Target identification method and device, electronic equipment and storage medium

Abstract

A target identification method and device, electronic equipment, and storage medium are disclosed. The method includes obtaining a segmentation result fusion image of a tubular part and a cross-sectional image corresponding to each cross-sectional position in the tubular part; determining a cross-sectional polar coordinate image corresponding to each cross-sectional position based on each cross-sectional image; obtaining a pixel value of a pixel point in each cross-sectional polar coordinate image and determining a target identification result of the tubular part based on the pixel value and a preset pixel threshold. The disclosed technical solution solves the problem of high misjudgment probability of the target identification result in the prior art, reduces the misjudgment probability of the identification result, and improves the identification accuracy.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and more particularly to target recognition methods, devices, electronic devices, and storage media. Background Technology

[0002] Coronary artery disease (CAD) has become one of the top three causes of death worldwide. Calcified plaques in the coronary arteries are a significant component of atherosclerotic plaques, causing narrowing of the coronary arteries, myocardial hypoxia, and decreased cardiac systolic and diastolic function, greatly increasing the likelihood of developing CAD. Therefore, the detection and identification of coronary artery calcified plaques plays a crucial role in the prevention and treatment of CAD.

[0003] Current automated algorithms, such as thresholding, set a threshold value. If the grayscale value of a region in CCTA exceeds this threshold, the region is considered a calcified area. The drawback of this approach is its limited applicability. CCTA images can have different grayscale distributions due to different CT scanner manufacturers or different imaging parameters. Therefore, using a fixed threshold to distinguish calcified plaques can lead to many misjudgments, resulting in reduced recognition accuracy. Summary of the Invention

[0004] This invention provides a target recognition method, device, electronic device, and storage medium to solve the problem of high misjudgment probability of target recognition results in tubular parts in the prior art, thereby reducing the probability of misjudgment of recognition results and improving recognition accuracy.

[0005] In a first aspect, embodiments of the present invention provide a target recognition method, the method comprising:

[0006] Obtain the segmented fused image of the tubular portion, as well as the cross-sectional images corresponding to each cross-sectional position in the tubular portion;

[0007] Based on each of the cross-sectional images, determine the polar coordinate image of the cross section corresponding to each of the cross-sectional positions;

[0008] Obtain the pixel values ​​of each pixel in the polar coordinate image of each cross section, and determine the target recognition result of the tubular part based on the pixel values ​​and a preset pixel threshold.

[0009] Optionally, obtaining the segmented result fused image of the tubular portion includes:

[0010] An initial medical image of the tubular portion is acquired, and the initial medical image is segmented to obtain a segmentation result image of the tubular portion;

[0011] The segmented image and the initial medical image are fused to obtain the fused segmented image of the tubular region.

[0012] Optionally, determining the polar coordinate image of the cross-section corresponding to each of the cross-sectional positions based on each of the cross-sectional images includes:

[0013] For any cross-sectional image, determine multiple preset diameters of the current cross-sectional image, and determine the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters respectively;

[0014] Image optimization processing is performed on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each cross-sectional position.

[0015] Optionally, determining the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters includes:

[0016] For any preset diameter in the current cross-sectional image, the pixel value of the diameter pixel point corresponding to the current preset diameter in the cross-sectional image is determined based on the diameter rotation angle corresponding to the current preset diameter;

[0017] Determine the stitching order corresponding to each diameter rotation angle, and stitch together the diameter pixels corresponding to the multiple preset diameters according to the stitching order to obtain the initial cross-sectional polar coordinate image of the current cross-sectional image.

[0018] Optionally, the optimization process includes edge optimization processing; the step of performing image optimization processing on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions includes:

[0019] The equivalent diameter corresponding to each of the cross-sectional images is determined respectively, and the cross-sectional image to be processed in each of the cross-sectional images is determined based on the preset equivalent diameter threshold and each of the equivalent diameters;

[0020] The cross-sectional polar coordinate image corresponding to the cross-sectional image to be processed is subjected to cross-sectional edge optimization processing to obtain the cross-sectional polar coordinate image of the tubular part.

[0021] Optionally, the optimization process includes dilation optimization.

[0022] The step of performing image optimization processing on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions includes:

[0023] Obtain preset dilation parameters, and perform dilation optimization processing on the cross-sectional polar coordinate images corresponding to each of the cross-sectional images based on the dilation parameters to obtain the cross-sectional polar coordinate images of the tubular part.

[0024] Optionally, obtaining the pixel values ​​of pixels in the polar coordinate images of each cross section, and determining the target recognition result of the tubular portion based on the pixel values ​​and a preset pixel threshold, includes:

[0025] For any cross-sectional position, the average pixel value and maximum pixel value of each pixel in the current cross-sectional position are determined based on the cross-sectional polar coordinate image corresponding to the current cross-sectional position;

[0026] The pixel ratio at the current cross-section position is determined based on the average pixel value and the maximum pixel value;

[0027] Obtain a preset pixel ratio threshold, and determine the target recognition result at the current cross-section position based on the pixel ratio threshold and the pixel ratio value;

[0028] The target identification result of the tubular part is determined based on the target identification results at each of the cross-sectional positions.

