Method and device for selecting a sequence of contrast images and storage medium

Abstract

The present disclosure relates to a method and device for selecting a contrast image sequence and a storage medium, the method comprising: segmenting an aorta region of a CTA contrast image sequence set to obtain an aorta binary mask image sequence set; determining a candidate aorta binary mask image sequence set and a candidate CTA contrast image sequence set according to the aorta binary mask image sequence set and an aorta length threshold; determining a three-dimensional matrix set retaining aorta contrast image information according to each candidate aorta binary mask image sequence and the corresponding candidate CTA contrast image sequence; determining a corresponding reaction tissue density average value of each three-dimensional matrix, and selecting a target CTA contrast image sequence according to the reaction tissue density average value. The purpose of the present disclosure is to automatically screen out a CTA contrast image sequence with better contrast effect from a CTA contrast image sequence set, and improve the screening efficiency of the CTA contrast image sequence.

Description

Technical Field

[0001] This disclosure relates to the field of medical image processing technology, specifically to a method, apparatus, and storage medium for selecting contrast-enhanced image sequences. Background Technology

[0002] Aortic dissection occurs when blood from within the aorta tears at the intima, entering the aortic media and causing separation of the media. This separation extends along the long axis of the aorta, creating a separation between the true and false lumens of the aortic wall. Aortic dissection is an acute, rapidly progressing, and serious cardiovascular disease with an incidence of approximately 1 in 100,000 per year, and its incidence has been increasing annually in recent years. To treat patients with aortic dissection, computed tomographic arteriography (CTA) is necessary. A single CTA examination typically generates multiple phases and different scanning sequences at various locations, each with different scanning sites, angiographic features, and other characteristics. Figure 2 As shown. Therefore, in order to accurately obtain the state of aortic dissection, it is necessary to select the CTA angiography image sequence with better aortic angiography results from the multiple generated CTA angiography image sequences.

[0003] In related technologies, radiologists select CTA angiography image sequences with better aortic angiography results from multiple CTA angiography image sequences. However, manually selecting CTA angiography image sequences with better aortic angiography results from multiple CTA angiography image sequences is time-consuming and inefficient. Summary of the Invention

[0004] The purpose of this disclosure is to provide a method, apparatus, and storage medium for selecting angiography image sequences, which automatically filters out CTA angiography image sequences with better angiography results from a set of CTA angiography image sequences, thereby solving the problem of low efficiency in manually screening CTA angiography image sequences with good aortic angiography results and improving the screening efficiency of CTA angiography image sequences.

[0005] The first aspect of this disclosure provides a method for selecting a sequence of contrast-enhanced images, the method comprising:

[0006] A set of CTA angiography image sequences is obtained, and the aortic region of the CTA angiography images in the set of CTA angiography image sequences is segmented to obtain a set of aortic binarized mask image sequences.

[0007] Based on the aortic binarized mask image sequence set and the preset aortic length threshold, a candidate aortic binarized mask image sequence set is determined;

[0008] For each candidate aortic binarized mask image sequence in the candidate aortic binarized mask image sequence set, a corresponding candidate CTA angiography image sequence is determined in the CTA angiography image sequence set. Based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, a three-dimensional matrix set retaining aortic angiography image information is determined, wherein the element values ​​of each three-dimensional matrix in the three-dimensional matrix set represent the reaction tissue density.

[0009] The average reactive tissue density corresponding to each of the three-dimensional matrices is determined, and a target CTA contrast image sequence is selected from the candidate CTA contrast image sequence set based on the average reactive tissue density.

[0010] Preferably, determining the candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and a preset aortic length threshold includes:

[0011] For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, the aortic region length in the aortic binarized mask image sequence is determined. If the aortic region length in the aortic binarized mask image sequence is greater than a preset aortic length threshold, the aortic binarized mask image sequence is added to an initially empty set to obtain a candidate aortic binarized mask image sequence set.

[0012] Preferably, determining the aortic region length in each aortic binarized mask image sequence within the aortic binarized mask image sequence set includes:

[0013] For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, perform the following operations:

[0014] Determine the interlayer spacing between each layer of the aortic binarized mask image in the aortic binarized mask image sequence, and determine the number of layers in the aortic binarized mask image sequence that include the aortic region;

[0015] The product of the interlayer spacing and the number of layers is determined as the length of the aortic region corresponding to the aortic binarized mask image sequence.

[0016] Preferably, determining the number of layers including the aortic region in the aortic binarized mask image sequence includes:

[0017] Determine the two-dimensional matrix of each layer of the aortic binarized mask image in the aortic binarized mask image sequence;

[0018] In the two-dimensional matrix, a target two-dimensional matrix is ​​determined to have matrix element values ​​of target values, wherein the target values ​​are pixel values ​​of the aortic region in the aortic binarized mask image sequence;

[0019] The number of the target two-dimensional matrices is used as the number of layers in the aortic binarized mask image sequence that include the aortic region.

[0020] Preferably, determining the three-dimensional matrix set retaining aortic angiography image information based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence includes:

[0021] Determine the three-dimensional matrix of each candidate aortic binarized mask image sequence, and determine the three-dimensional matrix of the candidate CTA angiography image sequence corresponding to each candidate aortic binarized mask image sequence;

[0022] For each candidate aortic binarized mask image sequence, the three-dimensional matrix of the candidate aortic binarized mask image sequence is multiplied by the three-dimensional matrix of the corresponding candidate CTA angiography image sequence to obtain a set of three-dimensional matrices that retain aortic angiography image information.

