CN114638843B - Method and device for identifying high-density characteristic image of middle cerebral artery - Google Patents
Method and device for identifying high-density characteristic image of middle cerebral artery Download PDFInfo
- Publication number
- CN114638843B CN114638843B CN202210273017.9A CN202210273017A CN114638843B CN 114638843 B CN114638843 B CN 114638843B CN 202210273017 A CN202210273017 A CN 202210273017A CN 114638843 B CN114638843 B CN 114638843B
- Authority
- CN
- China
- Prior art keywords
- dimensional image
- brain
- infarct
- region
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Abstract
The utility model relates to a middle cerebral artery high-density characteristic image recognition method and device, an electronic device and a storage medium, which determine at least one connected domain corresponding to a first infarct area segmented from a three-dimensional image by utilizing the mapping relation between the coordinates of the three-dimensional image and a two-dimensional image, determine a target connected domain in the connected domain, and then determine a second infarct area corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image; and finally, determining a target infarction region according to the volume of the second infarction region. According to the aspects of the disclosure, the influence on the identification of the middle cerebral artery high density sign image due to factors such as the characteristics of the CT image, the shooting level of the CT image, individual differences of patients and the small proportion of the middle cerebral artery high density sign on the CT image can be reduced, and the accuracy of the identification of the middle cerebral artery high density sign image can be improved.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for recognizing high-density images of middle cerebral artery, an electronic device, and a storage medium.
Background
The Middle Cerebral Artery high density signal (HMCAS) is represented in the Computed Tomography (CT) image of the brain as the density of the Middle Cerebral Artery is higher than that of the surrounding area. HMCAS is one of the important signs of early cerebral infarction. HMCAS is present in about 21% to 60% of cases of acute cerebral infarction.
The accuracy of identifying the HMCAS in the medical image is low due to the influences of medical image characteristics such as CT and the like and the characteristics of the patient.
Disclosure of Invention
In view of this, the present disclosure provides a middle cerebral artery high density feature image identification scheme.
According to an aspect of the present disclosure, there is provided a middle cerebral artery high density feature image identification method, including:
segmenting a first infarct area representing the high density of middle cerebral artery on a three-dimensional image, wherein the three-dimensional image corresponds to at least one layer of two-dimensional image;
determining at least one connected domain corresponding to the first peduncle area in the two-dimensional image according to the mapping relation between the three-dimensional image and the coordinates of the two-dimensional image;
determining a connected domain with the pixel average value smaller than a preset pixel value threshold value in the connected domain to obtain a target connected domain;
determining a second infarction region corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image;
determining a second infarct area in a brain area with a larger volume of the two brain areas as a target infarct area when a difference value of volumes of the respective second infarct areas of the two brain areas is greater than a preset volume difference threshold, wherein the two brain areas include: left brain region, right brain region.
In a possible implementation manner, the determining, in the connected component, a connected component whose pixel average value is smaller than a preset pixel value threshold to obtain a target connected component includes:
performing expansion treatment on the connected domain to obtain an expanded connected domain;
determining a first average value of pixel values of M pixels with minimum pixel values in each of at least two inflation connected domains under the condition that at least two inflation connected domains are intersected;
and taking the first average value with the smallest value as the pixel average value of each expansion connected domain in the at least two expansion connected domains.
In one possible implementation, the method further includes:
and in the case that the expansion connected domains do not intersect with each other, taking the average value of the pixel values of the N pixels with the minimum pixel value in the single expansion connected domain as the pixel average value of the single expansion connected domain.
In one possible implementation, the method further includes:
and under the condition that the volume difference value of the second infarct areas of the two brain areas is not larger than a preset volume difference value threshold, determining the second infarct area in the brain areas with larger average CT value of the second infarct areas in the two brain areas as a target infarct area.
In a possible implementation manner, in a case that a difference value between volumes of second infarct areas of the two brain areas is not greater than a preset volume difference threshold, determining a second infarct area in the brain area with a larger average CT value of the second infarct areas in the two brain areas as a target infarct area, includes:
acquiring an average CT value of a target connected domain in each layer of the two-dimensional image;
determining a second infarct region within the brain region having the largest said average CT value as the target infarct region.
In a possible implementation manner, before segmenting out a first infarct area representing a middle cerebral artery high density feature on the three-dimensional image, the method further includes:
determining a brain midline on the three-dimensional image, wherein the brain midline is used for dividing the brain into the left brain area and the right brain area;
and rotating the three-dimensional image to a preset position based on the middle line of the brain.
In one possible implementation, the method is implemented based on a neural network, and the segmenting a first infarct region representing a high density feature of middle cerebral arteries on the three-dimensional image comprises:
and segmenting the first infarct area in the three-dimensional image by utilizing the neural network after parameters are adjusted by a loss function, wherein the loss function is obtained by combining a weighted cross entropy and a Tverseky loss function.
According to another aspect of the present disclosure, there is provided a middle cerebral artery high density feature image recognition device, including:
the first infarct area segmentation unit is used for segmenting a first infarct area representing the high density characteristics of the middle cerebral artery on a three-dimensional image, and the three-dimensional image corresponds to at least one layer of two-dimensional image;
the connected domain determining unit is used for determining at least one connected domain corresponding to the first peduncle area in the two-dimensional image according to the mapping relation between the three-dimensional image and the coordinates of the two-dimensional image;
the target connected domain determining unit is used for determining a connected domain of which the pixel average value is smaller than a preset pixel value threshold value in the connected domain to obtain a target connected domain;
the second peduncle region determining unit is used for determining a second peduncle region corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image;
a first target infarct region determining unit, configured to determine, as a target infarct region, a second infarct region in a brain region having a larger volume of two brain regions when a difference between volumes of respective second infarct regions of the two brain regions is greater than a preset volume difference threshold, where the two brain regions include: left brain region, right brain region.