[0029] Optionally, the tubular portion includes a coronary artery, and the target identification result includes whether the coronary artery contains plaque.

[0030] Secondly, embodiments of the present invention also provide a target recognition device, the device comprising:

[0031] The image acquisition module is used to acquire the segmented result fused image of the tubular part, as well as the cross-sectional images corresponding to each cross-section position in the tubular part;

[0032] A cross-sectional polar coordinate image determination module is used to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions based on each of the cross-sectional images;

[0033] The target recognition result determination module is used to obtain the pixel value of each pixel in the polar coordinate image of each cross section, and determine the target recognition result of the tubular part based on the pixel value and a preset pixel threshold.

[0034] Thirdly, embodiments of the present invention also provide an electronic device, comprising:

[0035] At least one processor; and

[0036] A memory communicatively connected to the at least one processor; wherein,

[0037] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the target recognition method according to any embodiment of the present invention.

[0038] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions that are used to cause a processor to execute and implement the target recognition method described in any embodiment of the present invention.

[0039] The technical solution of this invention involves acquiring a segmented and fused image of the tubular portion, and cross-sectional images corresponding to each transverse position within the tubular portion; determining polar coordinate images of each transverse position based on each cross-sectional image; acquiring pixel values ​​of pixels in each polar coordinate image; and determining the target recognition result of the tubular portion based on the pixel values ​​and a preset pixel threshold. This technical solution does not set a fixed grayscale threshold, but instead sets a threshold representing the ratio of the maximum pixel value to the average pixel value. This makes the numerical comparison unaffected by the grayscale distribution range, making it more applicable and flexible. Furthermore, the method of merging and filtering recognition results effectively removes noise and false positives, resulting in higher accuracy and robustness of the recognition results.

[0040] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of a target recognition method provided in Embodiment 1 of the present invention;

[0043] Figure 2 This is a schematic diagram of a blood vessel segmentation result fusion image provided according to Embodiment 1 of the present invention;

[0044] Figure 3 This is a schematic diagram of a cross-sectional image of a blood vessel at a transverse position according to Embodiment 1 of the present invention;

[0045] Figure 4 This is a schematic diagram of a cross-sectional polar coordinate image of a blood vessel's transverse position according to Embodiment 1 of the present invention;

[0046] Figure 5 This is a schematic diagram of pixel values ​​and target recognition results at various cross-sectional positions of a blood vessel according to Embodiment 1 of the present invention;

[0047] Figure 6 This is a flowchart of a target recognition method provided according to Embodiment 2 of the present invention;

[0048] Figure 7 This is a schematic diagram illustrating the method for determining the pixel value of a diameter pixel according to Embodiment 1 of the present invention;

[0049] Figure 8 This is a schematic diagram of a branched blood vessel present in a blood vessel according to Embodiment 1 of the present invention;

[0050] Figure 9 This is a schematic diagram of the equivalent diameter of a blood vessel at various cross-sectional positions according to Embodiment 1 of the present invention;

[0051] Figure 10 This is a schematic diagram of the process of wiping away branch vessels in a blood vessel according to Embodiment 1 of the present invention;

[0052] Figure 11 This is a schematic diagram of the dilation process of an initial cross-sectional polar coordinate image according to Embodiment 1 of the present invention;

[0053] Figure 12 This is a schematic diagram of the structure of a target recognition device according to Embodiment 3 of the present invention;

[0054] Figure 13 This is a schematic diagram of the structure of an electronic device that implements the target recognition method of this invention. Detailed Implementation

[0055] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0056] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein.

[0057] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0058] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0059] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0060] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0061] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0062] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0063] Example 1

[0064] Figure 1 This is a flowchart illustrating a target recognition method according to Embodiment 1 of the present invention. This embodiment is applicable to the identification of calcified plaques in blood vessels, and optionally also to the identification of targets in other tubular locations. The method can be executed by a target recognition device, which can be implemented in hardware and / or software, and can be configured in a smart terminal or cloud server. Figure 1 As shown, the method includes:

[0065] S110. Obtain the segmented fused image of the tubular part, and the cross-sectional images corresponding to each cross-section position in the tubular part.

[0066] In this embodiment of the invention, a tubular region can be understood as a region in the human body that presents a tubular shape, such as blood vessels or trachea. The segmentation result fusion image can be understood as an image obtained by fusing the segmentation result image of the tubular region with a medical image of the tubular region.

[0067] Optionally, the method for obtaining the segmentation result fused image in this embodiment may include: obtaining an initial medical image of the tubular part, and performing segmentation processing on the initial medical image to obtain a segmentation result image of the tubular part; and fusing the segmentation result image and the initial medical image to obtain a segmentation result fused image of the tubular part.

