more clearly in clinical work, and the ATA collateral circulation assessment method is wider in application range and stronger in robustness.
Based on the unified inventive concept, an embodiment of the present invention further provides an intracranial collateral circulation automatic evaluation device, as shown in fig. 2, the intracranial collateral circulation automatic evaluation device includes:
a cerebral blood flow image obtaining module 210, configured to obtain a cerebral blood flow image based on the ASL image of multiple PLD times;
an ATA region segmentation module 220, configured to segment the cerebral blood flow image to obtain an artery arrival artifact region of interest;
a quantization parameter calculation module 230, configured to calculate a quantization parameter for performing intracranial collateral circulation evaluation based on the artery arrival artifact region of interest;
a result display module 240 for marking the artery arrival artifact interest region and the quantification parameter in the cerebral blood flow image.
In an optional embodiment of the present invention, the cerebral blood flow image obtaining module 210 may be further configured to:
acquiring ASL images of multiple PLD time, and registering the ASL images of the PLDs;
obtaining a cerebral blood flow image based on the ASL images of the PLD times after registration; the registered ASL images include an ASL proton density map and an ASL perfusion map.
In an optional embodiment of the present invention, the cerebral blood flow image obtaining module 210 may be further configured to:
for each ASL perfusion image after registration, taking the PLD time as a sample and the pixel value of the ASL perfusion image as a weight, and calculating the weighted average value of the PLD time pixel by pixel;
calculating the average value of the weighted average values corresponding to each PLD time as a time reference value;
and searching a target PLD time which is larger than the time reference value and has the minimum difference with the time reference value, and calculating according to the ASL proton density map and the ASL perfusion map corresponding to the target PLD time to obtain a cerebral blood flow image.
In an optional embodiment of the present invention, the ATA region splitting module 220 may further be configured to:
acquiring a brain structure image corresponding to the ASL image position, and registering by taking the ASL proton density map as a fixed image and the brain structure image as a floating image to obtain a registered target brain structure image;
acquiring a standard brain structure image, and registering the standard brain structure image and the target brain structure image so as to map an occipital lobe area, a frontal lobe area and a temporal lobe area in the standard brain structure image to the cerebral blood flow image;
determining an artery arrival artifact region of interest based on pixel values of an occipital lobe region, a frontal lobe region and a temporal lobe region in the cerebral blood flow image; the artery arrival artifact region of interest comprises an artery arrival artifact high signal area and an artery arrival artifact high signal area.
In an optional embodiment of the present invention, the ATA region splitting module 220 may further be configured to:
calculating the median value of pixels in the occipital lobe area in the cerebral blood flow image;
and comparing pixel values and the pixel median values in the frontal lobe area and the temporal lobe area in the cerebral blood flow image so as to mark an artery arrival artifact high signal area and an artery arrival artifact high signal area according to a comparison result.
In an optional embodiment of the present invention, the quantization parameter calculating module 230 may further be configured to:
counting a first abnormal area volume corresponding to the high signal area and a second abnormal area volume corresponding to the high signal area;
calculating a symmetry axis by adopting a symmetry axis detection algorithm, and marking a hemisphere where the high signal or the height-displaying signal is positioned as an abnormal side;
calculating a sum of abnormal lateral frontal and temporal lobe regional volumes, calculating a mismatch volume based on the sum of abnormal lateral frontal and temporal lobe regional volumes, a first abnormal regional volume, and a second abnormal regional volume;
and calculating the ratio of the first abnormal region volume, the second abnormal region volume and the sum of the abnormal lateral frontal lobe and temporal lobe region volumes as a mismatch ratio.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium is used to store a program code, and the program code is used to execute the intracranial collateral circulation automatic evaluation method according to the above embodiment.
An embodiment of the present invention further provides a computing device, where the computing device includes a processor and a memory: the memory is used for storing program codes and transmitting the program codes to the processor; the processor is used for executing the intracranial collateral circulation automatic evaluation method according to the embodiment according to the instructions in the program code.
It is clear to those skilled in the art that the specific working processes of the above-described systems, devices, modules and units may refer to the corresponding processes in the foregoing method embodiments, and for the sake of brevity, further description is omitted here.
In addition, the functional units in the embodiments of the present invention may be physically independent of each other, two or more functional units may be integrated together, or all the functional units may be integrated in one processing unit. The integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.
Those of ordinary skill in the art will understand that: the integrated functional units, if implemented in software and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computing device (e.g., a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention when the instructions are executed. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Alternatively, all or part of the steps of implementing the foregoing method embodiments may be implemented by hardware (such as a computing device, e.g., a personal computer, a server, or a network device) associated with program instructions, which may be stored in a computer-readable storage medium, and when the program instructions are executed by a processor of the computing device, the computing device executes all or part of the steps of the method according to the embodiments of the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present invention; such modifications or substitutions do not depart from the scope of the present invention.