In a possible implementation manner, the target connected component determining unit includes:
the expansion connected domain determining unit is used for performing expansion processing on the connected domain to obtain an expansion connected domain;
a first average value determination unit configured to determine a first average value of pixel values of M pixels having a smallest pixel value in each of at least two inflation connected domains, in a case where at least two inflation connected domains intersect;
a first expansion connected domain average pixel determination unit, configured to use the first average value with the smallest value as the pixel average value of each of the at least two expansion connected domains.
In one possible implementation, the apparatus further includes:
a second inflation connected domain average pixel determination unit, configured to, in a case where the inflation connected domains do not intersect with each other, take an average value of pixel values of N pixels having a smallest pixel value in a single inflation connected domain as the pixel average value of the single inflation connected domain.
In one possible implementation, the apparatus further includes:
and the second target infarct region determining unit is used for determining a second infarct region in the brain region with a larger average CT value of the second infarct regions in the two brain regions as the target infarct region under the condition that the volume difference value of the second infarct regions of the two brain regions is not larger than the preset volume difference value threshold.
In a possible implementation manner, the second target infarction region determining unit is configured to:
acquiring an average CT value of a target connected domain in each layer of the two-dimensional image;
determining a second infarct region within the brain region having the largest said average CT value as the target infarct region.
In one possible implementation, the apparatus further includes:
the brain midline determining unit is used for determining a brain midline on the three-dimensional image, and the brain midline is used for dividing the brain into the left brain area and the right brain area;
and the three-dimensional image rotating unit is used for rotating the three-dimensional image to a preset position based on the brain central line.
In a possible implementation manner, the first infarct area segmentation unit is further configured to:
and segmenting the first infarct area in the three-dimensional image by using the neural network after parameters are adjusted by a loss function, wherein the loss function is obtained by combining a weighted cross entropy and a Tverseky loss function.
According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the above-described method.
According to another aspect of the present disclosure, there is provided a computer program product comprising computer readable code, or a non-transitory computer readable storage medium carrying computer readable code, which when run in a processor of an electronic device, the processor in the electronic device performs the above method.
Determining at least one connected domain corresponding to a first peduncle region segmented from a three-dimensional image by utilizing a mapping relation between coordinates of the three-dimensional image and the two-dimensional image, determining a target connected domain in the connected domain, and determining a second peduncle region corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image; and finally, determining a target infarction region according to the volume of the second infarction region. According to the aspects of the disclosure, the influence on the identification of the middle cerebral artery high density sign image due to factors such as the characteristics of the CT image, the shooting level of the CT image, individual differences of patients and the small proportion of the middle cerebral artery high density sign on the CT image can be reduced, and the accuracy of the identification of the middle cerebral artery high density sign image can be improved.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.
Fig. 1 shows a flowchart of a middle cerebral artery high density feature image identification method according to an embodiment of the present disclosure.
Fig. 2 shows a schematic diagram of adjusting brain orientation according to an embodiment of the present disclosure.
FIG. 3 is a diagram illustrating a mapping relationship between a first infarct area and a connected domain according to an embodiment of the disclosure.
Fig. 4 shows a block diagram of a middle cerebral artery high density feature image recognition device according to an embodiment of the present disclosure.
Fig. 5 shows a block diagram of an electronic device according to an embodiment of the disclosure.
Fig. 6 illustrates a block diagram of an electronic device in accordance with an embodiment of the disclosure.
Detailed Description
Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers can indicate functionally identical or similar elements. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements and circuits that are well known to those skilled in the art have not been described in detail so as not to obscure the present disclosure.
The middle cerebral artery high density signal (HMCAS) is represented in a Computed Tomography (CT) image of the brain as the density of the middle cerebral artery being higher than that of the surrounding area. HMCAS is one of the important signs of early cerebral infarction. The proportion of HMCAS present in acute cerebral infarction cases is large. Therefore, accurate identification of HMCAS in CT images will play a key role in medical research related to acute cerebral infarction.
However, accurate identification of HMCAS from CT images is challenging due to various factors. For example: CT images have the characteristics of low signal-to-noise ratio and low contrast, and the difficulty of accurately segmenting the HMCAS is increased. 2. The relative positions of all patients and the CT machine cannot be kept consistent, and the accuracy of segmenting out the target object is reduced. 3. Because the middle artery is very close to a certain part of the skull (such as the anterior bed process) or the patient has middle artery calcification, the middle artery calcification part or the bone part is easy to be identified as the middle cerebral artery high-density character by mistake. The proportion of HMCAS in CT images is too small, which causes serious unbalance problems. The solution to this in the related art is mainly applicable to CT images acquired according to uniform parameters; for the case of performing HMCAS segmentation on CT images acquired according to different parameters, the effect is poor.
The embodiment of the disclosure finally completes the identification of the middle cerebral artery high-density characteristic image by preprocessing the three-dimensional image, associating the middle cerebral artery high-density characteristic image (namely, the first infarct area) primarily segmented in the three-dimensional image with the connected domain corresponding to the first infarct area in the two-dimensional image, and screening the primarily segmented middle cerebral artery high-density characteristic image for multiple times, thereby having higher accuracy.