[0068] Specifically, the tubular portion can be scanned using medical imaging equipment to obtain an initial medical image corresponding to the tubular portion. It should be noted that, since the problem addressed in this embodiment is to improve the recognition efficiency of 3D images, the medical imaging equipment used in this embodiment is a 3D scanning device to obtain an initial 3D medical image. For example, a CT imaging device can be used to scan the tubular portion to obtain a corresponding 3D medical image. Of course, the above examples are merely illustrative of this embodiment and are not intended to limit the technical solution of this embodiment. Other 3D imaging devices can also be used for scanning, and this embodiment does not limit this approach.

[0069] Based on the initial medical image of the tubular region, image segmentation processing is performed on the initial medical image to obtain the segmented image of the tubular region. Optionally, a pre-trained image segmentation model can be used to segment the initial medical image to obtain the segmented image, or a pre-processed image segmentation calculation expression can be used to process the initial medical image to obtain the segmented image of the tubular region.

[0070] Furthermore, the obtained segmentation result image and the obtained initial medical image are fused to obtain a fused segmentation result image of the tubular region. Optionally, the image fusion method can be to input the segmentation result image and the initial medical image into a pre-trained image fusion model to obtain the fused segmentation result image output by the fusion model. Of course, traditional image fusion processing can also be performed based on the pixel values ​​of the two images to obtain the fused segmentation result image.

[0071] It should be noted that, compared to directly using the initial medical image for target recognition, the technical solution of using segmented and fused images for subsequent target recognition can make tubular parts more obvious in the image, thereby reducing the probability of misidentification when identifying tubular parts and thus improving recognition efficiency.

[0072] Specifically, based on the fused image of the segmented tubular portion, the cross-sectional positions of the tubular portion are determined. The fused image is then truncated at any cross-sectional position to obtain the corresponding cross-sectional image. For example, the cross-sectional position and the fused image can be input into a pre-acquired 3D image rendering model to obtain the corresponding cross-sectional image.

[0073] In practical applications, if the tubular portion represents a coronary artery, a 3D scan of the coronary artery is performed to obtain a corresponding medical image of the vessel. Image segmentation is then performed on the medical image to obtain a segmented vessel image. Finally, the segmented vessel image and the medical image are fused to obtain a fused segmented vessel image. See the example below. Figure 2 , Figure 2 The segmented images of blood vessels are fused together.

[0074] Specifically, any cross-sectional location of the blood vessel is determined, and a cross-sectional image at that location is determined by fusing the image based on the segmentation results of the blood vessel. For example, see [link to example]. Figure 3 , Figure 3 This is a cross-sectional image of a blood vessel at any transverse position.

[0075] S120. Determine the polar coordinate image of each cross section position based on the cross section image.

[0076] In this embodiment of the invention, the tubular portion may include multiple cross-sectional positions. Correspondingly, the cross-sectional images corresponding to each cross-sectional position can be determined based on the segmentation result fusion image of the tubular portion. In routine operation, medical images obtained from medical imaging scanning equipment are all Cartesian coordinate system images. Therefore, the segmentation result fusion image obtained by processing the initial medical image in this embodiment is also a Cartesian coordinate system image. To reduce the amount of data processing required for subsequent image recognition and improve image recognition efficiency, this embodiment uses the method of converting the Cartesian coordinate system image to a polar coordinate system image. In other words, the cross-sectional images of each cross-sectional position in the tubular portion obtained based on the above implementation method are subjected to coordinate system transformation processing to obtain the cross-sectional polar coordinate images corresponding to each cross-sectional position.

[0077] Optionally, in this embodiment, a pre-trained coordinate transformation model can be used for image coordinate system transformation. The cross-sectional image is input into this model to obtain the polar coordinate image of the cross-section output by the model. Alternatively, traditional image processing expressions can be used for coordinate system transformation. Pixels from the cross-sectional image are input into these expressions to obtain processed image data. After rendering the image data, the polar coordinate image of the cross-section is obtained. Other coordinate transformation methods can also be used for image processing in this embodiment; however, the specific method for converting the cross-sectional image to a polar coordinate image is not limited in this embodiment.

[0078] In practical applications, in obtaining Figure 3 Based on a cross-sectional image of any transverse position of the blood vessel shown, a coordinate system transformation is performed on the cross-sectional image to obtain a corresponding polar coordinate image. See the example below. Figure 4 , Figure 4 This is a schematic diagram of a cross-sectional polar coordinate image of a blood vessel at a transverse location. Optionally, the same transformation method can be used when transforming cross-sectional images at other transverse locations to obtain the corresponding cross-sectional polar coordinate images, in order to ensure consistency in image processing and thus ensure the accuracy of subsequent identification.

[0079] S130. Obtain the pixel values ​​of the pixels in the polar coordinate images of each cross section, and determine the target recognition result of the tubular part based on the pixel values ​​and the preset pixel threshold.

