CN109671069B

CN109671069B - Method and system for measuring core infarction volume based on skull CT image

Info

Publication number: CN109671069B
Application number: CN201811531887.1A
Authority: CN
Inventors: 金海岚; 宋凌; 印胤; 杨光明; 秦岚
Original assignee: Qianglian Zhichuang Beijing Technology Co ltd
Current assignee: Qianglian Zhichuang Beijing Technology Co ltd
Priority date: 2018-12-14
Filing date: 2018-12-14
Publication date: 2021-02-19
Anticipated expiration: 2038-12-14
Also published as: CN109671069A

Abstract

The embodiment of the specification discloses a method and a system for measuring the volume of a core infarction based on a skull CT image, belonging to the field of medical images. The embodiment of the specification solves the problems of large subjective difference and poor accuracy of a measurement result of the core infarction volume in the prior art by using a method for measuring the core infarction volume based on a skull CT image. The measuring method comprises the following steps: determining a frame where a target image layer is located from multi-frame skull CT image data to be processed; extracting a target area; performing infarction judgment on each target area contained in the target area respectively, and outputting an infarction judgment result; the core infarct volume is measured based on the infarct determination result of the target area. The measuring method and the system provided by the embodiment of the specification can quickly and accurately measure the core infarction volume, eliminate or reduce the interference of subjective factors, and provide objective basis for medical research such as cerebral apoplexy and the like as a computer-aided method.

Description

Method and system for measuring core infarction volume based on skull CT image

Technical Field

The specification relates to the technical field of medical images and computers, in particular to a method and a system for measuring the volume of a core infarction based on a skull CT image.

Background

The core infarction is one of important indexes for evaluating Acute Ischemic Stroke (AIS). The brain tissue in which irreversible damage occurs as a core infarction means an area in which the cerebral blood flow is decreased by more than 30% as compared with the normal brain tissue, and is indicated as a low-density area on NCCT (non-contrast CT). The size of the infarcted core is closely related to the clinical prognosis of the patient. The smaller the infarct core, the greater the likelihood of a good prognosis for the patient. Meanwhile, the evaluation of the infarction core can also predict the risk of complications in intravascular treatment, so that accurate evaluation of the infarction core is helpful for screening out stroke patients suitable for intravascular treatment. The volume of the core infarction is one of the indexes for evaluating the core infarction, and is a prediction index related to good prognosis of AIS patients.

Currently, in the prior art, the measurement of the volume of the core infarction mainly comprises two methods, namely manual measurement and computer-aided measurement. The key point of manually measuring the core infarction volume is manually delineating the infarction focus, and the method is very complicated and has great limitations in the aspects of accuracy, repeatability and objectivity. Computer-aided measurement of the core infarction volume is mostly measured by adopting an image segmentation method, but the method can lose some information in the processing process and influence the measurement result.

Therefore, a method for measuring the core infarction volume is needed at present, which can measure the core infarction volume quickly and accurately, eliminate or reduce the interference of subjective factors, and provide objective basis for medical research such as stroke and the like as a computer-aided method.

Disclosure of Invention

The embodiment of the specification provides a method and a system for measuring the volume of a core infarction based on a skull CT image, which are used for solving the following technical problems: the method can quickly and accurately measure the core infarction volume, eliminate or reduce the interference of subjective factors, and provide objective basis for medical research such as cerebral apoplexy and the like as a computer-aided method. The embodiment of the specification provides a method for measuring the volume of a core infarction based on a skull CT image, which comprises the following steps:

determining a frame where a target image layer is located from multi-frame skull CT image data to be processed, wherein the target image layer is a layer where skull CT image data used for measuring the core infarction volume in the skull CT image is located;

extracting a target area based on the frame where the target image layer is located, wherein the target area is an area used for measuring the volume of the core infarct in the skull CT image data;

performing infarction judgment on each target area contained in the target area, and outputting an infarction judgment result of the target area;

measuring a core infarct volume based on the infarct determination result of the target area.

Preferably, determining a frame where the target image layer is located from the to-be-processed multi-frame skull CT image data specifically includes:

extracting image characteristics of the multi-frame skull CT image data to be processed, and determining the distance from each frame of skull CT image data in the multi-frame skull CT image data to each target image layer;

and determining the frame of the target image layer according to the distance from the skull CT image data of each frame to each target image layer.

Preferably, extracting a target area based on the frame where the target image layer is located specifically includes:

and based on the frame of the target image layer, performing digital label classification on each pixel point in the skull CT image data, and outputting a two-dimensional matrix with digital labels corresponding to each pixel point in the frame of the target image layer, wherein the digital labels correspond to the target areas one by one, and the two-dimensional matrix corresponds to the frame of the target image layer one by one.

Preferably, the infarction judgment of each target area included in the target area and the output of the infarction judgment result of the target area specifically include:

and carrying out infarction judgment based on each pixel point contained in the target area, and outputting an infarction judgment result of the target area.