Fig. 1 shows a flowchart of a middle cerebral artery high density feature image identification method according to an embodiment of the present disclosure. As shown in fig. 1, the method for recognizing high-density images of middle cerebral artery includes:
step S11, segmenting a first infarct area representing the high density characteristics of the middle cerebral artery on a three-dimensional image, wherein the three-dimensional image corresponds to at least one layer of two-dimensional image;
step S12, determining at least one connected domain corresponding to the first peduncle area in the two-dimensional image according to the mapping relation between the three-dimensional image and the coordinates of the two-dimensional image;
step S13, determining a connected domain with the pixel average value smaller than a preset pixel value threshold value in the connected domain to obtain a target connected domain;
step S14, determining a second infarction region corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image;
step S15, when a difference value between volumes of second infarct areas of two brain areas is greater than a preset volume difference threshold, determining a second infarct area in a brain area having a larger volume of the two brain areas as a target infarct area, where the two brain areas include: left brain region, right brain region.
In the embodiment of the disclosure, the image representing the high density characteristics of the middle cerebral artery can be segmented on the three-dimensional image acquired for the brain of the human body, and the first infarct area is obtained as the segmentation result. And a two-dimensional image can be obtained aiming at the same human brain. The three-dimensional image and the two-dimensional image are both the representation of the same object (human brain), and thus the three-dimensional image corresponds to the two-dimensional image. The three-dimensional image and the two-dimensional image may be Computed Tomography (CT) images of the brain of the human body. The number of the two-dimensional images corresponding to a certain area in the three-dimensional image is inversely related to the layer thickness of the two-dimensional image, wherein the layer thickness is the length covered by the two-dimensional image in the direction perpendicular to the scanned object.
In one possible implementation, step S11 may be implemented based on a neural network.
The embodiment of the disclosure can train the neural network first to obtain various parameters of the neural network. The neural network here may be a U-Net neural network.
An attention mechanism can be introduced in the training process of the neural network to more reasonably distribute the weight on each channel of the characteristic diagram in the neural network, so that the neural network can more specifically identify the first infarction region, the sensitivity of the neural network to the first infarction region is improved, and the accuracy of identifying the middle cerebral artery high-density characteristic image is further improved.
The embodiment of the disclosure can use the weighted cross entropy function as the loss function of the neural network training, obtain the difference between the predicted value and the true value according to the loss function, and then adjust the parameter of the neural network by using the difference.
In one possible implementation, segmenting a first infarct region representing a high density feature of middle cerebral arteries on the three-dimensional image comprises: and segmenting the first infarct area in the three-dimensional image by utilizing the neural network after parameters are adjusted by a loss function, wherein the loss function is obtained by combining a weighted cross entropy and a Tverseky loss function.
For convenience of understanding, in the embodiment of the present disclosure, the loss function obtained based on the combination of the weighted cross entropy and the Tversky loss function is expressed by using formula (1).
L Comb ＝αL WBCE +(1-α)L Tversky (1)
Wherein L is Comb Represents a loss function, L, in one embodiment of the disclosure WBCE Represents a weighted cross entropy function, L Tversky Denotes the Tversky loss function and alpha denotes a weight, which may be an empirical value.
Because the first infarcted area is smaller in volume than other areas on the image, the unbalanced-like problem is often generated during the neural network training. Therefore, the loss function obtained by combining the weighted cross entropy and the Tverseky loss function can reduce the probability of class imbalance to a certain extent, improve the accuracy of segmenting the first infarct area, and lay a good foundation for accurately identifying the middle cerebral artery high-density characteristic image in the follow-up process.
In the embodiment of the disclosure, the three-dimensional image and the two-dimensional image may be preprocessed before the first peduncle region is segmented.
In a possible implementation manner, before step S11, the method further includes: determining a brain midline on the three-dimensional image, wherein the brain midline is used for dividing the brain into the left brain area and the right brain area on the image; and rotating the three-dimensional image to a preset position based on the middle line of the brain.
In the embodiment of the present disclosure, a line segment dividing the brain into a left brain region and a right brain region on the image may be determined on the three-dimensional image, that is, the position of the line segment on the image is determined, and the line segment is defined as a brain midline.
In the disclosed embodiments, a brain midline segmentation model may be utilized to determine a brain midline. The method for determining the cerebral midline in the embodiments of the present disclosure is not limited.
Since the relative positions of the human body and the CT apparatus are often not consistent each time the three-dimensional image is acquired, the brain directions representing the directions of the brain portions on the three-dimensional images may be different, and the brain directions may be represented by the directions of the brain central lines. Therefore, before the first infarct area is divided, the three-dimensional images can be rotated so that the brain directions represented by the three-dimensional images are consistent. In this way, the accuracy of segmenting the first infarct region can be improved.
Fig. 2 is a schematic diagram illustrating adjustment of a brain direction according to an embodiment of the present disclosure, and as shown in fig. 2, in the embodiment of the present disclosure, an included angle ω between a brain midline l and a Y-axis direction may be obtained. The Y-axis direction may be a direction perpendicular to the ground direction. The three-dimensional image is then rotated so that the angle ω is 0 °, at which time the midline l of the brain coincides with the Y-axis. By the above processing, the brain direction indicated in each three-dimensional image coincides with the Y-axis direction. Therefore, the brain directions shown in the respective three-dimensional images are the same.
Since the part of the two-dimensional image outside the brain region is mainly the background region, the proportion of the part of the brain region on the two-dimensional image is slightly larger. In order to suppress the influence of the background region on the effect of the embodiment of the present disclosure, the embodiment of the present disclosure may perform binarization processing on the two-dimensional image according to a preset pixel threshold, where the preset pixel threshold enables a maximum connected domain determined on the binarized two-dimensional image to include the brain region. For example: the pixel threshold may be set to 100-200.
After the binarization of the two-dimensional images, the minimum area including the brain, namely the image area including the brain and having the minimum area value in each layer of two-dimensional images is determined. The smallest area that includes the brain is defined herein as the brain area. Each layer of the two-dimensional image may correspond to a brain region. The coordinates of the center point of the brain region in each layer of the two-dimensional image may be the same. Then, coordinates of each corner point of each brain region are obtained, and the brain region with the largest region area value is determined, namely the target brain region is determined. And cutting each layer of two-dimensional image by using the target brain area (each corner coordinate) to obtain a cut image, namely a brain image.