[0023] Preferably, determining the average reaction tissue density corresponding to each of the three-dimensional matrices includes:

[0024] For each three-dimensional matrix, the element values ​​of the three-dimensional matrix are added together to obtain the sum of the element values ​​of the three-dimensional matrix, and the number of element values ​​of the three-dimensional matrix is ​​determined.

[0025] For each three-dimensional matrix, the sum of the element values ​​of the three-dimensional matrix is ​​divided by the number of element values ​​of the three-dimensional matrix to obtain the average reactive tissue density of the three-dimensional matrix.

[0026] Preferably, the step of selecting the target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average density of the reactive tissue includes:

[0027] In the candidate CTA contrast image sequence set, the candidate CTA contrast image sequence with the largest average reaction tissue density is selected as the target CTA contrast image sequence.

[0028] A second aspect of this disclosure provides an apparatus for selecting a sequence of contrast-enhanced images, the apparatus comprising:

[0029] The segmentation module is used to acquire a set of CTA angiography image sequences and segment the aortic region of the CTA angiography images in the set of CTA angiography image sequences to obtain a set of aortic binarized mask image sequences.

[0030] The selection module is used to determine a candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and a preset aortic length threshold.

[0031] The processing module is configured to, for each candidate aortic binarized mask image sequence in the candidate aortic binarized mask image sequence set, determine the corresponding candidate CTA angiography image sequence in the CTA angiography image sequence set, and determine a three-dimensional matrix set that retains aortic angiography image information based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, wherein the element values ​​of each three-dimensional matrix in the three-dimensional matrix set represent the reaction tissue density;

[0032] The selection module is used to determine the average reactive tissue density corresponding to each of the three-dimensional matrices, and to select the target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average reactive tissue density.

[0033] A third aspect of this disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0034] A fourth aspect of this disclosure provides an electronic device, comprising:

[0035] A memory on which computer programs are stored;

[0036] A processor for executing the computer program in the memory to implement the steps of the method according to any one of the first aspects.

[0037] The above technical solution allows for the automatic selection of target CTA angiography image sequences from a set of CTA angiography image sequences, avoiding human intervention in the selection process and thus improving the selection efficiency. Specifically, segmenting the aortic region of the CTA angiography images in the set reduces the amount of subsequent image data processing, enhancing selection efficiency. Furthermore, segmentation of the aortic region reduces interference from other image regions during the selection process, ensuring accuracy. On the other hand, setting an aortic length threshold to select candidate aortic binarized mask image sequences reduces the number of subsequent image sequences requiring processing, further improving selection efficiency. Moreover, by using the average tissue density, selection errors caused by uneven distribution of contrast agent in the aortic region are avoided, further improving selection accuracy.

[0038] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0039] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings:

[0040] Figure 1 This is a flowchart of a method for selecting a sequence of contrast images according to an exemplary embodiment;

[0041] Figure 2 This is a schematic diagram of the imaging effect corresponding to different scanning positions in the same scanning sequence according to an exemplary embodiment;

[0042] Figure 3 This is a schematic diagram of a CTA angiography image sequence provided according to an exemplary embodiment;

[0043] Figure 4 This is a schematic diagram of a CTA angiography image provided according to an exemplary embodiment;

[0044] Figure 5 This is a schematic diagram of a binarized mask image of the aorta, including the aortic region, provided according to an exemplary embodiment.

[0045] Figure 6 This is a schematic diagram of the transformation from a candidate CTA angiography image sequence to a three-dimensional matrix according to an exemplary embodiment;

[0046] Figure 7 This is a block diagram of a device for selecting a sequence of contrast images according to an exemplary embodiment;

[0047] Figure 8 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0048] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0049] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0050] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0051] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0052] As mentioned in the background section, CTA (Continuous Aortic Radiography) is required in the treatment of patients with aortic dissection to obtain aortic angiographic images that reflect the state of the aortic dissection. The angiographic quality of the aortic angiography images is related to the amount of contrast agent filling the aorta. However, in a single CTA examination, the contrast agent is not uniformly distributed in the aorta. Therefore, to obtain better CTA angiographic images, multiple CTA angiographic image sequences are typically generated in a single CTA examination (each CTA angiographic image sequence may correspond to the same or different scanning sites, and the angiographic quality of each CTA angiographic image sequence is not entirely the same). This allows radiologists to subsequently select the CTA angiographic image sequence that most accurately reflects the state of the aortic dissection from multiple CTA angiographic image sequences. However, manually selecting the CTA angiographic image sequence with better aortic angiographic quality from multiple CTA angiographic image sequences is time-consuming and inefficient. Moreover, since the screening process relies on the experience of radiologists, if the radiologists lack sufficient experience, the selected CTA angiography image sequences may not accurately reflect the condition of aortic dissection, thus requiring rescreening and affecting screening efficiency.

[0053] To address the aforementioned technical problems, this disclosure provides a method, apparatus, and storage medium for selecting angiographic image sequences, which can automatically and quickly filter out target CTA angiographic image sequences with good aortic angiography results from multiple CTA angiographic image sequences, thereby improving the screening efficiency of CTA angiographic image sequences.