[0080] In this embodiment of the invention, target identification of the tubular portion can be understood as identifying whether the tubular portion contains a target object. Optionally, directly identifying the target in the tubular portion involves a large amount of data processing computation, resulting in reduced identification efficiency. Furthermore, this embodiment includes multiple cross-sectional positions, so the target identification result for each cross-sectional position can be determined separately, and the results of the target identification at each cross-sectional position can be combined and statistically analyzed to obtain the target identification result for the tubular portion.

[0081] Optionally, the method for determining the target recognition result of the tubular portion in this embodiment may include: for any cross-sectional position, determining the average pixel value and maximum pixel value of each pixel in the current cross-sectional position based on the cross-sectional polar coordinate image corresponding to the current cross-sectional position; determining the pixel ratio of the current cross-sectional position based on the average pixel value and the maximum pixel value; obtaining a preset pixel ratio threshold; determining the target recognition result of the current cross-sectional position based on the pixel ratio threshold and the pixel ratio; the target recognition result includes whether the current cross-sectional position contains a target object; and determining the target recognition result of the tubular portion based on the target recognition results of each cross-sectional position.

[0082] Specifically, for any cross-sectional polar coordinate image corresponding to a cross-sectional position, the pixel value of each pixel in the cross-sectional polar coordinate image is determined. The largest pixel value among these pixel values ​​is taken as the maximum pixel value at the current cross-sectional position. Further, the average pixel value at the current cross-sectional position is obtained by averaging the pixel values. The average pixel value is compared with the maximum pixel value to obtain the pixel ratio at the current cross-sectional position. A preset pixel ratio threshold is obtained, and the obtained pixel ratio is compared with the pixel ratio threshold. If the pixel ratio is greater than the pixel ratio threshold, it indicates that the current cross-sectional position contains a target object; conversely, if the pixel ratio is less than the pixel ratio threshold, it indicates that the current cross-sectional position does not contain a target object. Optionally, based on the above implementation method, it is determined whether each cross-sectional position contains a target object. In the above implementation method, different thresholds can be set based on different medical images scanned by different imaging devices, overcoming the limitations of the traditional fixed threshold method, and using a simple and lightweight method to identify target objects, thus improving the accuracy of the identification results.

[0083] It should be noted that if the position threshold between the current cross-sectional position containing the target object and its adjacent cross-sectional position is less than a preset merging threshold, the two cross-sectional positions are merged. After merging the cross-sectional positions that meet the merging conditions, each target object corresponding to the merged cross-sectional position is considered a complete target object, thus ensuring the accuracy of the number and size of the identified target objects. It should also be noted that if the position threshold between the current cross-sectional position containing the target object and its adjacent cross-sectional position is greater than a preset filtering threshold, the object identified at that cross-sectional position is considered noise, and therefore, that cross-sectional position does not contain a target object. The beneficial effect of the above implementation is that it can filter out false positives in the identification results, reduce the probability of misidentification in the identification results, and thus improve the accuracy of the identification results.

[0084] In practical applications, the maximum and average pixel values ​​corresponding to each cross-sectional position are obtained. See examples below. Figure 5 , Figure 5This section presents the pixel values ​​and target recognition results for each cross-sectional location of the aforementioned blood vessel. Line A represents the maximum pixel value at each cross-sectional location, line B represents the average pixel value at each cross-sectional location, and square wave C represents the calcified plaque recognition result at each cross-sectional location. Specifically, a threshold value t is set for the ratio of the maximum pixel value to the average pixel value. If the ratio is greater than t, the cross-sectional location is considered to contain calcified plaques, i.e., a calcified region, and the corresponding square wave is activated. Conversely, if the ratio is less than t, the cross-sectional location is considered not to contain calcified plaques, i.e., a normal region, and the corresponding square wave is not activated. Further, the identified calcified regions undergo appropriate processing, such as merging adjacent calcifications and filtering out isolated points as noise. The final output is the location of the calcified plaque.

[0085] The technical solution of this invention obtains a segmented fused image of the tubular portion and cross-sectional images corresponding to each transverse position within the tubular portion; determines the polar coordinate image of each transverse position based on each cross-sectional image; obtains the pixel values ​​of pixels in each polar coordinate image; and determines the target recognition result of the tubular portion based on the pixel values ​​and a preset pixel threshold. This technical solution does not set a fixed grayscale threshold, but instead sets a threshold for the ratio of the maximum pixel value to the average pixel value. This makes the numerical comparison unaffected by the grayscale distribution range, making it more applicable and flexible. Furthermore, the method of merging and filtering recognition results effectively removes noise and false positives, resulting in higher accuracy and robustness of the recognition results.