Then, the brain image can be grayed according to the CT value of each pixel in the brain image, and the result after graying is still a two-dimensional image.
For example, the CT value of the pixel in the brain image with the CT value smaller than the first CT threshold may be set as the first CT threshold, and the CT value of the pixel in the brain image with the CT value larger than the second CT threshold may be set as the second CT threshold. For example: the first CT threshold may be set to 0HU and the second CT threshold to 100 HU. Then, the pixel value of each pixel is converted from the CT value of each pixel. Illustratively, the pixel value of each pixel may be obtained using equation (2).
Wherein the minimum CT value represents the minimum value of the CT values of each pixel in the brain image. Similarly, the maximum CT value represents the maximum value of the CT values of each pixel in the brain image.
In practical application, the positive samples containing the middle cerebral artery high density characteristics in the samples for training the neural network are far less than the negative samples not containing the middle cerebral artery high density characteristics. Therefore, after preprocessing the three-dimensional image and the two-dimensional image, the embodiments of the present disclosure can flip and rotate the CT image (three-dimensional image and two-dimensional image) containing the middle cerebral artery high-density feature, so that the number of the positive samples and the negative samples is close to 1: 1.
FIG. 3 is a diagram illustrating a mapping relationship between a first infarct area and a connected domain according to an embodiment of the disclosure.
As shown in fig. 3, since the three-dimensional image a corresponds to at least one two-dimensional image (two-dimensional images B1-B4) and the three-dimensional image a and the two-dimensional images B1-B4 represent the same object, the coordinates of the voxels of the three-dimensional image a and the coordinates of the pixels of the two-dimensional images B1-B4 may have a mapping relationship.
The same first dead region a may correspond to a plurality of connected domains. According to the mapping relation, areas corresponding to the first infarction area a can be determined on the two-dimensional images B1-B4, and the areas are defined as connected areas B1-B4. The connected domains b1-b4 can be distributed on a multilayer two-dimensional image. Connected domains b1-b4 may characterize cross-sections of different portions of the same first infarct region.
In the embodiment of the present disclosure, after determining the connected domain corresponding to the first pedestrial area, the first pedestrial area may be preliminarily screened.
For example, a first threshold of the number of pixels may be preset, the number of pixels in the connected component may be determined, and the segmentation result (the first infarct region) corresponding to the connected component whose number of pixels is smaller than the first threshold of the number of pixels may be excluded. Therefore, wrong segmentation results can be eliminated, for example, due to the low signal-to-noise ratio and low contrast of the CT image, the noise is identified as the first infarcted area by mistake; or erroneously identify a tissue having a cross-sectional area that is too small to be the first infarct region as the first infarct region. Therefore, the probability of misjudging the first infarct region can be reduced; the accuracy of determining the first infarct region is improved.
For example, a third CT threshold may be preset, the average value of CT values of pixels in the connected domain is determined, and the segmentation result with the average value of CT values greater than the third CT threshold is excluded. Therefore, the segmentation result of identifying the skeleton or the calcified blood vessel as the first infarction region by mistake can be eliminated, the probability of judging the first infarction region by mistake is reduced, and the accuracy of determining the first infarction region is improved.
Because the high-density characteristics of the middle cerebral artery mostly appear in the lateral cerebrum fissure area, the first infarct area can be further screened according to the local position of the first infarct area.
Generally, the pixel value of the lateral fissure region of the brain in the CT image is smaller, so the embodiment of the present disclosure can determine whether the first infarct region corresponding to the connected domain is located in the lateral fissure according to the pixel values of the region where the connected domain is located and the peripheral region thereof, and further exclude the first infarct region not located in the lateral fissure position.
In the embodiment of the present disclosure, it is determined whether the connected component is located in the lateral fissure, and the connected component located in the lateral fissure is determined as the target connected component by step S13. Therefore, first infarcted areas with identification errors can be eliminated, and accuracy of identifying the middle cerebral artery high-density characteristic image is improved.
In one possible implementation, step S13 includes: performing expansion treatment on the connected domain to obtain an expanded connected domain; determining a first average value of pixel values of M pixels with minimum pixel values in each of at least two inflation connected domains under the condition that at least two inflation connected domains are intersected; and taking the first average value with the smallest value as the pixel average value of each expansion connected domain in the at least two expansion connected domains.
In one possible implementation, in a case where the inflation connected domains do not intersect with each other, an average value of pixel values of N pixels having a smallest pixel value in a single inflation connected domain is taken as the pixel average value of the single inflation connected domain.
The dilation process may be performed by performing a convolution process on a partial region (connected component) of the image using a convolution kernel to obtain a dilated connected component. Therefore, the pixel values of the pixels in the range of the dilated connected component are merged with the pixel values of the connected component and the area surrounding the connected component. It can be determined whether the connected component is located in the lateral fissure or not by calculating the average of the pixels of the dilated connected component.
If the dilated connected component does not intersect with any connected component, embodiments of the present disclosure may sort the pixel values in the connected component from small to large, select the first N smallest pixels, determine the pixel average value of the N pixels, and use the pixel average value of the N pixels as the average pixel value of the dilated connected component, where N is a positive integer.
If at least two expansion communication domains intersect, two or more expansion communication domains have overlapping portions. The pixel values in the intersected expansion connected domains are sorted from small to large respectively, the first M pixels with the smallest pixel values in the intersected expansion connected domains are determined, the pixel average values of the M pixels are determined respectively, namely the first average values of the intersected expansion connected domains are obtained, and the first average value with the smallest value is used as the average pixel value of the intersected expansion connected domains. Therefore, the calculation amount of calculating the average pixel value can be reduced, the determined value can represent the average pixel value of the connected domain, and the step of judging whether the first peduncle area is located in the lateral cerebrum fissure is simplified.