Further, performing an infarction judgment on each target area included in the target area, and outputting an infarction judgment result of the target area, further including:

judging whether the target area has infarction in the left and right brains according to the infarction judgment result of the target area;

and if the target area has infarctions in the left and right brains, further judging which side has the infarctions, and correcting the infarction judgment result of the target area to serve as a new infarction judgment result of the target area.

Preferably, the measuring of the core infarct volume based on the infarct determination result of the target area specifically includes:

determining the number of pixel points with cerebral infarction in each target area with infarction based on the infarction judgment result of the target area;

and determining the core infarction volume of each target area in which the cerebral infarction occurs according to the pixel point distance of the CT image and the layer thickness of the CT image and the number of the pixel points in each target area in which the cerebral infarction occurs, wherein the sum of the core infarction volumes of each target area is used as a measurement result of the core infarction volume.

The embodiment of this description provides a measurement system of volume of infarct based on head CT image, includes:

the region extraction module is used for extracting a target region from multi-frame skull CT image data to be processed, wherein the target region is a region for measuring the core infarction volume in the skull CT image data;

the infarction identification module is used for carrying out infarction judgment on each target area contained in the target area and outputting an infarction judgment result of the target area;

and the measuring module is used for measuring the core infarction volume based on the infarction judgment result of the target area.

Preferably, the extracting the target region from the multi-frame skull CT image data to be processed specifically includes:

respectively determining frames where a target image layer is located from multi-frame skull CT image data to be processed, wherein the target image layer is a layer where the skull CT image data used for measuring the core infarction volume in the skull CT image are located;

and extracting a target area based on the frame where the target image layer is located, wherein the target area is an area used for measuring the volume of the core infarct in the skull CT image data.

The embodiment of the specification adopts at least one technical scheme which can achieve the following beneficial effects:

the embodiment of the specification extracts the target areas from the multi-frame CT image data to be processed to respectively judge the infarctions, further measures the volume of the core infarctions, can quickly and accurately measure the volume of the core infarctions, eliminates or reduces interference of subjective factors, and provides objective basis for medical research such as stroke and the like as a computer-aided method.

Drawings

In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.

Fig. 1 is a frame diagram of a method for measuring a volume of a core infarct based on a cranial CT image according to an embodiment of the present disclosure;

fig. 2a is a schematic diagram of a target area for measuring core infarct volume in a nuclear bolus layer provided in an embodiment of the present disclosure;

fig. 2b is a schematic diagram of a target area for measuring core infarct volume in the upper layer of the nucleus provided in the embodiments of the present specification;

fig. 3a is a schematic diagram of the pons and bilateral cerebellar regions for post-circulatory measurement of core infarct volume provided in an embodiment of the present disclosure;

fig. 3b is a schematic diagram of a midbrain region for post-circulatory measurement of core infarct volume provided in an embodiment of the present description;

FIG. 3c is a schematic diagram of the bilateral thalamus and posterior cerebral artery donor area for post-circulatory measurement of core infarct volume provided in an embodiment of the present disclosure;

fig. 4 is a flowchart of a method for measuring a core infarct volume based on a skull CT image according to an embodiment of the present disclosure;

fig. 5 is a schematic diagram of determining a frame in which a layer of a kernel and a layer above the kernel are located according to an embodiment of the present disclosure;

fig. 6 is a schematic diagram of infarction determination provided in the embodiments of the present disclosure;

fig. 7 is a schematic diagram of a system for measuring a volume of a core infarct based on a cranial CT image according to an embodiment of the present disclosure.

Description of the acronyms

C-tail nucleus head

L bean-shaped core

IC inner capsule hind limb

I island belt

M1 anterior cortical region of middle cerebral artery

M2 lateral cortex region of the island lobe of the middle cerebral artery

M3 posterior cortex region of middle cerebral artery

Middle cerebral artery cortex above M4M 1

Middle cerebral artery cortex above M5M 2

Middle cerebral artery cortex above M6M 3

Description of the invention

In FIG. 3a, from top to bottom, the pons and bilateral cerebellum are shown;

FIG. 3c shows, from top to bottom, the bilateral thalamus and posterior cerebral artery supply regions;

the

numbers

1 and 2 in fig. 3 represent the weight of each region.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any inventive step based on the embodiments of the present disclosure, shall fall within the scope of protection of the present application.

Fig. 1 is a frame diagram of a method for measuring a volume of a core infarct based on a cranial CT image according to an embodiment of the present disclosure, including the following steps:

step S101: and respectively determining the frames of the target image layer from the multi-frame skull CT image data to be processed.

The CT image is composed of a certain number of pixels with different gray scales from black to white arranged in a matrix. The CT value is used for reflecting the density value of the corresponding tissue of the human body. Since the single-frame CT image is a slice image having a certain thickness, a cross-section is generally used, and a plurality of consecutive slice images are required in order to display the entire organ. Therefore, when the volume measurement of the core infarction is carried out, a series of multi-frame images are combined for observation and evaluation, and the skull CT image data to be processed in the embodiment is multi-frame.