[0086] Example 2

[0087] Figure 6 This is a flowchart of a target recognition method provided in Embodiment 2 of the present invention. Based on the above embodiments, this embodiment optionally includes determining the polar coordinate image of each cross-sectional position based on each cross-sectional image, including:

[0088] For any cross-sectional image, determine multiple preset diameters of the current cross-sectional image, and determine the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters;

[0089] Image optimization processing was performed on the polar coordinate images of each initial cross-section to determine the polar coordinate images of the cross-sections corresponding to each cross-sectional position. For example... Figure 6 As shown, the method includes:

[0090] S210. Obtain the segmented fused image of the tubular portion, and the cross-sectional images corresponding to each cross-sectional position in the tubular portion.

[0091] S220. For any cross-sectional image, determine multiple preset diameters of the current cross-sectional image, and determine the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters.

[0092] In this embodiment of the invention, when the tubular portion is a blood vessel, the cross-sectional image of the blood vessel is approximately circular. Since the plaque to be identified is located on the outer wall of the blood vessel, a preset circle that can encompass the cross-sectional image of the blood vessel is pre-defined. To ensure that the obtained circle can enclose the current cross-sectional image, the set ray length must be greater than the radius from the center point to the boundary of the cross-sectional image, facilitating subsequent identification of the target object. Specifically, the preset circle can be centered at the center point of the blood vessel, and its radius can be the sum of the radius from the center point to the boundary of the cross-sectional image and the preset boundary length. The preset circle contains multiple preset diameters, and the initial cross-sectional polar coordinate image corresponding to the current cross-sectional image can be determined based on these multiple preset diameters.

[0093] Optionally, the method for determining the initial cross-sectional polar coordinate image corresponding to the current cross-sectional image in this embodiment may include: for any preset diameter in the current cross-sectional image, determining the pixel value of the diameter pixel point corresponding to the current preset diameter in the cross-sectional image based on the diameter rotation angle corresponding to the current preset diameter; determining the splicing order corresponding to each diameter rotation angle, and splicing the diameter pixels corresponding to multiple preset diameters according to the splicing order to obtain the initial cross-sectional polar coordinate image of the current cross-sectional image.

[0094] Specifically, taking the center point of the current cross-sectional image as the center, rays of a preset length are emitted at preset degree intervals. The number of emitted rays is determined based on one rotation, and the endpoints of each ray are connected to obtain a preset circle. The starting points of two rays with an included angle of 180° within the preset circle are connected to form a preset diameter of the circle. Optionally, multiple preset diameters of the preset circle can be obtained based on the above method.

[0095] In this embodiment of the invention, the diameter rotation angle is the angle between the right radius of the current preset diameter and the radius of the first ray. The diameter rotation angle corresponding to the current preset diameter is obtained based on the process of forming a preset circle, and then the pixel point corresponding to the current preset diameter in the current cross-section can be determined based on this diameter rotation angle. Optionally, the method for determining the diameter pixel point may include: for any point on the current preset diameter, determining multiple adjacent pixels adjacent to that point, and determining the nearest pixel point among the adjacent pixels based on a preset expression, and using the pixel value of the nearest pixel point as the pixel value corresponding to the diameter pixel point of that point in the current preset diameter. Optionally, the pixel values ​​of each diameter pixel point corresponding to each preset diameter are determined respectively based on the above method.

[0096] Specifically, the stitching order can be determined based on the rotation angle of each diameter. Then, based on the stitching order, the diameter pixels corresponding to each preset diameter are stitched together to obtain the initial polar coordinate image of the current cross-section image.

[0097] In practical applications, 36 rays are emitted every 10°, with the center of the cross-sectional image of the blood vessel as the rotation center. The starting point of each ray is the midline point, and the ending point is a circle with a radius of 16 pixels. For each ray, 16 pixels are formed in the polar coordinate image, and the value of each pixel is the Cartesian coordinate value of the nearest pixel to its ray. See also the exemplary examples. Figure 7 , Figure 7 This is a diagram illustrating how to determine the pixel value of a diameter pixel. If the coordinates of a point on the ray are (x, y), calculate the distances between this point and its four nearest neighbors, and take the pixel value of the neighbor with the smallest distance as the pixel value of this point. For example, the pixel value of the diameter pixel can be determined based on the following expression:

[0098]

[0099] Where P(x,y) represents the pixel value of the point on the ray corresponding to the polar coordinate image to be generated, P is the polar coordinate image, and D(u,v) represents the pixel value of the point with coordinates (u,v) in the Cartesian coordinate system of the image (original image).

[0100] Specifically, a 36x16 polar coordinate image can be generated by rotating the polar coordinates counterclockwise by 360°, and a 36x16 polar coordinate image can be generated by rotating the polar coordinates clockwise by 360°. Combining the two will give you the initial 36x32 polar coordinate image of each cross section.

[0101] S230. Perform image optimization processing on the polar coordinate images of each initial cross section to determine the polar coordinate images of the cross section corresponding to each cross section position.

[0102] In this embodiment of the invention, to make the subsequent target recognition results of the tubular part based on the polar coordinate image more accurate, the technical solution of this embodiment further optimizes the initial cross-sectional polar coordinate image to obtain an optimized cross-sectional polar coordinate image. The optimization process in this embodiment includes edge optimization and dilation optimization.