In step 14, the embodiment of the present disclosure determines a second infarct region corresponding to the target connected domain in the three-dimensional image by using the mapping relationship between the three-dimensional image and the two-dimensional image. The embodiment of the disclosure realizes the purpose of screening the first peduncle area. The second peduncle area is a segmentation result obtained by screening the first peduncle area.
Since the middle cerebral artery high density features generally appear in only one brain region (left brain region or right brain region), the second infarct region can be further screened according to whether the second infarct region appears in only a unilateral brain region.
Generally, if a doctor can judge which side brain area of a patient the middle cerebral artery high density features appear in through a patient case or by other auxiliary means, and can obtain clinical affected side information, the second infarct area of the other side brain area on the image can be directly excluded.
If the clinical affected side information cannot be obtained, the first infarct area can be screened according to the volume of the second infarct area in each brain area.
In a possible implementation manner, in the embodiment of the present disclosure, under the condition that a difference value between volumes of respective second infarct areas of two brain areas is not greater than a preset volume difference threshold, a second infarct area in the brain area with a larger average CT value of the second infarct areas in the two brain areas is determined as the target infarct area.
In the embodiment of the present disclosure, the volumes of the second infarct areas in the left brain area and the right brain area may be determined, and then the difference between the volume of the second infarct area in the left brain area and the volume of the second infarct area in the right brain area may be determined.
And if the difference is larger than a preset volume difference threshold value, determining a second infarct area in a lateral cerebral area of a second infarct area with a larger volume as a target infarct area.
If the difference value is not larger than the threshold value of the preset volume difference, the average CT value of the second infarct areas in the left cerebral area and the right cerebral area is continuously determined, and then the second infarct area in the cerebral area on one side with the larger average CT value of the second infarct areas in the two cerebral areas is determined as the target infarct area. Therefore, the second infarct area in the lateral brain area can be excluded, and the accuracy of identifying the high-density characteristic image can be improved.
In a possible implementation manner, in a case that a difference value between volumes of second infarct areas of the two brain areas is not greater than a preset volume difference threshold, determining a second infarct area in the brain area with a larger average CT value of the second infarct areas in the two brain areas as a target infarct area, includes: acquiring an average CT value of a target connected domain in each layer of the two-dimensional image; determining a second infarct region within the brain region having the largest said average CT value as the target infarct region.
As previously described, the first infarct area may correspond to a plurality of connected domains, so the second infarct area may also correspond to a plurality of target connected domains. Embodiments of the present disclosure may determine an average CT value for a target connected component corresponding to a second infarct zone in each brain region. Then, comparing the average CT values, and determining a second infarct area corresponding to the target connected domain with the maximum average CT value as a target infarct area; and determining the brain area where the target connected component with the maximum average CT value is positioned as the brain area where the high density of middle cerebral artery appears.
The embodiment of the disclosure can also screen the target infarction region after the target infarction region is determined. According to the mapping relation, the target infarction region can correspond to a plurality of connected domains on the two-dimensional image, and the connected domain corresponding to the target infarction region is defined as a first connected domain.
In a possible implementation manner, the embodiment of the disclosure may perform modeling according to the characteristics of the image omics, and screen the target infarction region. The cinematology features may include: number of pixels, CT value, etc.
For example, the second pixel number threshold and the fourth CT threshold may be preset, and the pixel number of the single first connected domain and the CT average value of the single first connected domain may be determined. And judging whether a plurality of first communication domains corresponding to the target infarction region have first communication domains of which the pixel number of the first communication domains is greater than a second pixel number threshold value or the CT average value of the first communication domains is greater than a fourth CT threshold value. If the connected domain exists, the target infarct area corresponding to the first connected domain is reserved, and if the connected domain does not exist, the target infarct area corresponding to the first connected domain is excluded.
By screening the target infarction region, the sensitivity of identifying the middle cerebral artery high-density sign image can be ensured, the false positive target infarction region can be eliminated, and the accuracy of identifying the middle cerebral artery high-density sign image is improved.
In the embodiment of the disclosure, at least one connected domain corresponding to a first peduncle region segmented from a three-dimensional image is determined by using a mapping relation between coordinates of the three-dimensional image and the two-dimensional image, a target connected domain is determined in the connected domain, and then a second peduncle region corresponding to the target connected domain in the three-dimensional image is determined according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image; and finally, determining a target infarction region according to the volume of the second infarction region. Therefore, the segmented first infarct area can be screened for many times, the influence on the identification of the middle cerebral artery high-density sign image due to the factors of the characteristics of the CT image, the CT image shooting level, the individual difference of a patient, the small proportion of the middle cerebral artery high-density sign on the CT image and the like is reduced, and the accuracy of the identification of the middle cerebral artery high-density sign image is improved.
Fig. 4 is a block diagram of a middle cerebral artery high-density feature image recognition apparatus according to an embodiment of the present disclosure, and as shown in fig. 4, the apparatus 20 includes:
the first infarct area segmentation unit 21 is configured to segment a first infarct area representing a high density characteristic of a middle cerebral artery on a three-dimensional image, where the three-dimensional image corresponds to at least one layer of two-dimensional image;
a connected component determining unit 22, configured to determine, according to a mapping relationship between the three-dimensional image and the coordinates of the two-dimensional image, at least one connected component in the two-dimensional image that corresponds to the first peduncle area;
a target connected component determining unit 23, configured to determine, in the connected component, a connected component whose pixel average value is smaller than a preset pixel value threshold, so as to obtain a target connected component;
a second stalk zone determining unit 24, configured to determine, according to a mapping relationship between coordinates of the three-dimensional image and the two-dimensional image, a second stalk zone in the three-dimensional image, which corresponds to the target connected domain;
a first target infarct region determining unit 25, configured to determine, as a target infarct region, a second infarct region in a brain region with a larger volume of two brain regions when a difference between volumes of respective second infarct regions of the two brain regions is greater than a preset volume difference threshold, where the two brain regions include: left brain region, right brain region.