In one embodiment of the present specification, 10 regions for volume measurement of the core infarct in the anterior circulation are located in the nucleus pulposus layer and the upper nucleus pulposus layer, respectively, as shown in fig. 2. Therefore, when the pre-cycle core infarction volume measurement is performed, frames where the nucleus layer and the upper layer of the nucleus are located need to be determined from the multi-frame skull CT image data to be processed, and the target image layer is the nucleus layer and the upper layer of the nucleus.

In another embodiment of the present specification, of the 8 regions used for the volume measurement of the post-circulatory core infarct, the pons and bilateral cerebellum are located in the cerebellum layer, the midbrain is located in the midbrain layer, and the bilateral thalamus and posterior cerebral artery supply regions are located in the cerebral layer, as shown in fig. 3 in particular. Therefore, when measuring the volume of the post-circulation core infarction, the frames of the cerebellum layer, the midbrain layer and the brain layer are determined from the multi-frame skull CT image data to be processed, and the target image layer is the cerebellum layer, the midbrain layer and the brain layer.

Step S103: and extracting a target area.

After the frame of the target image layer is determined in step S101, the image data of the frame of the target image layer needs to be further analyzed to determine whether each pixel point on the frame belongs to the target region and which region belongs to the target region, where the target region is a region for measuring the core infarct volume in the frame of the target image layer, and the measurement of the core infarct volume of the skull CT image can be finally achieved by determining the cerebral infarct of the target region. In one embodiment of the present description, the target regions may be 20 regions of the caudate nucleus head, the lenticular nucleus, the inner capsular hind limb, the zona insularis, and the M1-M6 region of the left and right brain. In another embodiment of the present disclosure, the target region may be a pons, bilateral cerebellum, midbrain, bilateral thalamus, and a region of blood supply to the posterior cerebral artery. And when the pixel does not belong to the target area, judging that the pixel belongs to the background area.

Step S105: and carrying out infarction judgment on the target area, and outputting an infarction judgment result of the target area.

After the target region is extracted in step S103, it needs to be further analyzed and determined which regions have infarcted. In the CT image, one CT value corresponds to one gray level. There are several ways to determine infarction, one exemplary way is based on the symmetry of the human brain: under normal conditions of human brain tissue, the mean values of the CT values at symmetric positions of the left and right brain are substantially the same, i.e. the gray values are substantially the same. However, if the brain tissue on one side is infarcted, the gray value is reduced, so that the infarct can be judged by comparing the gray value with the gray value of the corresponding area on the opposite side, so as to determine the infarct area, and further realize the infarct judgment of each target contained in the target area.

Besides the above method, another exemplary method is to extract image features, such as gray information, gray information of adjacent pixels, gradient information of adjacent pixels, and the like, and then train a model in a supervised learning manner, and can also be used for determining infarction.

Under normal circumstances, when cerebral infarction occurs, infarction occurs on only one side, and if infarction occurs on both sides, the brain tissue may have died. Therefore, if the infarction determination result in the target area is that both the left and right brains have infarcted, the infarction identification result in one side may be caused by noise such as image artifacts. Based on the above reasons, if the infarction judgment result is that the left and right brains are both infarcted, it is necessary to further judge which side is infarcted, so as to correct the infarction judgment result. In the specific correction, judgment based on gray scale change can be adopted. In the target area, the smaller the value of the ratio or the difference of the gray scale or the combination of the ratio and the difference, the higher the confidence, so comparing the minimum value of the ratio or the difference of the gray scale or the combination of the ratio and the difference in the target area, the side with the smaller value, i.e. the side with the higher confidence, is judged as infarcted, the other side is judged as non-infarcted, and the output infarct judgment result is corrected according to the confidence result.

Step S107: the core infarct volume is measured based on the infarct determination result of the target area.

Based on the infarction determination results of the target areas of the left and right brains in the foregoing step S105, the infarction state of each pixel point in the target area of infarction needs to be further determined. Specifically, a gray threshold is set, and the number of pixels with cerebral infarction in each target area is determined. And then calculating the volume of cerebral infarction in each target area according to the pixel point distance of the CT image, the layer thickness of the CT image and the number of the pixel points of cerebral infarction in each target area, wherein the sum of the volumes of the core infarction in all the target areas is used as a measurement result of the volume of the core infarction.

By adopting the method provided by the embodiment of the specification, the core infarction volume is measured on the skull CT image data, the core infarction volume can be measured quickly and accurately, the interference of subjective factors is eliminated or reduced, and the method is used as a computer-aided method and provides objective basis for medical research such as cerebral apoplexy.

To further illustrate the method for measuring the volume of the core infarct based on the cranial CT image, fig. 4 is a flowchart of a method for measuring the volume of the core infarct based on the cranial CT image provided in the embodiment of the present specification, and the measurement process is described in detail by taking the measurement of the volume of the core infarct in the previous cycle as an example.