[0054] Figure 1 This is a flowchart illustrating a method for selecting an imaging image sequence according to an exemplary embodiment of the present disclosure, such as... Figure 1 As shown, the method may include:

[0055] S1: Obtain the CTA angiography image sequence set, and segment the aortic region of the CTA angiography images in the CTA angiography image sequence set to obtain the aortic binarized mask image sequence set.

[0056] The CTA angiography image sequence set includes multiple CTA angiography image sequences, and each CTA angiography image sequence includes multiple CTA angiography images, such as... Figure 3 As shown. Each CTA image is obtained after the patient undergoes a CTA examination. CTA examination uses CT (Computed Tomography) technology, which introduces a contrast agent to reduce the permeability of blood to X-rays, making blood vessels appear as high density on the CT image, thereby distinguishing blood vessels from other tissues.

[0057] It should be understood that the aorta is the largest artery in the human body, with a wide distribution. It originates from the left ventricle, ascends anterosuperiorly to the right, reaches the second costal cartilage on the right side, then turns posterosuperiorly to the left, descends inferiorly to the left of the lower border of the fourth thoracic vertebra, and runs along the anterior aspect of the spine, branching into many smaller arteries within the thoracic and abdominal cavities. Therefore, CTA examinations require scanning a large number of body parts or a wide area, resulting in CTA images that include regions other than the aortic region, such as... Figure 4 As shown. To avoid interference from other regions in the selection of the aortic region, and also to reduce the amount of data processing required for subsequent CTA angiography images, step S1, after obtaining the CTA angiography image sequence set, segments the aortic region of the CTA angiography images in the CTA angiography image sequence set, thereby obtaining a binarized aortic mask image that retains the aortic region and removes other regions, as shown. Figure 5 As shown, the white area corresponds to the aortic region, and the black area corresponds to the background region.

[0058] The segmentation of the aortic region in CTA angiography images within a CTA angiography image sequence set can be achieved based on an automatic aortic segmentation model.

[0059] In a possible implementation, a sample CTA angiography image can be used as input, and a sample aortic binarized mask image of the aortic region segmented based on the sample CTA angiography image can be used as output to train an automatic aortic segmentation model. The training method for this automatic aortic segmentation model can refer to relevant technologies and will not be elaborated here. Correspondingly, after obtaining a set of CTA angiography image sequences, the CTA angiography images included in the set can be input into the trained automatic aortic segmentation model to obtain the aortic binarized mask image output by the automatic aortic segmentation model for each CTA angiography image, thereby obtaining a set of aortic binarized mask image sequences. Alternatively, the aortic region segmentation can also be achieved based on the level set method, which can refer to relevant technologies and will not be elaborated here. It should be understood that the above methods for segmenting the aortic region are merely illustrative examples. In practical applications, other image segmentation methods can also be used to segment the aortic region of the CTA angiography images in the CTA angiography image sequence set, and this disclosure does not limit this approach.

[0060] S2: Determine the candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and the preset aortic length threshold.

[0061] It should be understood that during an examination of aortic patients, CTA scans are performed on different sites, such as the heart, pulmonary artery, and aorta, resulting in multiple CTA angiographic image sequences. Because the aorta is widely distributed, parts of the aorta may be scanned during CTA scans not specifically targeting the aorta. Since the aorta corresponding to this scanned area is incomplete, it is not conducive to obtaining information about aortic dissection in aortic patients. Therefore, to ensure the integrity of the aorta and reduce subsequent data processing, improving screening efficiency, an aortic length threshold can be set to filter out incomplete aortic segments. For example, during a cardiac CTA scan, although the contrast agent is injected into the coronary arteries, it can flow throughout the coronary arteries and enter parts of the aorta near the heart. This results in a small segment of the aorta being included in the cardiac CTA angiographic image sequence. Observing this small segment of the aorta cannot accurately obtain information about aortic dissection in aortic patients. Therefore, a pre-set aortic length threshold can be used to filter out this small segment of the aorta.

[0062] The aortic length threshold can be set to 400mm, 450mm, or 500mm. It should be understood that this is merely illustrative; in actual applications, the threshold can be set as needed, and this disclosure does not impose any limitations on it.

[0063] S3: For each candidate aortic binarized mask image sequence in the candidate aortic binarized mask image sequence set, determine the corresponding candidate CTA angiography image sequence in the CTA angiography image sequence set. Based on the candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, determine a three-dimensional matrix set that retains aortic angiography image information. The element values ​​of each three-dimensional matrix in the three-dimensional matrix set represent the tissue density.

[0064] In this context, candidate CTA angiography image sequences refer to CTA angiography image sequences where the aortic length exceeds the aortic length threshold. After obtaining the set of candidate aortic binarized mask image sequences, the candidate CTA angiography image sequence corresponding to each candidate aortic binarized mask image sequence can be determined from the initial CTA angiography image sequence set. For example, in practical applications, each initial CTA angiography image sequence can carry a sequence identifier. Correspondingly, after obtaining the candidate aortic binarized mask image sequence based on the initial CTA angiography image sequence, the candidate aortic binarized mask image sequence also has a corresponding sequence identifier. Therefore, if the sequence identifier of a certain initial CTA angiography image sequence is the same as the sequence identifier of a candidate aortic binarized mask image sequence, then that CTA angiography image sequence is a candidate CTA angiography image sequence corresponding to the candidate aortic binarized mask image sequence. Thus, the candidate CTA angiography image sequence corresponding to each candidate aortic binarized mask image sequence can be obtained.