In a possible implementation manner, the target connected component determining unit 23 includes:
the expansion connected domain determining unit is used for performing expansion processing on the connected domain to obtain an expansion connected domain;
a first average value determination unit that determines a first average value of pixel values of M pixels having a smallest pixel value among the respective inflation connected domains of the at least two inflation connected domains, in a case where the at least two inflation connected domains intersect;
a first expansion connected domain average pixel determination unit, configured to use the first average value with the smallest value as the pixel average value of each of the at least two expansion connected domains.
In one possible implementation, the apparatus 20 further includes:
a second inflation connected domain average pixel determination unit, configured to, in a case where the inflation connected domains do not intersect with each other, take an average value of pixel values of N pixels having a smallest pixel value in a single inflation connected domain as the pixel average value of the single inflation connected domain.
In a possible implementation manner, the apparatus 20 further includes:
and the second target infarct region determining unit is used for determining a second infarct region in the brain region with a larger average CT value of the second infarct regions in the two brain regions as the target infarct region under the condition that the volume difference value of the second infarct regions of the two brain regions is not larger than the preset volume difference value threshold.
In a possible implementation manner, the second target infarction region determining unit is configured to:
acquiring an average CT value of a target connected domain in each layer of the two-dimensional image;
determining a second infarct region within the brain region having the largest said average CT value as the target infarct region.
In one possible implementation, the apparatus 20 further includes:
the brain midline determining unit is used for determining a brain midline on the three-dimensional image, and the brain midline is used for dividing the brain into the left brain area and the right brain area;
and the three-dimensional image rotating unit is used for rotating the three-dimensional image to a preset position based on the brain central line.
In a possible implementation manner, the first infarct area segmentation unit 21 is further configured to:
and segmenting the first infarct area in the three-dimensional image by utilizing the neural network after parameters are adjusted by a loss function, wherein the loss function is obtained by combining a weighted cross entropy and a Tverseky loss function.
In some embodiments, functions of or modules included in the apparatus provided in the embodiments of the present disclosure may be used to execute the method described in the above method embodiments, and specific implementation thereof may refer to the description of the above method embodiments, and for brevity, will not be described again here.
Embodiments of the present disclosure also provide a computer-readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the above-mentioned method. The computer readable storage medium may be a volatile or non-volatile computer readable storage medium.
An embodiment of the present disclosure further provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-described method when executing the memory-stored instructions.
The disclosed embodiments also provide a computer program product comprising computer readable code or a non-transitory computer readable storage medium carrying computer readable code, which when run in a processor of an electronic device, the processor in the electronic device performs the above method.
Fig. 5 is a block diagram illustrating an apparatus 800 for mid-cerebral artery high density feature image recognition, according to an exemplary embodiment. For example, the apparatus 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.
Referring to fig. 5, the apparatus 800 may include one or more of the following components: processing component 802, memory 804, power component 806, multimedia component 808, audio component 810, input/output (I/O) interface 812, sensor component 814, and communications component 816.
The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing components 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.
The memory 804 is configured to store various types of data to support operations at the apparatus 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The multimedia component 808 includes a screen that provides an output interface between the device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.
The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.
The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed status of the device 800, the relative positioning of components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in the position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration/deceleration of the device 800, and a change in the temperature of the device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object in the absence of any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 816 is configured to facilitate communications between the apparatus 800 and other devices in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic components for performing the above-described methods.
In an exemplary embodiment, a non-transitory computer-readable storage medium, such as the memory 804, is also provided that includes computer program instructions executable by the processor 820 of the device 800 to perform the above-described methods.
Fig. 6 is a block diagram illustrating an apparatus 1900 for mid-cerebral artery high density feature image recognition, according to an exemplary embodiment. For example, the apparatus 1900 may be provided as a server or terminal device. Referring to FIG. 6, the device 1900 includes a processing component 1922 further including one or more processors and memory resources, represented by memory 1932, for storing instructions, e.g., applications, executable by the processing component 1922. The application programs stored in memory 1932 may include one or more modules that each correspond to a set of instructions. Further, the processing component 1922 is configured to execute instructions to perform the methods described above.
The device 1900 may also include a power component 1926 configured to perform power management of the device 1900, a wired or wireless network interface 1950 configured to connect the device 1900 to a network, and an input/output (I/O) interface 1958. The device 1900 may operate based on an operating system stored in memory 1932, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.
In an exemplary embodiment, a non-transitory computer readable storage medium, such as the memory 1932, is also provided that includes computer program instructions executable by the processing component 1922 of the apparatus 1900 to perform the above-described methods.
The present disclosure may be systems, methods, and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied thereon for causing a processor to implement various aspects of the present disclosure.
The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present disclosure may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, the electronic circuitry that can execute the computer-readable program instructions implements aspects of the present disclosure by utilizing the state information of the computer-readable program instructions to personalize the electronic circuitry, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA).
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, 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/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or technical improvements to the market, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (10)
1. A middle cerebral artery high-density feature image identification method is characterized by comprising the following steps:
segmenting a first infarct area representing the high density of middle cerebral artery on a three-dimensional image, wherein the three-dimensional image corresponds to at least one layer of two-dimensional image;
determining at least one connected domain corresponding to the first peduncle area in the two-dimensional image according to the mapping relation between the three-dimensional image and the coordinates of the two-dimensional image;
determining a connected domain with the pixel average value smaller than a preset pixel value threshold value in the connected domain to obtain a target connected domain;
determining a second infarction region corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image;
determining a second infarct area in a brain area with a larger volume of the two brain areas as a target infarct area when a difference value of volumes of the respective second infarct areas of the two brain areas is greater than a preset volume difference threshold, wherein the two brain areas include: left brain region, right brain region.