Step S401: acquiring skull CT image data to be processed.

Since the size of the pixels in the CT image is different, the corresponding number is different. For example, the pixel size may vary from 1.0mm by 1.0mm, 0.5mm by 0.5mm, and the corresponding number may be 256 by 256, or 512 by 512. Obviously, the smaller the pixel size, the greater the number, and the higher the spatial resolution. Since the CT image is a slice image, a plurality of consecutive slice images are required to display the entire organ, and thus the skull CT image to be processed is a multi-frame image.

Step S403: and determining the frame where the nuclear group layer and the upper layer of the nuclear group are located.

In the multi-frame skull CT image data to be processed, the interest is the frame where the nuclear mass layer and the upper layer of the nuclear mass are located. Therefore, it is first required to determine the frame where the nuclear mass layer and the upper layer of the nuclear mass are located, and a neural network or other algorithms can be adopted to extract the features of a certain frame of image, wherein the features can be the gray level of the image, for example, an M region (an M1-M6 region is collectively referred to as an M region) is a cerebral cortex, mainly a gray level, and the gray level range is about 35; or the shape of the image, for example, the bean-shaped core is bean-shaped, the island strip is strip-shaped, and the gray level difference exists between the island strip and the periphery; the image may also be characterized by texture information, such as sulci and gyrus in the M region; position information can also be used as an extraction feature, such as an island band adjacent to an M region, which is proximate to the skull. And further estimating the distance from the frame image to the nucleus layer and the upper layer of the nucleus, determining the position of the frame image in the brain, and further determining the frame where the nucleus layer and the upper layer of the nucleus are located.

Fig. 5 is a schematic diagram illustrating distances from each frame of CT image to a nuclear bolus layer and a nuclear bolus upper layer according to an embodiment of the present disclosure. Specifically, one image sequence, with a layer spacing of 10mm, is 12 frames in total. The distances from each frame to the nuclei layer and the upper layer of the nuclei are calculated by using a neural network or other algorithms, as shown in fig. 5. It can be seen that the 5 th frame is closest to the bolus layer and the 7 th frame is closest to the upper layer of the bolus, so the 5 th frame is the bolus layer and the 7 th frame is the upper layer of the bolus. The distances from each frame to the nucleus layer and the nucleus upper layer are determined by other similar methods such as mean square error averaging, absolute error averaging and the like, and the method can also be used for determining the frames where the nucleus layer and the nucleus upper layer are located, and other similar methods are regarded as the same mode.

In an embodiment of the present specification, a nuclear group layer and a nuclear group upper layer in a CT image data sequence are artificially labeled, trained in a supervised learning manner, and a training model is obtained for classifying the nuclear group layer and the nuclear group upper layer and outputting [ P1, P2, P3], where P1 represents a probability that the image belongs to neither the nuclear group layer nor the nuclear group upper layer, P2 represents a probability of belonging to the nuclear group layer, P3 represents a probability of belonging to the nuclear group upper layer, and P1+ P2+ P3 is 1. When inputting a n frame CT image data, it will output a matrix of n x 3, if the P2 value of the k row is maximum and the P3 value of the P row is maximum, then the k frame is considered as the nuclear group layer, and the P frame is the nuclear group upper layer. By adopting the method, the frame where the nuclear group layer and the upper layer of the nuclear group are positioned can also be determined.

Step S405: based on the frame where the nuclear group layer and the upper nuclear group layer are located, target regions of the left and right brains are extracted.

In order to extract the target areas of the left and right brains, it is necessary to perform digital label classification on each pixel point in the frame where the nucleus layer and the nucleus upper layer are located, and determine to which area of the target areas of the left and right brains the pixel point belongs. In one embodiment of the present disclosure, 0 represents background, 1 represents left M1, 2 represents left M2, 3 represents left M3, 4 represents left caudate nucleus head, 5 represents left islet band, 6 represents left inner capsular limb, 7 represents left putamen, 8 represents right M1, 9 represents right M2, 10 represents right M3, 11 represents right caudate nucleus head, 12 represents right islet band, 13 represents right inner capsular limb, 14 represents right putamen, 15 represents left M4, 16 represents left M5, 17 represents left M6, 18 represents right M4, 19 represents right M5, and 20 represents right M6. The images of the frame where the nuclear group layer and the upper nuclear group layer are located can output a two-dimensional matrix with digital labels through a neural network or other algorithms, the values in the matrix are 0-20, and 1-20 represent the target area. If the input data is 512 × 512 images, a two-dimensional matrix with digital labels is output, the labels are 512 × 512 two-dimensional matrices, the two-dimensional matrices respectively correspond to the frame where the kernel group layer and the upper layer of the kernel group are located one by one, the digital labels in the two-dimensional matrix corresponding to the frame where the kernel group layer is located are 0-14, and the digital labels in the two-dimensional matrix corresponding to the frame where the upper layer of the kernel group is located are 0, 15-20. According to the two-dimensional matrix, the region to which each pixel point in the frame belongs can be obtained.