[0065] For example, the initial CTA angiography image sequence set is N, where N = {N1, N2, N3, N4, N5}, and the sequence identifiers corresponding to CTA angiography image sequences N1, N2, N3, N4, and N5 are sequence identifier 1, sequence identifier 2, sequence identifier 3, sequence identifier 4, and sequence identifier 5, respectively. The candidate aortic binarized mask image sequence set is M, where M = {M1, M3, M4}, and the sequence identifiers corresponding to candidate aortic binarized mask image sequences M1, M3, and M4 are sequence identifier 1, sequence identifier 3, and sequence identifier 4, respectively. Therefore, when selecting candidate CTA angiography image sequence sets from the initial CTA angiography image sequence set based on the candidate aortic binarized mask image sequence set, the CTA angiography image sequences corresponding to sequence identifiers 1, 3, and 4 should be selected from the initial CTA angiography image sequence set. That is, the candidate CTA angiography image sequence set is constructed using CTA angiography image sequences N1, N3, and N4.

[0066] S4: Determine the average value of the reactive tissue density corresponding to the three-dimensional matrix, and select the target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average value of the reactive tissue density.

[0067] It should be understood that in CTA examinations, the contrast effect of the CTA image is related to the amount of contrast agent filling the examined area; that is, the more contrast agent filling the examined area, the better the contrast effect of the CTA image; the less contrast agent filling the examined area, the worse the contrast effect of the CTA image. The amount of contrast agent filling the examined area is reflected by the reactive tissue density; that is, the more contrast agent filling the examined area, the greater the reactive tissue density; the less contrast agent filling the examined area, the lower the reactive tissue density. Therefore, the contrast effect of CTA images can be judged by comparing reactive tissue densities, and the sequence with the most abundant contrast agent in the target examined area can be selected. It should also be considered that in CTA examinations, the contrast agent distribution in the examined area may be uneven, with some areas having excessive contrast agent and others having insufficient contrast agent. Therefore, to avoid selection errors caused by uneven distribution of contrast agent in the examination area, the above problem can be overcome by obtaining the average reaction tissue density of each three-dimensional matrix. Furthermore, the average density can also exclude sequences containing contrast agent in non-target examination areas. For example, if the target examination area is the aortic region, and the acquired cardiac CTA image sequence (non-target detection area sequence) includes an incomplete aortic region whose length exceeds the preset aortic region length, this incomplete aortic region is not what we need. Therefore, calculating the average reaction tissue density can further eliminate this incomplete aortic region length, improving the accuracy of the screening.

[0068] In summary, by employing the above method, when multiple CTA angiography image sequences are acquired, the target CTA angiography image sequence with better aortic angiography results can be automatically and accurately selected based on a preset aortic length threshold and average reactive tissue density. Specifically, by removing redundant portions from the CTA angiography image and retaining only the aortic region, the size of the CTA angiography image is reduced, thereby reducing the amount of subsequent image data processing and improving the selection efficiency of the CTA angiography image sequence. Simultaneously, segmenting the aortic region in the CTA angiography image reduces interference from other image regions or other parts of the CTA angiography sequence, thus ensuring the accuracy of the CTA angiography image sequence selection. Furthermore, setting an aortic length threshold to select sequences with a relatively complete aorta not only reduces the number of CTA angiography image sequences that need to be processed subsequently, further improving selection efficiency, but also accurately obtains the aortic dissection status of the patient. Finally, by using the average reaction tissue density, not only can screening errors caused by uneven distribution of contrast agent in the aortic region be avoided, but sequences with contrast agent in non-target examination areas can also be excluded, further improving the screening accuracy of CTA angiography image sequences.

[0069] To enable those skilled in the art to better understand the method for selecting contrast image sequences provided in this disclosure, the above steps are illustrated in detail below.

[0070] Optionally, determining a candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and a preset aortic length threshold may include:

[0071] For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, the length of the aortic region in the aortic binarized mask image sequence is determined. If the length of the aortic region in the aortic binarized mask image sequence is greater than the preset aortic length threshold, the aortic binarized mask image sequence is added to the initially empty set to obtain a candidate aortic binarized mask image sequence set.

[0072] For example, the preset aortic length threshold is 400 mm, and the acquired CTA angiography image sequence is N, where N = {N1, N2, N3}. The aortic regions in the CTA angiography image sequences N1, N2, and N3 are segmented to obtain a set of aortic binary mask image sequences P, where P = {P1, P2, P3}, and the aortic region lengths corresponding to the three aortic binary mask image sequences P1, P2, and P3 are 410 mm, 230 mm, and 480 mm, respectively. Since the aortic region lengths corresponding to the aortic binary mask image sequences P1 and P3 are greater than the preset aortic length threshold, the aortic binary mask image sequences P1 and P3 are added to the initially empty candidate set to obtain a candidate aortic binary mask image sequence set M, where M = {M1, M3}.

[0073] Optionally, determining the length of the aortic region in each aortic binarized mask image sequence within the aortic binarized mask image sequence set may include:

[0074] For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, perform the following operations: determine the interlayer spacing between each aortic binarized mask image in the aortic binarized mask image sequence, and determine the number of layers in the aortic binarized mask image sequence that include the aortic region; the product of the interlayer spacing and the number of layers is determined as the length of the aortic region corresponding to the aortic binarized mask image sequence.