[0103] Optionally, the method for edge optimization processing of the initial cross-sectional image in this embodiment may include: determining the equivalent diameter corresponding to each cross-sectional image, and determining the cross-sectional image to be processed in each cross-sectional image based on a preset equivalent diameter threshold and each equivalent diameter; performing cross-sectional edge optimization processing on the cross-sectional polar coordinate image corresponding to the cross-sectional image to be processed to obtain the cross-sectional polar coordinate image of the tubular part.

[0104] In this embodiment of the invention, the equivalent diameter can be understood as the diameter of the near-circular shape corresponding to the current cross-sectional image. Specifically, for the current cross-sectional image, the number of pixels with a value of 1 along each row of the vertical axis is calculated, which is the "diameter" in each direction. The average of these diameters is then calculated as the equivalent diameter of the cross-section.

[0105] In practical applications, branch vessels appear as bulges on polar coordinate images, leading to an overestimation of the calculated equivalent diameter and impacting subsequent calculations. Therefore, they need to be removed. In other words, the process of removing branch vessels is the image edge optimization process described in the above embodiment. See the example below. Figure 8 , Figure 8 This is a schematic diagram of branching vessels within a blood vessel. To remove branching vessels, it is necessary to determine the equivalent diameter corresponding to the cross-sectional image of the vessel at each cross-sectional position. See the example below. Figure 9 , Figure 9 This is a schematic diagram showing the equivalent diameter of a blood vessel at various cross-sectional positions. Figure 9 In the image, the horizontal axis represents the distance from the start to the end of the blood vessel, and the vertical axis represents the equivalent diameter. The blue curve represents the equivalent diameter of each cross-section, and some outliers are the branch points. Then, a spline is used to fit the curve, and the blue curve is smoothed to obtain the orange fitted curve. Then, the polar coordinate image is processed using the fitted equivalent diameter to erase those local diameters that are larger than the fitted diameter, thus "shrinking" the protruding branch positions, resulting in the processed cross-sectional polar coordinate image.

[0106] Specifically, unlike Cartesian coordinates, branch erasure of polar coordinate images can be performed directly using assignment methods, greatly reducing the computational load. For example, see... Figure 10 , Figure 10 This is a schematic diagram of the process of wiping away branch vessels in a blood vessel. Figure 10 The pixel values ​​outside the positions of the yellow line (7 and 20) are set to 0 in advance, that is, the pixel values ​​of x coordinates <7 and >20 are all 0, so that the polar coordinate image of the erased cross section can be obtained.

[0107] Optionally, the method for performing dilation optimization processing on the initial cross-sectional image in this embodiment may include: obtaining a preset dilation parameter, and performing dilation optimization processing on the cross-sectional polar coordinate images corresponding to each cross-sectional image based on the dilation parameter to obtain the cross-sectional polar coordinate image of the tubular part.

[0108] In practical applications, since the segmentation results of coronary arteries only cover the lumen area, dilation is necessary to include as much calcified plaque as possible. In Cartesian coordinates, image dilation requires complex geometric calculations, while in polar coordinates, it only requires horizontal expansion. This implementation reduces the computational load, thus improving calculation speed. See the example below. Figure 11 , Figure 11 This is a schematic diagram of the dilation process of the initial cross-sectional polar coordinate image. Figure 11 In the image, the arrow points to the first row. The diameter before dilation is 12 (12 pixels equal to 1). To dilate by 6 pixels (3 pixels on each side), simply assign a value of 1 to the first three pixels of the left boundary and a value of 1 to the last three pixels of the right boundary. This will yield the dilated cross-sectional polar coordinate image. This method of assigning values ​​to dilate the initial cross-sectional polar coordinate image reduces computational load, thereby improving image processing efficiency and ultimately enhancing overall recognition efficiency.

[0109] It should be noted that the execution order of edge optimization and dilation optimization is not limited in this embodiment.

[0110] S240. Obtain the pixel values ​​of the pixels in the polar coordinate images of each cross section, and determine the target recognition result of the tubular part based on the pixel values ​​and the preset pixel threshold.

[0111] The technical solution of this invention, based on obtaining a cross-sectional image of the tubular portion's cross-sectional location, pre-determines the equivalent diameter in the cross-sectional image and stitches together the pixels of the equivalent diameter to obtain a converted initial cross-sectional polar coordinate image. Then, by performing edge optimization and dilation optimization on the cross-sectional polar coordinate image, a new cross-sectional polar coordinate image is obtained. Finally, pixel value processing is performed on the cross-sectional polar coordinate images corresponding to each cross-sectional position to obtain the target recognition result of the tubular portion. This embodiment reduces the computational load of data processing, improves data processing efficiency, and thus improves the overall recognition efficiency.