Step S407: and carrying out infarction judgment on each target area contained in the left and right brain target areas.

The CT value is an index for measuring relative density in CT images, and has a Hounsfield Unit (HU), and the CT value of human tissue has a fixed range, such as a CT value of 0 for water, 60-100Hu for blood, 30-35Hu for gray matter, 20-25Hu for white matter, 20-40Hu for muscle, and >1000Hu for bone. In the CT image, one CT value corresponds to one gray level. Due to the symmetry of human brain, under normal conditions, the mean values of CT values at symmetric positions of left and right brain are substantially the same, i.e. the gray values are substantially the same. However, if one side of the brain tissue is infarcted, the gray value is reduced, so that the infarct can be judged by comparing the gray value of the corresponding area of the opposite side to determine the infarcted area.

Based on the digital labels output in step S405, the same target regions of the left and right brains in the image data are located, all the pixels in the target regions are traversed, meanwhile, in order to eliminate interference such as noise, thresholds thre0 and thre1 are set, pixels with gray values between thre0 and thre1 are selected, thre0 may be set to 30Hu, and thre1 may be set to 50 Hu. And calculating an average value after summing the gray values of the pixel points meeting the conditions, taking the average value as an average gray value of the target area, and comparing the gray value of one side with the gray value of the opposite side so as to judge whether the target area is infarcted.

In one embodiment of the present specification, a method of extracting image features is used to perform infarction judgment. Specifically, image features such as gray information, gray information of adjacent pixels, gradient information and gradient information of adjacent pixels are extracted, artificial marking is carried out, and training is carried out by using a mode of supervised learning such as an SVM (support vector machine) and a random forest. And (3) applying the trained model to extract image characteristics of each pixel point in the image to obtain an infarction judgment result of each pixel point. Specifically, with this method, the outputted infarction judgment result is in a two-class manner, i.e., 0 indicates no infarction, and 1 indicates infarction.

Fig. 6 is a schematic diagram of infarction determination provided in the embodiments of the present specification. Specifically, the determination may be performed by a difference method, a ratio method, a combination of a ratio and a difference, and other infarction determination methods based on gray scale change are all regarded as the same methods as the present invention. If judging whether the lenticular nucleus is infarcted, locating the point (the number label is 7) of the left lenticular nucleus and the point (the number label is 14) of the right lenticular nucleus in the image data, traversing all the pixel points corresponding to the number labels 7 and 14, and respectively screening the pixel points with the gray values between 30Hu and 50Hu in the number labels 7 and 14. By using the screened pixel points of the digital label 7 and the digital label 14, calculating an average value of gray values of the pixel points of the digital label 7 as a gray value of the left lenticular nucleus, calculating an average value of gray values of the pixel points of the digital label 14 as a gray value of the right lenticular nucleus, and comparing the gray value of the left lenticular nucleus with the gray value of the right lenticular nucleus, thereby judging whether the lenticular nucleus is infarcted or not and which side of the lenticular nucleus is infarcted. Other calculation methods of the average gray value, such as gray stretching and calculating the average gray value after performing certain operation conversion on the gray value of each pixel point, are all regarded as the same methods as the present invention.

In an embodiment of the present specification, when the difference between the gray-scale value of one side and the gray-scale value of the opposite side in the same target area on the left and right brain sides is smaller than the preset threshold diff, the corresponding area in the CT image of the skull is considered to be infarcted, and the side with the relatively lower gray-scale value is the infarcted side. Generally, the threshold diff is considered to be-2. As in the example of fig. 6, the difference between the gray scale value of the left M2 region and the gray scale value of the right M2 region was-2.7, which is less than the threshold value of-2, and thus an infarction was considered to have occurred in the left M2 region. The infarct was judged in the same manner as left M3 and right M4.

In an embodiment of the present specification, a ratio method is used for infarction determination, and when a ratio of a gray-scale value of one side to a gray-scale value of the opposite side in the same target area on both sides of the left and right brains is smaller than a preset threshold ratio, it is considered that a corresponding area in a skull CT image is infarcted, and the side with a relatively low gray-scale value is an infarcted side. Generally, the threshold ratio is considered to be 0.96. As in the example of fig. 6, the ratio of the left-side lenticular nucleus to the right-side lenticular nucleus is 0.99, which is greater than 0.96, and it is considered that the lenticular nucleus has no infarction, which is consistent with the judgment result of the difference method. The ratio of the gray value of the left M2 area to the gray value of the right M2 area was 0.92 and less than 0.96, and the left M2 area was considered to be infarcted. In the same way, the left M3 area was judged to have infarct.

In one embodiment of the present disclosure, the infarction is determined by a combination of the ratio and the difference. The basis for infarction is considered to be that the ratio is less than 0.96 and the difference is less than-2. In the embodiment of fig. 6, the method is used for determining the infarction of the target area in the skull CT image, and the left M2, M3 and right M4 infarctions. The infarction judgment basis can be that the ratio is less than 0.96 or the difference is less than-2. In the embodiment of fig. 6, the infarction is judged by adopting the method, and the left M2 and M3 areas are infarcted.