[0075] The slice interval refers to the distance between two adjacent layers of the aortic binarized mask image. The slice interval can be the same or different in different aortic binarized mask image sequences. However, the slice interval is the same within the same aortic binarized mask image sequence. Furthermore, the slice interval of each aortic binarized mask image sequence can be directly read from the TAG information of the CTA angiography image sequence. Each CTA angiography image sequence corresponds to a TAG, which describes the content of the CTA angiography image sequence, including the patient information corresponding to the CTA angiography image sequence, the name of the CTA angiography image sequence, the number of layers contained in the CTA angiography image sequence, and the slice interval between adjacent layers, etc.

[0076] It should be understood that since a sequence of aortic binarized mask images includes multiple aortic binarized mask images, and each aortic binarized mask image is a two-dimensional image, it can only reflect whether the scanned area contains the aortic region and the area of ​​the contained aortic region, but not the length of the contained aortic region. The interlayer spacing, on the other hand, refers to the distance between two adjacent aortic binarized mask images, and therefore can be used to reflect the length of the aortic region. In summary, by obtaining the interlayer spacing of an aortic binarized mask image sequence and the number of layers containing the aortic region, the length of the corresponding aortic region can be calculated based on the interlayer spacing and the number of layers.

[0077] Pm L =Pm N *Pm spacing

[0078] Among them, Pm L Pm represents the length of the aortic region. N Pm indicates the number of layers including the aortic region. spacing Indicates the interlayer spacing.

[0079] Optionally, determining the number of layers in the aortic binarized mask image sequence that include the aortic region may include:

[0080] Determine the two-dimensional matrix of each layer of the aortic binarized mask image in the aortic binarized mask image sequence; identify target two-dimensional matrices in the two-dimensional matrix whose matrix element values ​​are target values, where the target values ​​are the pixel values ​​of the aortic region in the aortic binarized mask image sequence; use the number of target two-dimensional matrices as the number of layers in the aortic binarized mask image sequence that include the aortic region.

[0081] The two-dimensional matrix of the aortic binarization mask image is obtained by extracting the corresponding two-dimensional pixel matrix of the aortic binarization mask image. It should be understood that when acquiring aortic CTA angiography images, contrast agent needs to be injected into the patient's aorta, causing the aorta to appear as high-density areas in the aortic CTA image, while the rest appears as low-density areas. Therefore, when segmenting the aortic region of the CTA angiography image to obtain the aortic binarization mask image, the pixel values ​​of the aortic region in the aortic binarization mask image differ from those of the other regions (background regions). By judging the pixel values ​​of the aortic binarization mask image, it is possible to determine whether the aortic region exists within the aortic binarization mask image.

[0082] Based on the above principles, by extracting a two-dimensional pixel matrix from the binarized aortic mask image, a corresponding two-dimensional matrix is ​​generated. The presence of a target value matrix element in the generated two-dimensional matrix allows for the determination of whether an aortic region exists within the binarized aortic mask image. Specifically, if the binarized aortic mask image contains an aortic region, the generated two-dimensional matrix should contain two distinct matrix element values: one corresponding to the pixel value of the aortic region, and the other corresponding to the pixel value of the remaining region. If the binarized aortic mask image does not contain an aortic region, the generated two-dimensional matrix should contain only one matrix element value, corresponding to the pixel value of the remaining region.

[0083] For example, 1 represents the aortic region and 0 represents the remaining regions. If a binary mask image of the aorta is used to extract a two-dimensional pixel matrix, resulting in a two-dimensional matrix Pm-S1: If the two-dimensional matrix contains two distinct element values, 0 and 1, it indicates that the aortic region exists in the binarized aortic mask image. If the generated two-dimensional matrix is ​​Pm-S2: If the two-dimensional matrix has only one element value, and the corresponding pixel value of the rest of the region is 0, it means that there is no aortic region in the binarized aortic mask image.

[0084] In summary, by counting the number of two-dimensional matrices containing the target value, the number of layers containing the aortic region in the aortic binarized mask image sequence can be obtained. Subsequently, the length of the aortic region can be determined based on the number of layers and the interlayer spacing. Then, combined with the preset aortic length threshold, a candidate aortic binarized mask image sequence set can be determined, reducing the number of image sequences in subsequent processing and improving the screening efficiency.

[0085] After obtaining the candidate aortic binarized mask image sequence set, the corresponding candidate CTA angiography image sequence can be determined based on each candidate aortic binarized mask image sequence. Then, based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, a three-dimensional matrix set that retains the aortic angiography image information can be determined.

[0086] Optionally, based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, a three-dimensional matrix set retaining aortic angiography image information is determined, which may include:

[0087] The three-dimensional matrix of each candidate aortic binarized mask image sequence is determined, and the three-dimensional matrix of the candidate CTA angiography image sequence corresponding to each candidate aortic binarized mask image sequence is also determined. For each candidate aortic binarized mask image sequence, the three-dimensional matrix of the candidate aortic binarized mask image sequence and the three-dimensional matrix of the corresponding candidate CTA angiography image sequence are multiplied to obtain a set of three-dimensional matrices that retain the aortic angiography image information.

[0088] In a possible implementation, the three-dimensional matrix of each candidate CTA angiography image sequence can be generated as follows:

[0089] First, extract the two-dimensional pixel matrix 'a' of each candidate CTA angiography image from the candidate CTA angiography image sequence. Among them, a nn This represents the pixel value in the nth row and nth column. Then, according to the order of the candidate CTA angiography images in the candidate CTA angiography image sequence, the extracted two-dimensional pixel matrix 'a' is arranged in order, thus obtaining the three-dimensional matrix M0 for each candidate CTA angiography image sequence, as shown below. Figure 6 As shown, a1, a2, and a3 represent pixel values:

[0090]

[0091] The dimension of the three-dimensional matrix M0 is D×n×n, where D represents the number of layers in the candidate CTA angiography image sequence, and n×n represents the size of each two-dimensional pixel matrix.

[0092] Similarly, the above processing is performed on each candidate aortic binarized mask image sequence to obtain the three-dimensional matrix M1 of the candidate aortic binarized mask image sequence:

[0093]

[0094] In this diagram, 1 represents the aortic region in the candidate aortic binarized mask image, and 0 represents the remaining regions in the candidate aortic binarized mask image. The dimensions of the 3D matrix M1 are the same as those of the 3D matrix M0. It should be understood that this is only for illustration, used to explain the structure of the 3D matrix M1. In actual applications, the specific structure of the 3D matrix should be set according to the actual situation.

[0095] Finally, the three-dimensional matrix M0 is multiplied by the three-dimensional matrix M1 to generate a three-dimensional matrix M2 that retains the aortic angiography image information, i.e.:

[0096] M2 = M0 × M1;

[0097]

[0098] It should be understood that the element values ​​in the three-dimensional matrix represent the reactive tissue density, and the magnitude of the reactive tissue density reflects the contrast effect of the CTA angiography image. Therefore, the contrast effect of the CTA angiography image can be judged by comparing the reactive tissue density. However, because the uneven distribution of contrast agent in the aortic region can cause selection errors, after obtaining the three-dimensional matrix that retains the aortic angiography image information, the above problem can be overcome by determining the average reactive tissue density corresponding to each three-dimensional matrix.

[0099] Optionally, for each three-dimensional matrix, the element values ​​are summed to obtain the total sum of the element values ​​of the three-dimensional matrix, and the number of element values ​​of the three-dimensional matrix is ​​determined. Then, for each three-dimensional matrix, the total sum of the element values ​​of the three-dimensional matrix is ​​divided by the number of element values ​​of the three-dimensional matrix to obtain the average reactive tissue density of the three-dimensional matrix.

[0100] Since the element values ​​of each three-dimensional matrix in the three-dimensional matrix set represent the reactive tissue density, the average reactive tissue density of each three-dimensional matrix can be calculated by summing the element values ​​and counting the number of element values ​​in each matrix. That is:

[0101]

[0102] Among them, AVG HU HU represents the average reactive tissue density of the three-dimensional matrix, Q represents the element value in the three-dimensional matrix, and Q represents the number of element values ​​in the three-dimensional matrix.

[0103] It should be understood that the average reactive tissue density represents the average distribution of contrast agent in the CTA contrast image sequence. Therefore, the higher the average reactive tissue density, the greater the average distribution of contrast agent in the CTA contrast image sequence, and the better the corresponding contrast effect. Thus, after obtaining the average reactive tissue density, the target CTA contrast image sequence can be selected from the candidate CTA contrast image sequence set based on this average reactive tissue density.

[0104] Optionally, the candidate CTA contrast image sequence with the highest average tissue density can be selected from the candidate CTA contrast image sequence set as the target CTA contrast image sequence. This allows for the automatic selection of the target CTA contrast image sequence with better contrast effects.

[0105] Furthermore, it should be understood that since a mean reactive tissue density corresponds to a three-dimensional matrix of aortic angiography image information, and a three-dimensional matrix of aortic angiography image information corresponds to a candidate aortic binarized mask image sequence and a candidate CTA angiography image sequence, the target CTA angiography image sequence can be selected from the candidate CTA angiography image sequence set based on the mean reactive tissue density.

[0106] After selecting the target CTA angiography image sequence using any of the above methods, the target CTA angiography image sequence can be pushed to subsequent applications, such as inputting the target CTA angiography image sequence into a deep learning-based aortic dissection recognition model. Since the recognition effect of the aortic dissection recognition model depends on the angiography effect of the input data, and the target CTA angiography image sequence selected using any of the above methods has good angiography effect, inputting the target CTA angiography image sequence selected using any of the above methods into the aortic dissection recognition model can improve the recognition accuracy of the aortic dissection recognition model. Of course, the selected target CTA angiography image sequence can also be used for other applications, and this disclosure does not limit this application.

[0107] Based on the same concept, embodiments of this disclosure also provide a device for selecting an imaging image sequence, which can be part or all of an electronic device through software, hardware, or a combination of both. For example... Figure 7 As shown, the selection device 700 includes:

[0108] The segmentation module 701 is used to acquire a set of CTA angiography image sequences and segment the aortic region of the CTA angiography images in the set of CTA angiography image sequences to obtain a set of aortic binarized mask image sequences.

[0109] Selection module 702 is used to determine a candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and a preset aortic length threshold;

[0110] The processing module 703 is used to determine the corresponding candidate CTA angiography image sequence in the CTA angiography image sequence set for each candidate aortic binarized mask image sequence in the candidate aortic binarized mask image sequence set, and to determine a three-dimensional matrix set that retains aortic angiography image information based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, wherein the element value of each three-dimensional matrix in the three-dimensional matrix set represents the reaction tissue density;

[0111] Module 704 is selected to determine the average value of the reactive tissue density corresponding to each three-dimensional matrix, and to select the target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average value of the reactive tissue density.

[0112] Optionally, the selection module 702 is also used for:

[0113] For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, the length of the aortic region in the aortic binarized mask image sequence is determined. If the length of the aortic region in the aortic binarized mask image sequence is greater than the preset aortic length threshold, the aortic binarized mask image sequence is added to the initially empty set to obtain a candidate aortic binarized mask image sequence set.

[0114] Optionally, for each aortic binarized mask image sequence in the aortic binarized mask image sequence set, the selection module 702 is further configured to:

[0115] Determine the interlayer spacing between each layer of the aortic binarized mask image in the aortic binarized mask image sequence, and determine the number of layers in the aortic binarized mask image sequence that include the aortic region;

[0116] The product of the interlayer spacing and the number of layers is used to determine the length of the aortic region corresponding to the binarized mask image sequence of the aorta.

[0117] Optionally, the selection module 702 is also used for:

[0118] Determine the two-dimensional matrix of each layer of the aortic binarized mask image in the aortic binarized mask image sequence;

[0119] In a two-dimensional matrix, a target two-dimensional matrix is ​​determined to exist where the matrix element value is the target value, where the target value is the pixel value of the aortic region in the aortic binarized mask image sequence.

[0120] The number of target two-dimensional matrices is used as the number of layers in the aortic binarization mask image sequence that include the aortic region.

[0121] Optionally, the processing module 703 is also used for:

[0122] Determine the three-dimensional matrix of each candidate aortic binarized mask image sequence, and determine the three-dimensional matrix of the candidate CTA angiography image sequence corresponding to each candidate aortic binarized mask image sequence;

[0123] For each candidate aortic binarized mask image sequence, the three-dimensional matrix of the candidate aortic binarized mask image sequence is multiplied with the three-dimensional matrix of the corresponding candidate CTA angiography image sequence to obtain a set of three-dimensional matrices that retain the aortic angiography image information.

[0124] Optionally, the selection module 704 is also used for:

[0125] For each three-dimensional matrix, sum the element values ​​of the three-dimensional matrix to obtain the total sum of the element values ​​of the three-dimensional matrix, and determine the number of element values ​​of the three-dimensional matrix;

[0126] For each three-dimensional matrix, the sum of the element values ​​of the three-dimensional matrix is ​​divided by the number of element values ​​of the three-dimensional matrix to obtain the average value of the reactive tissue density of the three-dimensional matrix.

[0127] Optionally, the selection module 704 is also used for:

[0128] In the candidate CTA contrast image sequence set, the candidate CTA contrast image sequence with the largest average tissue density is selected as the target CTA contrast image sequence.

[0129] In summary, using the aforementioned device, when multiple CTA angiography image sequences are acquired, the target CTA angiography image sequence with better aortic angiography results can be automatically and accurately selected based on a preset aortic length threshold and average reactive tissue density. Specifically, by removing redundant portions from the CTA angiography image and retaining the aortic region, the size of the CTA angiography image is reduced, thereby reducing the amount of subsequent image data processing and improving the selection efficiency of the CTA angiography image sequence. Simultaneously, segmenting the aortic region in the CTA angiography image reduces interference from other image regions or other parts of the CTA angiography sequence, thus ensuring the accuracy of the CTA angiography image sequence selection. Furthermore, setting an aortic length threshold to select sequences with a more complete aorta not only reduces the number of CTA angiography image sequences that need to be processed subsequently, further improving selection efficiency, but also accurately obtains the aortic dissection status of the patient. Finally, by using the average reaction tissue density, not only can screening errors caused by uneven distribution of contrast agent in the aortic region be avoided, but sequences with contrast agent in non-target examination areas can also be excluded, further improving the screening accuracy of CTA angiography image sequences.

[0130] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0131] Figure 8 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Figure 8 As shown, the electronic device 800 may include a processor 801 and a memory 802. The electronic device may also include one or more of a multimedia component 803, an input / output (I / O) interface 804, and a communication component 805.

[0132] The processor 801 controls the overall operation of the electronic device 800 to complete all or part of the steps in the above-described method for selecting the imaging image sequence. The memory 802 stores various types of data to support the operation of the electronic device 800. This data may include, for example, instructions for any application or method operating on the electronic device 800, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, 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 storage, flash memory, magnetic disk, or optical disk. Multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in memory 802 or transmitted via communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. Communication component 805 is used for wired or wireless communication between the electronic device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IoT, eMTC, or other 5G technologies, or combinations thereof, is not limited here. Therefore, the corresponding communication component 805 may include: a Wi-Fi module, a Bluetooth module, an NFC module, etc.

[0133] In an exemplary embodiment, the electronic device 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, microcontrollers, microprocessors, or other electronic components to perform the above-described method for selecting the contrast image sequence.

[0134] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the above-described method for selecting a contrast image sequence. For example, the computer-readable storage medium may be the memory 802 including the program instructions, which may be executed by the processor 801 of the electronic device 800 to complete the above-described method for selecting a contrast image sequence.

[0135] In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable device, the computer program having a code portion for performing the above-described method for selecting the contrast image sequence when executed by the programmable device.

[0136] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0137] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0138] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A method for selecting a sequence of contrast-enhanced images, characterized in that, The method includes: A set of CTA angiography image sequences is obtained, and the aortic region of the CTA angiography images in the set of CTA angiography image sequences is segmented to obtain a set of aortic binarized mask image sequences. Based on the aortic binarized mask image sequence set and the preset aortic length threshold, a candidate aortic binarized mask image sequence set is determined; For each candidate aortic binarized mask image sequence in the candidate aortic binarized mask image sequence set, a corresponding candidate CTA angiography image sequence is determined in the CTA angiography image sequence set. Based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, a three-dimensional matrix set retaining aortic angiography image information is determined, wherein the element values ​​of each three-dimensional matrix in the three-dimensional matrix set represent the reaction tissue density. Determine the average reactive tissue density corresponding to each of the three-dimensional matrices, and select the target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average reactive tissue density. The step of selecting a target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average density of the reactive tissue includes: In the candidate CTA contrast image sequence set, the candidate CTA contrast image sequence with the largest average reaction tissue density is selected as the target CTA contrast image sequence.

2. The method according to claim 1, characterized in that, The step of determining the candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and a preset aortic length threshold includes: For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, the aortic region length in the aortic binarized mask image sequence is determined. If the aortic region length in the aortic binarized mask image sequence is greater than a preset aortic length threshold, the aortic binarized mask image sequence is added to an initially empty set to obtain a candidate aortic binarized mask image sequence set.

3. The method according to claim 2, characterized in that, The step of determining the length of the aortic region in each aortic binarized mask image sequence in the aortic binarized mask image sequence set includes: For each aortic binarized mask image sequence in the aortic binarized mask image sequence set, perform the following operations: Determine the interlayer spacing between each layer of the aortic binarized mask image in the aortic binarized mask image sequence, and determine the number of layers in the aortic binarized mask image sequence that include the aortic region; The product of the interlayer spacing and the number of layers is determined as the length of the aortic region corresponding to the aortic binarized mask image sequence.

4. The method according to claim 3, characterized in that, Determining the number of layers in the aortic binarized mask image sequence that include the aortic region includes: Determine the two-dimensional matrix of each layer of the aortic binarized mask image in the aortic binarized mask image sequence; In the two-dimensional matrix, a target two-dimensional matrix is ​​determined to have matrix element values ​​of target values, wherein the target values ​​are pixel values ​​of the aortic region in the aortic binarized mask image sequence; The number of the target two-dimensional matrices is used as the number of layers in the aortic binarized mask image sequence that include the aortic region.

5. The method according to any one of claims 1-4, characterized in that, The step of determining a three-dimensional matrix set that retains aortic angiography image information based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence includes: Determine the three-dimensional matrix of each candidate aortic binarized mask image sequence, and determine the three-dimensional matrix of the candidate CTA angiography image sequence corresponding to each candidate aortic binarized mask image sequence; For each candidate aortic binarized mask image sequence, the three-dimensional matrix of the candidate aortic binarized mask image sequence is multiplied by the three-dimensional matrix of the corresponding candidate CTA angiography image sequence to obtain a set of three-dimensional matrices that retain aortic angiography image information.

6. The method according to any one of claims 1-4, characterized in that, Determining the average reactive tissue density corresponding to each of the three-dimensional matrices includes: For each three-dimensional matrix, the element values ​​of the three-dimensional matrix are added together to obtain the sum of the element values ​​of the three-dimensional matrix, and the number of element values ​​of the three-dimensional matrix is ​​determined. For each three-dimensional matrix, the sum of the element values ​​of the three-dimensional matrix is ​​divided by the number of element values ​​of the three-dimensional matrix to obtain the average reactive tissue density of the three-dimensional matrix.

7. A device for selecting a sequence of contrast-enhanced images, characterized in that, The device includes: The segmentation module is used to acquire a set of CTA angiography image sequences and segment the aortic region of the CTA angiography images in the set of CTA angiography image sequences to obtain a set of aortic binarized mask image sequences. The selection module is used to determine a candidate aortic binarized mask image sequence set based on the aortic binarized mask image sequence set and a preset aortic length threshold. The processing module is configured to, for each candidate aortic binarized mask image sequence in the candidate aortic binarized mask image sequence set, determine the corresponding candidate CTA angiography image sequence in the CTA angiography image sequence set, and determine a three-dimensional matrix set that retains aortic angiography image information based on each candidate aortic binarized mask image sequence and the corresponding candidate CTA angiography image sequence, wherein the element values ​​of each three-dimensional matrix in the three-dimensional matrix set represent the reaction tissue density; The selection module is used to determine the average value of the reactive tissue density corresponding to each of the three-dimensional matrices, and to select a target CTA contrast image sequence from the candidate CTA contrast image sequence set based on the average value of the reactive tissue density. In the candidate CTA contrast image sequence set, the candidate CTA contrast image sequence with the largest average value of the reactive tissue density is selected as the target CTA contrast image sequence.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1-6.

9. An electronic device, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 1-6.

Images