2. The method according to claim 1, wherein the determining, in the connected component, a connected component having a pixel average value smaller than a preset pixel value threshold to obtain a target connected component comprises:
performing expansion treatment on the connected domain to obtain an expanded connected domain;
determining a first average value of pixel values of M pixels with minimum pixel values in each of at least two inflation connected domains under the condition that at least two inflation connected domains are intersected;
and taking the first average value with the minimum value as the pixel average value of each expansion connected domain in the at least two expansion connected domains.
3. The method of claim 2, further comprising:
and in the case that the expansion connected domains do not intersect with each other, taking the average value of the pixel values of the N pixels with the minimum pixel value in the single expansion connected domain as the pixel average value of the single expansion connected domain.
4. The method of claim 1, further comprising:
and under the condition that the volume difference value of the second infarct areas of the two brain areas is not larger than a preset volume difference value threshold, determining the second infarct area in the brain areas with larger average CT value of the second infarct areas in the two brain areas as a target infarct area.
5. The method according to claim 4, wherein in a case that the difference value of the volumes of the second infarct areas of the two brain areas is not greater than the preset volume difference value threshold, determining the second infarct area in the brain area with the larger average CT value of the second infarct areas in the two brain areas as the target infarct area comprises:
acquiring an average CT value of a target connected domain in each layer of the two-dimensional image;
determining a second infarct region within the brain region having the largest said average CT value as the target infarct region.
6. The method according to claim 1, wherein before segmenting the first infarct region characterizing the high density of the middle cerebral artery on the three-dimensional image, further comprising:
determining a brain midline on the three-dimensional image, wherein the brain midline is used for dividing the brain into the left brain area and the right brain area;
and rotating the three-dimensional image to a preset position based on the middle line of the brain.
7. The method of claim 1, wherein the method is implemented based on a neural network, and the segmenting of the first infarct region representing the high density feature of the middle cerebral artery on the three-dimensional image comprises:
and segmenting the first infarct area in the three-dimensional image by utilizing the neural network after parameters are adjusted by a loss function, wherein the loss function is obtained by combining a weighted cross entropy and a Tverseky loss function.
8. An image recognition device is characterized in that the device comprises:
the first infarct area segmentation unit is used for segmenting a first infarct area representing the high density characteristics of the middle cerebral artery on a three-dimensional image, and the three-dimensional image corresponds to at least one layer of two-dimensional image;
the connected domain determining unit is used for determining at least one connected domain corresponding to the first peduncle area in the two-dimensional image according to the mapping relation between the three-dimensional image and the coordinates of the two-dimensional image;
the target connected component determining unit is used for determining a connected component of which the pixel average value is smaller than a preset pixel value threshold value in the connected component to obtain a target connected component;
the second peduncle region determining unit is used for determining a second peduncle region corresponding to the target connected domain in the three-dimensional image according to the mapping relation between the coordinates of the three-dimensional image and the two-dimensional image;
a target infarct region determining unit, configured to determine, as a target infarct region, a second infarct region in a brain region having a larger volume of the two brain regions when a difference between volumes of respective second infarct regions of the two brain regions is greater than a preset volume difference threshold, where the two brain regions include: left brain region, right brain region.
9. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the method of any one of claims 1 to 7 when executing the memory-stored instructions.
10. A non-transitory computer readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210273017.9A CN114638843B (en) | 2022-03-18 | 2022-03-18 | Method and device for identifying high-density characteristic image of middle cerebral artery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210273017.9A CN114638843B (en) | 2022-03-18 | 2022-03-18 | Method and device for identifying high-density characteristic image of middle cerebral artery |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114638843A CN114638843A (en) | 2022-06-17 |
| CN114638843B true CN114638843B (en) | 2022-09-06 |
Family
ID=81950626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210273017.9A Active CN114638843B (en) | 2022-03-18 | 2022-03-18 | Method and device for identifying high-density characteristic image of middle cerebral artery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114638843B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109671067A (en) * | 2018-12-14 | 2019-04-23 | 强联智创（北京）科技有限公司 | A kind of measurement method and system of the core Infarction volume based on head CT images |
| CN109671066A (en) * | 2018-12-14 | 2019-04-23 | 强联智创（北京）科技有限公司 | A kind of method and system of the cerebral infarction judgement based on head CT images |
| CN110569854A (en) * | 2019-09-12 | 2019-12-13 | 上海商汤智能科技有限公司 | Image processing method and device, electronic equipment and storage medium |
| JP2019213785A (en) * | 2018-06-14 | 2019-12-19 | 社会福祉法人 恩賜財団済生会熊本病院 | Medical image processor, method and program |