Generally, the CT value of newly infarcted tissue is not lower than 20Hu, and if the brain tissue on the other side is not infarcted, the average gray value is about 40Hu, the ratio of the two sides is not less than 0.5, and the absolute value of the difference diff is not more than 20. In one embodiment of the present specification, if the target area of the CT image of the skull is determined to be infarcted using c ═ 0.5)/0.7+ (diff +20)/50, and if the value is equal to or less than 1, the target area of the lateral brain tissue is considered to be infarcted, and if the value is greater than 1, the target area is not infarcted. In the embodiment of fig. 6, the difference between the right M4 regions is-1.6, and the ratio is 0.95, calculated using the method described above: (0.95-0.5)/0.7+ (-1.6+20)/50 ═ 1.02>1, therefore the M4 zone was considered to have not infarcted. The method is adopted to judge that the left M2 and M3 areas have infarct.

After multi-frame skull CT image data to be processed finish infarction judgment, outputting 20 infarction judgment results of target areas for subsequent measurement of the core infarction volume. The infarction judgment result is an array which comprises three elements, wherein the first element of the array is a Boolean type number, 0 represents non-infarction, and 1 represents infarction; the second element is a floating point number, and the numerical value is the average gray value of the side divided by the average gray value of the opposite side; the third element is a floating point number, and the numerical value is the average gray value of the side minus the average gray value of the opposite side. The 1 st output is [0,1,0], where 0 indicates no infarction in the left M1 region, the mean gray value of the left M1 region divided by the mean gray value of the right M1 region is 1, and the mean gray value of the left M1 region minus the mean gray value of the right M1 region is 0.

Step S409: and carrying out denoising treatment on infarct pixel points in target areas of the left and right brains.

The foregoing step S407 realizes classification of each pixel point in the target area, and isolated infarct pixel points may occur among the pixel points in these areas, and these isolated infarct pixel points may be caused by factors such as image quality and device, and belong to interference noise, so that these noises need to be removed to avoid the influence of interference on subsequent measurement of the core infarct volume.

And (4) correcting the removed infarction judgment result of the isolated pixel points. If the infarction judgment result in the step 405 is output as [1,1], the pixel point is found to be an isolated pixel point by adopting a region-connected denoising method, so that the pixel point is considered not to have an infarction, and the infarction judgment result in the step S405 is corrected to [1,0 ].

Step S411: and further judging the infarction judgment results of the target areas of the left and right brains.

As a result of the infarction determination of the 20 target regions output in the foregoing step S409, there may be a case where infarctions occur in both sides of the target regions. Since only one side should be infarcted under normal conditions, it is necessary to further determine which side is infarcted if there is a bilateral infarction. The judgment can be carried out according to the fact that the probability of infarction on the side with smaller gray value change is higher. In the target areas of the left and right brains, the smaller the value of the ratio or the difference of the gray levels or the combination of the ratio and the difference, the higher the confidence level, so as to compare the minimum value of the ratio or the difference of the gray levels or the combination of the ratio and the difference in the target areas of the left and right brains, and consider that the side with the smaller value, i.e., the side with the higher confidence level, is an infarct, and the other side is not infarcted, and correct the infarct judgment result output in the step S409.

According to the method described in step S409, the embodiment shown in fig. 6 is infarct-judged by using a ratio method or a difference method, and the output infarct judgment results of the target areas 1-20 are as follows: 1:[0,0.99, -0.5],2:[1,0.92, -2.7],3:[1,0.93, -2.4],4:[0,1,0],5:[0,1.01,0.2],6:[0,1.03,0.9],7:[0,0.99, -0.5],8:[0,1.02,0.5],9:[0,1.09,2.7],10:[0,1.08,2.4],11:[0,1,0],12:[0,0.99, -0.2],13:[0,0.97, -0.9],14:[0,1.01,0.5],15:[0,1.05,1.6],16:[0,1,0],17:[0,1,0],18:[1,0.95, -1.6],19:[0,1,0],20:[0,1,0].

In an embodiment of the present invention, the outputted infarction determination result is further determined by using a ratio-based confidence method, where the minimum ratio of the left target area is 0.92, and the minimum ratio of the right target area is 0.95. Infarction occurred due to the high confidence on the side with a small ratio, and therefore left infarction. Furthermore, the outputted infarction judgment result is corrected to 18: [1,0.95, -1.6] as 18: [0,0.95, -1.6], other infarction judgment results are not changed, and the corrected infarction judgment results of the target areas of the left and right brains are used as new infarction judgment results for measuring the volume of subsequent core infarctions.