[0112] Example 3

[0113] Figure 12 This is a schematic diagram of the structure of a target recognition device provided in Embodiment 3 of the present invention. Figure 12As shown, the device includes: an image acquisition module 310, a cross-sectional polar coordinate image determination module 320, and a target recognition result determination module 330; wherein,

[0114] Image acquisition module 310 is used to acquire the segmentation result fused image of the tubular part, and the cross-sectional images corresponding to each cross-section position in the tubular part;

[0115] The cross-sectional polar coordinate image determination module 320 is used to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions based on each of the cross-sectional images;

[0116] The target recognition result determination module 330 is used to obtain the pixel value of each pixel in the polar coordinate image of each cross section, and determine the target recognition result of the tubular part based on the pixel value and a preset pixel threshold.

[0117] Based on the above embodiments, optionally, the image acquisition module 310 includes:

[0118] The segmentation result image acquisition unit is used to acquire an initial medical image of the tubular part and perform segmentation processing on the initial medical image to obtain a segmentation result image of the tubular part;

[0119] The segmentation result fusion image acquisition unit is used to fuse the segmentation result image and the initial medical image to obtain the segmentation result fusion image of the tubular part.

[0120] Based on the above embodiments, optionally, the cross-sectional polar coordinate image determination module 320 includes:

[0121] The initial cross-sectional polar coordinate image acquisition unit is used to determine multiple preset diameters of the current cross-sectional image for any cross-sectional image, and to determine the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters respectively;

[0122] The cross-sectional polar coordinate image acquisition unit is used to perform image optimization processing on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions.

[0123] Optionally, based on the above embodiments, the initial cross-sectional polar coordinate image acquisition unit includes:

[0124] The diameter pixel point determination subunit is used to determine the pixel value of the diameter pixel point corresponding to any preset diameter in the current cross-sectional image based on the diameter rotation angle corresponding to the current preset diameter;

[0125] The initial cross-sectional polar coordinate image is used to obtain sub-units to determine the splicing order corresponding to the rotation angle of each diameter. Based on the splicing order, the diameter pixels corresponding to the multiple preset diameters are spliced ​​to obtain the initial cross-sectional polar coordinate image of the current cross-sectional image.

[0126] Based on the above embodiments, optionally, the optimization process includes edge optimization processing;

[0127] The unit for obtaining the polar coordinate image of the cross section includes:

[0128] The subunit for obtaining cross-sectional images to be processed is used to determine the equivalent diameter corresponding to each of the cross-sectional images respectively, and to determine the cross-sectional image to be processed in each of the cross-sectional images based on a preset equivalent diameter threshold and each of the equivalent diameters;

[0129] The first cross-sectional polar coordinate image acquisition subunit is used to perform cross-sectional edge optimization processing on the cross-sectional polar coordinate image corresponding to the cross-sectional image to be processed, so as to obtain the cross-sectional polar coordinate image of the tubular part.

[0130] Optionally, based on the above embodiments, the optimization process includes an expansion optimization process;

[0131] The second cross-sectional polar coordinate image acquisition subunit is used for the cross-sectional polar coordinate image acquisition unit, including: acquiring a preset dilation parameter, and performing dilation optimization processing on the cross-sectional polar coordinate images corresponding to each of the cross-sectional images based on the dilation parameter to obtain the cross-sectional polar coordinate image of the tubular part.

[0132] Based on the above embodiments, optionally, the target recognition result determination module 320 includes:

[0133] The pixel value acquisition unit is used to determine the average pixel value and the maximum pixel value of each pixel point in the current cross-section position based on the cross-sectional polar coordinate image corresponding to the current cross-section position for any cross-section position.

[0134] A pixel ratio determination unit is used to determine the pixel ratio at the current cross-sectional position based on the average pixel value and the maximum pixel value;

[0135] A cross-sectional target recognition result determination unit is used to obtain a preset pixel ratio threshold and determine the target recognition result at the current cross-sectional position based on the pixel ratio threshold and the pixel ratio value;

[0136] The part target recognition result determination unit is used to determine the target recognition result of the tubular part based on the target recognition result of each of the cross-sectional positions.

[0137] Based on the above embodiments, optionally, the tubular portion includes a coronary artery, and the target identification result includes whether the coronary artery contains plaque.

[0138] The target recognition device provided in the embodiments of the present invention can execute the target recognition method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0139] Example 4

[0140] Figure 13 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0141] like Figure 13 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0142] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0143] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as object recognition methods.

[0144] In some embodiments, the target recognition method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the target recognition method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the target recognition method by any other suitable means (e.g., by means of firmware).

[0145] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0146] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0147] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0148] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0149] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0150] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0151] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0152] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A target recognition method, characterized in that, include: Obtain the segmented fused image of the tubular portion, as well as the cross-sectional images corresponding to each cross-sectional position in the tubular portion; Based on each of the cross-sectional images, determine the polar coordinate image of the cross section corresponding to each of the cross-sectional positions; For any cross-sectional position, the average pixel value and maximum pixel value of each pixel in the current cross-sectional position are determined based on the cross-sectional polar coordinate image corresponding to the current cross-sectional position; The pixel ratio at the current cross-section position is determined based on the average pixel value and the maximum pixel value; A preset pixel ratio threshold is obtained, and the target recognition result at the current cross-sectional position is determined based on the pixel ratio threshold and the pixel ratio value; wherein, the target recognition result includes whether the current cross-sectional position contains a target object; The target identification result of the tubular part is determined based on the target identification result at each of the cross-sectional positions; The method further includes: If the position threshold between the current cross-section containing the target object and the adjacent cross-section containing the target object is less than a preset merging threshold, then the current cross-section and the adjacent cross-section are merged. If the position threshold between the current cross-section containing the target object and the adjacent cross-section containing the target object is greater than a preset filtering threshold, then the target identification result at the current cross-section is filtered out as noise.

2. The method according to claim 1, characterized in that, The process of obtaining the segmented and fused image of the tubular portion includes: An initial medical image of the tubular portion is acquired, and the initial medical image is segmented to obtain a segmentation result image of the tubular portion; The segmented image and the initial medical image are fused to obtain the fused segmented image of the tubular region.

3. The method according to claim 1, characterized in that, The step of determining the polar coordinate image of each cross-section position based on each of the cross-sectional images includes: For any cross-sectional image, determine multiple preset diameters of the current cross-sectional image, and determine the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters respectively; Image optimization processing is performed on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each cross-sectional position.

4. The method according to claim 3, characterized in that, The step of determining the initial cross-sectional polar coordinate image of the current cross-sectional image based on the diameter pixels corresponding to the multiple preset diameters includes: For any preset diameter in the current cross-sectional image, the pixel value of the diameter pixel point corresponding to the current preset diameter in the cross-sectional image is determined based on the diameter rotation angle corresponding to the current preset diameter; Determine the stitching order corresponding to each diameter rotation angle, and stitch together the diameter pixels corresponding to the multiple preset diameters according to the stitching order to obtain the initial cross-sectional polar coordinate image of the current cross-sectional image.

5. The method according to claim 3, characterized in that, The optimization process includes edge optimization processing; The step of performing image optimization processing on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions includes: The equivalent diameter corresponding to each of the cross-sectional images is determined respectively, and the cross-sectional image to be processed in each of the cross-sectional images is determined based on the preset equivalent diameter threshold and each of the equivalent diameters; the cross-sectional edge optimization processing is performed on the cross-sectional polar coordinate image corresponding to the cross-sectional image to be processed to obtain the cross-sectional polar coordinate image of the tubular part.

6. The method according to claim 3, characterized in that, The optimization process includes dilation optimization; The step of performing image optimization processing on each of the initial cross-sectional polar coordinate images to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions includes: Obtain preset dilation parameters, and perform dilation optimization processing on the cross-sectional polar coordinate images corresponding to each of the cross-sectional images based on the dilation parameters to obtain the cross-sectional polar coordinate images of the tubular part.

7. The method according to claim 1, characterized in that, The tubular portion includes coronary arteries, and the target identification result includes whether the coronary arteries contain plaques.

8. A target recognition device, characterized in that, include: The image acquisition module is used to acquire the segmented result fused image of the tubular part, as well as the cross-sectional images corresponding to each cross-section position in the tubular part; A cross-sectional polar coordinate image determination module is used to determine the cross-sectional polar coordinate image corresponding to each of the cross-sectional positions based on each of the cross-sectional images; The target recognition result determination module is used to obtain the pixel value of each pixel in the polar coordinate image of each cross section, and determine the target recognition result of the tubular part based on the pixel value and a preset pixel threshold. The target recognition result determination module includes: The pixel value acquisition unit is used to determine the average pixel value and the maximum pixel value of each pixel point in the current cross-section position based on the cross-sectional polar coordinate image corresponding to the current cross-section position for any cross-section position. A pixel ratio determination unit is used to determine the pixel ratio at the current cross-sectional position based on the average pixel value and the maximum pixel value; A cross-sectional target recognition result determination unit is used to obtain a preset pixel ratio threshold and determine the target recognition result at the current cross-sectional position based on the pixel ratio threshold and the pixel ratio value; wherein, the target recognition result includes whether the current cross-sectional position contains a target object; A part target recognition result determination unit is used to determine the target recognition result of the tubular part based on the target recognition result of each of the cross-sectional positions; The device is also used for: If the position threshold between the current cross-section containing the target object and the adjacent cross-section containing the target object is less than a preset merging threshold, then the current cross-section and the adjacent cross-section are merged; if the position threshold between the current cross-section containing the target object and the adjacent cross-section containing the target object is greater than a preset filtering threshold, then the target identification result at the current cross-section is filtered out as noise.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the target recognition method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the target recognition method according to any one of claims 1-7.

Images