| CN110992377A (en) * | 2019-12-02 | 2020-04-10 | 北京推想科技有限公司 | Image segmentation method, device, computer-readable storage medium and equipment |
| CN112070781A (en) * | 2020-08-13 | 2020-12-11 | 沈阳东软智能医疗科技研究院有限公司 | Processing method and device of craniocerebral tomography image, storage medium and electronic equipment |
| CN112529918A (en) * | 2020-12-23 | 2021-03-19 | 沈阳东软智能医疗科技研究院有限公司 | Method, device and equipment for ventricular region segmentation in brain CT image |
| CN112927239A (en) * | 2021-02-22 | 2021-06-08 | 北京安德医智科技有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN113077479A (en) * | 2021-03-25 | 2021-07-06 | 深圳市铱硙医疗科技有限公司 | Automatic segmentation method, system, terminal and medium for acute ischemic stroke focus |
| CN113177943A (en) * | 2021-06-29 | 2021-07-27 | 中南大学 | Cerebral apoplexy CT image segmentation method |
| CN113876345A (en) * | 2021-09-28 | 2022-01-04 | 武汉中科医疗科技工业技术研究院有限公司 | Method and device for identifying ischemic penumbra, electronic device and storage medium |
| CN114121236A (en) * | 2021-11-22 | 2022-03-01 | 上海商汤智能科技有限公司 | Image detection method and related device, electronic equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7129869B2 (en) * | 2018-10-01 | 2022-09-02 | 富士フイルム株式会社 | Disease area extraction device, method and program |
-
2022
- 2022-03-18 CN CN202210273017.9A patent/CN114638843B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2019213785A (en) * | 2018-06-14 | 2019-12-19 | 社会福祉法人 恩賜財団済生会熊本病院 | Medical image processor, method and program |
| CN109671067A (en) * | 2018-12-14 | 2019-04-23 | 强联智创（北京）科技有限公司 | A kind of measurement method and system of the core Infarction volume based on head CT images |
| CN109671066A (en) * | 2018-12-14 | 2019-04-23 | 强联智创（北京）科技有限公司 | A kind of method and system of the cerebral infarction judgement based on head CT images |
| CN110569854A (en) * | 2019-09-12 | 2019-12-13 | 上海商汤智能科技有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN110992377A (en) * | 2019-12-02 | 2020-04-10 | 北京推想科技有限公司 | Image segmentation method, device, computer-readable storage medium and equipment |
| CN112070781A (en) * | 2020-08-13 | 2020-12-11 | 沈阳东软智能医疗科技研究院有限公司 | Processing method and device of craniocerebral tomography image, storage medium and electronic equipment |
| CN112529918A (en) * | 2020-12-23 | 2021-03-19 | 沈阳东软智能医疗科技研究院有限公司 | Method, device and equipment for ventricular region segmentation in brain CT image |
| CN112927239A (en) * | 2021-02-22 | 2021-06-08 | 北京安德医智科技有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN113077479A (en) * | 2021-03-25 | 2021-07-06 | 深圳市铱硙医疗科技有限公司 | Automatic segmentation method, system, terminal and medium for acute ischemic stroke focus |
| CN113177943A (en) * | 2021-06-29 | 2021-07-27 | 中南大学 | Cerebral apoplexy CT image segmentation method |
| CN113876345A (en) * | 2021-09-28 | 2022-01-04 | 武汉中科医疗科技工业技术研究院有限公司 | Method and device for identifying ischemic penumbra, electronic device and storage medium |
| CN114121236A (en) * | 2021-11-22 | 2022-03-01 | 上海商汤智能科技有限公司 | Image detection method and related device, electronic equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| Automatic segmentation of cerebral infarcts in follow-up computed tomography images with convolutional neural networks;renan sales Barros 等;《Ischemic Stroke》;20191223;1-6 * |
| CT 对超急性期脑梗死的诊断价值;张颖;《医疗装备》;20200630;第33卷(第12期);12-14 * |
| 肺部CT影像慢阻肺病灶三维定量化分析研究;李梭;《中国优秀博硕士学位论文全文数据库(博士)信息科技辑》;20150115(第（2015）01期);I138-34 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114638843A (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829920B (en) | Image processing method and device, electronic equipment and storage medium | |
| US9665945B2 (en) | Techniques for image segmentation | |
| CN109978886B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN110569854B (en) | Image processing method and device, electronic equipment and storage medium | |
| WO2020211293A1 (en) | Image segmentation method and apparatus, electronic device and storage medium | |
| WO2022151755A1 (en) | Target detection method and apparatus, and electronic device, storage medium, computer program product and computer program | |
| CN111899268B (en) | Image segmentation method and device, electronic equipment and storage medium | |
| CN112767329A (en) | Image processing method and device and electronic equipment | |
| CN113222038B (en) | Breast lesion classification and positioning method and device based on nuclear magnetic image | |
| CN112967291B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN113034491B (en) | Coronary calcified plaque detection method and device | |
| CN113012166A (en) | Intracranial aneurysm segmentation method and device, electronic device, and storage medium | |
| CN112862916B (en) | CT perfusion function map quantitative parameter processing equipment and method | |
| CN114820584B (en) | Lung focus positioner | |
| WO2023050691A1 (en) | Image processing method and apparatus, and electronic device, storage medium and program | |
| CN112075927A (en) | Method and device for classifying causes of cerebral apoplexy | |
| CN112927239A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN113160947A (en) | Medical image display method and device, electronic equipment and storage medium | |
| JP2022518583A (en) | Neural network training and image segmentation methods, devices, equipment | |
| CN113012816A (en) | Brain partition risk prediction method and device, electronic equipment and storage medium | |
| CN114638843B (en) | Method and device for identifying high-density characteristic image of middle cerebral artery | |
| JP2022548453A (en) | Image segmentation method and apparatus, electronic device and storage medium | |
| CN106469446B (en) | Depth image segmentation method and segmentation device | |
| CN116843647A (en) | Method and device for determining lung field area and evaluating lung development, electronic equipment and medium | |
| CN114387436B (en) | Wall coronary artery detection method and device, electronic device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Image recognition method and device of middle cerebral artery high-density sign Effective date of registration: 20221115 Granted publication date: 20220906 Pledgee: Beijing Jiahesen Health Technology Co.,Ltd. Pledgor: BEIJING ANDE YIZHI TECHNOLOGY Co.,Ltd. Registration number: Y2022990000806 |