In an embodiment of the present invention, the outputted infarction determination result is further determined by a confidence method based on difference, wherein the minimum difference of the left target area is-2.7, and the minimum difference of the right target area is-1.6. Infarction occurs because of the high confidence on the side with small ratio, -2.7< -1.6, so left infarction. Further, the outputted infarction judgment result is corrected to 18: [1,0.95, -1.6] as 18: [0,0.95, -1.6], other infarction judgment results are not changed, and the corrected infarction judgment result is used as a new infarction judgment result for measuring the volume of the subsequent core infarction.

In an embodiment of the present invention, a confidence method combining the ratio and the difference is adopted to further determine the outputted infarction determination result. The method uses the function c ═ (ratio-0.5)/0.7+ (diff +20)/50 to calculate the c value of the infarction judgment result. Since the lower the c value, the higher the probability of infarction, i.e., the higher the confidence, the minimum c values on both sides are compared, and infarction occurs on the side with the lower minimum c value. According to the infarction judgment result output by the embodiment shown in fig. 6 by adopting a ratio method or a difference method, the c value of the target area is calculated, and the c values of the target areas 1-20 are as follows in sequence: 1.09, 0.95, 0.97, 1.11, 1.13, 1.17, 1.09, 1.15, 1.30, 1.28, 1.11, 1.10, 1.05, 1.14, 1.22, 1.11, 1.01, 1.11, from the above c value results, it can be seen that the minimum c value on the left side is 0.95, the minimum c value on the right side is 1.01, and the left side is infarcted because 0.95< 1.01. Further, the outputted infarction judgment result is corrected to 18: [1,0.95, -1.6] as 18: [0,0.95, -1.6], other infarction judgment results are not changed, and the corrected infarction judgment result is used as a new infarction judgment result for measuring the volume of the subsequent core infarction.

If the infarction judgment result output in step S407 is that one side has an infarction, further judgment is not needed, and the method can be directly used for measuring the volume of subsequent core infarction.

Step S413: and measuring the core infarction volume based on the infarction judgment results of the target areas of the left and right brains.

Based on the infarction determination results of the target areas of the left and right brains output in the step S411, the number of pixels with cerebral infarction in the target area is determined according to a preset gray threshold, the volume with cerebral infarction in each target area is calculated according to the pixel pitch of the CT image, the layer thickness of the CT image and the number of pixels with cerebral infarction in each target area, and the sum of the volumes of the core infarctions in all the target areas is used as the measurement result of the volume of the core infarctions.

In one embodiment of the present specification, the infarction is judged by using a difference method, and the infarction judgment result is that the left lenticular-shaped nucleus has infarct. Assuming that the average gray value of the left bean-shaped nucleus is h1, the average gray value of the right bean-shaped nucleus is h2, and h2-2 is taken as a gray threshold, the number of pixels of which the gray value is lower than the gray threshold in the left bean-shaped nucleus is determined asAnd N is added. Assuming that the size of the skull CT image is 512 × 512, the pixel pitch is 0.5 × 0.5mm, and the layer thickness of the skull CT image is 5mm, the volume of each pixel is 0.5 × 5 ═ 0.125mm³The volume of the core infarct of the lenticular nucleus is then N x 0.125mm³. The infarct volume of each target area in the infarct judgment result is calculated by adopting the method, and the sum of the infarct volumes of all the target areas is the measurement result of the infarct volume.

In one embodiment of the present specification, the infarct is judged by using a ratio method, and the infarct is judged as the left lenticular-shaped core infarct. Assuming that the average gray value of the left lenticular nucleus is h1, the average gray value of the right lenticular nucleus is h2, and h1/h2 is s, then (h2 (s +1)/2) is taken as the gray threshold, and the number of pixels in the left lenticular nucleus whose gray value is lower than the gray threshold is determined to be N. Assuming that the size of the skull CT image is 512 × 512, the pixel pitch is 0.5 × 0.5mm, and the layer thickness of the skull CT image is 5mm, the volume of each pixel is 0.5 × 5 ═ 0.125mm³The volume of the core infarct of the lenticular nucleus is then N x 0.125mm³. The infarct volume of each target area in the infarct judgment result is calculated by adopting the method, and the sum of the infarct volumes of all the target areas is the measurement result of the infarct volume.

In an embodiment of the present disclosure, if the infarction is determined using C ═ 0.5)/0.7+ (diff +20)/50, and the infarction determination result is that the left lenticular nucleus has infarcted, the average gray value h1 of the left lenticular nucleus is smaller than the average gray value h2 of the right lenticular nucleus, and the number of pixels whose gray values are lower than the gray threshold in the left lenticular nucleus is determined as N according to the result of settlement of (ratio-0.5)/0.7+ (diff +20)/50 for the left lenticular nucleus as C and the value h1 × C as the gray threshold. Assuming that the size of the skull CT image is 512 × 512, the pixel pitch is 0.5 × 0.5mm, and the layer thickness of the skull CT image is 5mm, the volume of each pixel is 0.5 × 5 ═ 0.125mm³The volume of the core infarct of the lenticular nucleus is then N x 0.125mm³. Calculating the infarct volume of each target area in the infarct judgment result by adopting the method, wherein the sum of the core infarct volumes in all the target areas isMeasurement of the volume of the core infarct.

In an embodiment of the present specification, after performing infarction judgment, the number of pixels that have infarcted in each target area is calculated directly according to the infarction judgment result by using a method of extracting image features. And then calculating the volume of cerebral infarction in each target area according to the pixel point distance of the CT image, the layer thickness of the CT image and the number of the pixel points of cerebral infarction in each target area, wherein the sum of the volumes of the core infarction in all the target areas is used as a measurement result of the volume of the core infarction.

In one embodiment of the present specification, the outputted measurement of the core infarct volume may be a value of the core infarct volume based on the results of two infarct determinations of the target regions of the left and right brains. Or two values of the core infarct volume, respectively representing the core infarct volume of the left and right brain. If the measurement result is represented by two core infarct volumes, in specific application, the measurement result of the core infarct volume on the corresponding side can be directly output according to the infarct side determined by known symptoms without performing two infarct judgments on the target areas of the left and right brains. If the symptom is known to be left hemiplegia, the right brain is known to have infarcted, and the measurement result of the core infarct volume of the right brain can be directly output when the core infarct volume is measured.

The aforementioned method for measuring the volume of a core infarct based on a cranial CT image for the anterior circulation can also be used for the posterior circulation. In the specific implementation, the method for measuring the volume of the post-circulatory core infarction based on the skull CT image may have some differences from the pre-circulatory, such as the target areas of the post-circulatory pons, bilateral cerebellum, midbrain, bilateral colliculus, and bilateral cerebral posterior artery blood supply areas, so the numerical labels are: the background is represented by 0, the pons is represented by 1, the left cerebellum is represented by 2, the right cerebellum is represented by 3, the midbrain is represented by 4, the left thalamus is represented by 5, the right thalamus is represented by 6, the left posterior cerebral artery blood supply region is represented by 7, and the right posterior artery blood supply region is represented by 8. When used for target extraction, the extracted features are also different. Since the target region of the posterior circulation is not present in all regions in the left and right brains, if the midbrain and the pons do not need to be subjected to bilateral infarction judgment, certain differences exist when infarction judgment is performed.

Whether the measurement of the volume of the anterior circulation core infarction or the volume of the posterior circulation core infarction is aimed at, all the ideas based on the specification or the improved method provided by the invention belong to the protection scope of the invention.

The foregoing details describe a method for measuring a core infarct volume based on a cranial CT image, and accordingly, the present application further provides a system for measuring a core infarct volume based on a cranial CT image, as shown in fig. 7, which specifically includes:

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the apparatus, the electronic device, and the nonvolatile computer storage medium, since they are substantially similar to the embodiments of the method, the description is simple, and the relevant points can be referred to the partial description of the embodiments of the method.

The apparatus, the electronic device, the nonvolatile computer storage medium and the method provided in the embodiments of the present description correspond to each other, and therefore, the apparatus, the electronic device, and the nonvolatile computer storage medium also have similar advantageous technical effects to the corresponding method.

In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the various elements may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.

As will be appreciated by one skilled in the art, the present specification embodiments may be provided as a method, system, or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.

The description has been presented with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only an example of the present specification, and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for measuring the volume of core infarction based on a skull CT image is characterized by comprising the following steps:

and determining the core infarction volume of each target area in which the cerebral infarction occurs according to the pixel point distance of the CT image, the layer thickness of the CT image and the number of the pixel points in which the cerebral infarction occurs in each target area in which the cerebral infarction occurs, wherein the sum of the core infarction volumes of all the target areas is used as a measurement result of the core infarction volume.

2. The method according to claim 1, wherein the determining a frame of a target image layer from the to-be-processed multi-frame skull CT image data specifically comprises:

3. The method of claim 1, wherein the extracting the target region based on the frame where the target image layer is located specifically comprises:

4. The method according to claim 1, wherein the determining infarction of each target area included in the target area and outputting the infarction determination result of the target area specifically includes:

5. The method according to claim 1 or 4, wherein the infarction determination of each target area included in the target areas and outputting the infarction determination result of the target areas, further comprises:

6. A system for measuring the volume of a core infarct based on a cranial CT image, comprising:

the measuring module is used for determining the number of pixel points of cerebral infarction in each target area of the infarction based on the infarction judgment result of the target area;

7. The system of claim 6, wherein the extracting the target region from the multi-frame skull CT image data to be processed specifically comprises:

8. The system according to claim 7, wherein the determining the frames of the target image layer from the multiple frames of the skull CT image data to be processed respectively comprises:

9. The system of claim 7, wherein the extracting the target region based on the frame where the target image layer is located specifically comprises:

10. The system according to claim 6, wherein the determining of the infarction of each target area included in the target area and outputting the result of the infarction determination of the target area specifically include:

11. The system according to claim 6 or 10, wherein the determining of the infarction in each target area included in the target areas and outputting the results of the infarction determination in the target areas further comprises: