Blood vessel image processing method, interaction displaying method and computer device

[0089] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following further describes the present invention in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

[0090] In the embodiments of this application, a method for processing blood vessel images is provided, which can be used in different application scenarios according to the needs of different operators for the blood vessel segments contained in the blood vessel tree presented in the blood vessel image. The selected display can also enhance the richness of the image display; specifically:

[0091] figure 1 It is a schematic flowchart of a method for processing a blood vessel image in a blood vessel image in an optional embodiment. Such as figure 1 As shown, a method for processing blood vessel images can be applied to selectively display blood vessel segments contained in tree-like blood vessels (ie, blood vessel trees) in blood vessel images such as pulmonary blood vessels and liver blood vessels to meet different application scenarios. Selective display requirements for different blood vessel segments; the above processing method may include the following steps:

[0092] Step S01: Obtain a blood vessel image, the blood vessel image includes blood vessels in a blood vessel tree structure, and the blood vessel tree may include multiple blood vessel segments.

[0093] Specifically, the above-mentioned blood vessel image can be obtained by extraction such as a blood vessel tracking model.

[0094] Step S02: Establish and use the vascular segment classification rules to classify the blood vessel segments of the blood vessel tree to obtain a first classification result.

[0095] Specifically, the corresponding blood vessel segment grading rules can be set based on specific needs, and the blood vessel segment grading rules can be used to classify all the blood vessel segments included in the blood vessel tree in the blood vessel image obtained above, such as each The blood vessel segment is marked with a grading level to obtain the first grading result.

[0096] Step S03: Perform grading display of the blood vessel segment according to the first grading result.

[0097] Specifically, according to the display rules, all the blood vessel segments included in the blood vessel tree in the blood vessel image can be hierarchically displayed according to the first grading result obtained above, so as to realize the selective display of each blood vessel segment.

[0098] figure 2 It is a schematic flowchart of a method for processing a blood vessel image in a blood vessel image in another optional embodiment. Such as figure 2 As shown, in another optional embodiment, a method for processing a blood vessel image can be applied to the selective display of blood vessel segments included in the blood vessel tree in the blood vessel image, and the processing method may include:

[0099] Step S11: Obtain a blood vessel image, and obtain image information of the blood vessel image and obtain parameter information of the blood vessel segment included in the blood vessel tree in the blood vessel image.

[0100] Specifically, a blood vessel tracking method can be used to obtain the aforementioned blood vessel image, image information, parameter information, etc. In addition, it is also possible to obtain the centerline of each blood vessel segment by performing morphological skeletonization on the binary mask image of the segmentation result of the blood vessel tree, or to process the blood vessel image by the distance transformation method or the fast travel method. The distance field is used to extract the center line of each blood vessel segment; the geometric parameters of each center line are then calculated to obtain the above-mentioned parameter information.

[0101] The above-mentioned parameter information may include the attributes of the blood vessel segment, the angle between the current blood vessel segment and the parent branch blood vessel segment, the center position of the blood vessel segment, the extension direction of the blood vessel segment, and the blood vessel caliber at the center of the blood vessel segment. In addition, the attributes of the blood vessel segment may include root main branches, intermediate branches, and peripheral branches. The corresponding parameter information may also include information such as the center points of the peripheral branches of the current blood vessel segment. At the same time, the image information may include the contrast parameter of the blood vessel segment, the matching degree parameter of the blood vessel segment, and the average gray value of the area around the blood vessel segment.

[0102] It should be noted that the above-mentioned image information and parameter information can be obtained at the same time as the above-mentioned blood vessel image, or the image information and/or parameter information can be obtained as needed before the subsequent corresponding processing steps, so as to reduce the acquisition of image information and Processing resources and complexity of parameter information.

[0103] Step S12: Perform noise reduction processing on the blood vessel tree according to the image information and/or parameter information.

[0104] Specifically, due to the complicated distribution of blood vessels in various organs, generally in a tree-like distribution, the traditional vascular tree of various internal organs extracted based on blood vessel image segmentation may have defects such as spikes, short branch ends, and overflow. When angiography is used to obtain vascular images of internal organs, the traditional method is to denoise the vascular tree based on the morphology of the image pixel space. Since it can only obtain one-dimensional information of the image plane image, it cannot be effective. Eliminate defects such as spikes and short branch ends of the vascular tree. In response to the above problems, in this embodiment, based on the parameter information of each vessel segment on the vessel tree, or combined with the image information of the vessel image, noise defects such as the removal of spurs and/or short branch ends can be effectively removed from the vessel tree.

[0105] For example, the blood vessel tree may be denoised according to the parameter information of the blood vessel segment and/or the image information of the blood vessel image. Specifically, the first noise reduction rule can be preset, and then according to the parameter information of the blood vessel segment, the first noise reduction rule can be used to remove the terminal spurs and/or short branches of the vascular tree, thereby avoiding the appearance of the vascular tree in the blood vessel image such as Defects such as spikes, short branch ends and overflow.

[0106] In an optional embodiment, when performing noise reduction processing on the vascular tree based on parameter information, the parameter threshold may be set first, and then the vascular segment whose parameter information is less than the above-mentioned parameter threshold is deleted to remove spurs on the vascular tree. And short branch tips. Since the above-mentioned parameter information is based on the three-dimensional three-dimensional size information of each blood vessel segment obtained by the geometric model, it can make all-round judgments on spurs and short branch ends. Compared with the traditional one-dimensional size information obtained only on blood vessel images, It can judge and remove noises such as spurs and short branch ends more accurately.

[0107] In another optional embodiment, multiple iterations of noise reduction processing may be performed on the blood vessel tree to further improve the accuracy and effectiveness of the noise reduction processing.

[0108] Step S13: Establish a blood vessel segment classification rule, and use the blood vessel segment classification rule to classify the blood vessel segments of the blood vessel tree to obtain a first classification result.

[0109] Step S14, setting the blood vessel segment display classification level.

[0110] Specifically, the above-mentioned blood vessel segment display hierarchical level can be set according to specific application scenarios and the blood vessel tree display requirements.

[0111] Step S15: According to the blood vessel segment display grading level and the first grading result, the blood vessel segment contained in the blood vessel tree is adjusted, and the blood vessel segment contained in the adjusted blood vessel tree is displayed. The blood vessel segment can be set in advance. Display the grading level, and then according to the blood vessel segment display grading level and the first grading result, adjust the blood vessel segment contained in the blood vessel tree, and display the blood vessel segment contained in the blood vessel tree after adjustment to realize the blood vessel The outline of the tree.

[0112] In an optional embodiment, the aforementioned adjustment operation may be an operation of adding and/or deleting a blood vessel segment according to the display hierarchical level of the blood vessel segment. For example, the blood vessel segment with a classification level greater than the blood vessel segment display classification level is deleted, and/or the blood vessel segment with a classification level less than or equal to the blood vessel segment display classification level is added.

[0113] Step S16: Sort the first grading result according to the image information and/or parameter information to obtain the second grading result.

[0114] Specifically, the second grading result can be obtained by sorting the first grading result according to the size of the grading level; wherein, for blood vessel segments with the same grading level, the ranking can also be performed based on parameter information and/or image information.

[0115] It should be noted that the blood vessel segment of the vascular tree after the noise reduction process can be graded using the vascular segment classification rule to obtain the first classification result; or the blood vessel segment of the vascular tree can be classified first to obtain the above-mentioned first classification result. After the first classification result, noise reduction processing is performed on the first classification result. At the same time, the subsequent second grading result can be directly obtained by sorting the first grading result, or the first grading result can be denoised first, and then the second grading result can be obtained by sorting.

[0116] Step S17, setting the completeness of the blood vessel tree display.

[0117] Specifically, according to specific application scenarios and the vascular tree display requirements, the completeness of the vascular tree display can be set to correspond to the application scenarios and display requirements, and to further improve the accuracy of displaying corresponding vascular segments in the vascular tree.

[0118] In step S18, the blood vessel segments included in the blood vessel tree are selectively displayed according to the completeness of the blood vessel tree display and the second classification result.

[0119] Specifically, the addition and deletion operations can be performed in sequence according to the arrangement order of the blood vessel segments in the second classification result, until the number of blood vessel segments contained in the blood vessel tree is relative to the number of blood vessel segments contained in the blood vessel tree before the addition or deletion operation (or relative to the initial The number of blood vessel segments contained in the blood vessel tree in the blood vessel image) satisfies the completeness level of the blood vessel tree display, and then the blood vessel segments contained in the blood vessel tree after the addition and deletion operations are displayed to achieve selective display of the blood vessel segments; wherein, The adding and deleting operations include sequentially deleting and/or adding the blood vessel segments in the second grading result. In addition, each blood vessel segment in the second grading result may be added to a first-in-first-out queue according to the sorting order, and deleted sequentially according to the order of the queue.

[0120] It should be noted that in other optional embodiments, the blood vessel segment can be selectively displayed according to the completeness of the blood vessel tree display and the second classification result combined with the noise reduction processing result to improve the display of the blood vessel segment Richness. For example, the blood vessel segment obtained by the blood vessel segment tracking method can be used as the first level (that is, the highest level of completeness) of the blood vessel tree display, and the blood vessel segments contained in the blood vessel tree after the noise reduction process can be displayed as the blood vessel tree The second level of completeness, and then according to the second grading result of the vascular segment, the number of levels of the vascular tree showing the degree of completeness is sequentially increased (that is, as the number of levels of the vascular tree showing the degree of completeness increases, the displayed vascular segment of the vascular tree Completeness gradually decreases).

[0121] image 3 It is a schematic flowchart of a method for setting the completeness level of the blood vessel tree display in an optional embodiment. Such as image 3 As shown, in order to increase the richness of the blood vessel image display, the following steps can be used to set the completeness level of the blood vessel tree display, specifically:

[0122] In step S21, the blood vessel segments included in the blood vessel tree obtained by the blood vessel tracking method are set as the first level of completeness (that is, the top level of completeness).

[0123] Step S22: Set the blood vessel segments included in the blood vessel tree after the noise reduction process to the second level of integrity (ie, the second level of integrity).

[0124] Step S23, according to the above-mentioned first grading result, combined with the second grading result, according to the law that the degree of completeness gradually decreases (that is, the number of completeness levels gradually increases), the completeness settings of the next level and subsequent levels are sequentially performed.

[0125] In an optional embodiment, between step S22 and step S23, multiple integrity levels can be set according to the number of noise reduction iterations. Among them, the more the number of noise reduction iterations, the higher the degree of completeness of the blood vessel segments contained in the obtained blood vessel tree is set, and the lower the completeness (that is, the richness of the blood vessel tree).

[0126] Figure 4 It is a schematic flowchart of an interactive display method of a blood vessel image in another optional embodiment. Such as Figure 4 As shown, a method for interactive display of blood vessel images can be based on any embodiment of the method for processing blood vessel images in this application to perform hierarchical display of blood vessel images and/or perform vascular segment display based on the completeness of the blood vessel tree display Selective display of, that is, the interactive display method can be applied to the display of blood vessel images in a blood vessel tree structure, and the blood vessel tree includes multiple blood vessel segments, the method may include the following steps:

[0127] Step S31, pre-store the classification information of each blood vessel segment in the blood vessel tree.

[0128] Specifically, the aforementioned grading information may include a first grading result, and the grading level of the parent branch blood vessel segment in the first grading result is smaller than the grading level of the child branch blood vessel segment.

[0129] Step S32: Obtain the display rule information set by the user; wherein, the display rule information may include the vascular segment display gradation level.

[0130] Step S33, based on the above-mentioned classification information, display the blood vessel segments in the blood vessel tree that satisfy the display rule information.

[0131] Specifically, the blood vessel segments included in the blood vessel tree can be adjusted according to the display grade of the blood vessel segments, and the blood vessel segments included in the blood vessel tree after adjustment can be displayed. Wherein, the above adjustment operation may include deleting blood vessel segments with a classification level greater than the blood vessel segment display classification level, and/or adding blood vessel segments with a classification level less than or equal to the blood vessel segment display classification level.

[0132] In an optional embodiment, the aforementioned grading information may further include a first grading result and a second grading result, and the display rule information includes the completeness level of the blood vessel tree display; the aforementioned step S33 may specifically include:

[0133] First, obtain the parameter information of each blood vessel segment in the blood vessel tree and the image information of the blood vessel image; wherein, the image information may include the contrast parameter of the blood vessel segment, the matching degree parameter of the blood vessel segment, and the average gray value of the area around the blood vessel segment, etc. At least one of them.

[0134] Secondly, use the parameter information and/or the image information to sort the first grading result to obtain a second grading result; wherein, each blood vessel segment in the second grading result is performed according to the size of the grading series Sorting, and for blood vessel segments with the same hierarchical level, sorting may be performed based on the parameter information and/or the image information.

[0135] Finally, the blood vessel segment is selectively displayed according to the second classification result. For example, the level of completeness of the blood vessel tree display can be set first, and then additions and deletions are performed in sequence according to the arrangement order of the blood vessel segments in the second grading result, until the number of blood vessel segments contained in the blood vessel tree is relative to the number before the addition and deletion operation. The number of blood vessel segments contained in the blood vessel tree meets the level of completeness displayed by the blood vessel tree; finally, the blood vessel segments contained in the blood vessel tree after the addition and deletion operations are displayed; wherein, the aforementioned addition and deletion operations include following the second classification result The arrangement order of each blood vessel segment is sequentially deleted and/or added. For example, each blood vessel segment in the second grading result can be added to a first-in first-out queue according to the sorting order, and sequentially deleted according to the order of the queue.

[0136] Figure 5 It is a schematic structural diagram of a computer device in another optional embodiment. Such as Figure 5 As shown, a computer device includes a memory 101, a processor 102, and a computer program stored on the memory 101. The processor 102 is used to execute the computer program stored on the memory 101 as described in this application. Steps in any embodiment of the method for processing blood vessel images, and/or any embodiment of the method for interactive display of blood vessel images.

[0137] In order to facilitate those skilled in the art to understand the technical content of the present invention, the following takes processing operations such as segmentation and display of liver blood vessel images as an example to describe in detail the blood vessel image processing methods in the embodiments of the present application:

[0138] At present, in the process of analyzing and processing liver vascular images, the precise segmentation of hepatic portal vein and hepatic vein for the liver vascular tree is the basis for applications such as liver segmentation and surgical planning.

[0139] However, in the traditional hepatic venous and delayed scanning process, the contrast of the hepatic portal vein and hepatic vein is lower than that of the abdominal artery and the noise is larger. At the same time, they are also affected by the partial volume effect, which makes the traditional Based on the one-dimensional information such as the center line obtained from the liver blood vessel image, the portal vein and hepatic vein obtained by segmentation and extraction of the liver blood vessel tree will have defects such as spurs, short branch ends or overflow. At the same time, based on different application scenarios, different operators have different requirements for the display of the results after the segmentation process, but the traditional liver vascular tree segmentation and extraction results show a relatively single richness, and cannot achieve interactive display, which cannot meet the requirements of different application scenarios for segmentation. Display requirements for extraction results.

[0140] In view of the above-mentioned technical problems, the method for processing the vascular tree in the vascular image provided in the embodiments of the present application is to obtain the grading result by grading and marking each vascular segment of the vascular tree, and displaying the vascular tree according to the grading result. It can also be combined with the above-mentioned parameter information and/or image information of the vascular segment for selective display on the basis of grading marks, so as to realize the corresponding interactive interaction of the extraction results of hepatic portal vein and hepatic vein for different clinical application scenarios Display, to effectively increase the richness of the extraction result display; for example, according to the richness (and/or completeness) of the blood vessel, several levels can be set for each blood vessel segment, and then the user can select the blood vessel extraction result to be presented by himself, and then Realize the interactive richness adjustment of the extraction result display.

[0141] In addition, in the embodiments of the present application, multiple dimensions of data information such as image information based on the blood vessel image and/or parameter information of each blood vessel segment on the liver blood vessel tree can also be used to reduce the noise of the blood vessel tree segmentation result. , Can effectively avoid the noise defects such as spurs, short branch ends or overflow in the portal vein and hepatic vein obtained by segmentation and extraction of the liver vascular tree; at the same time, it can also be combined with the blood vessel segment contained in the vascular tree after the noise reduction process to set the display integrity level , To further enhance the richness of blood vessel segment display.

[0142] Image 6 Is a schematic flow chart of a method for processing a blood vessel tree in a blood vessel image in another optional embodiment, Figure 7 Is a schematic diagram of the blood vessel tree obtained by using the portal vein automatic algorithm in an optional embodiment, Picture 8 Is a schematic diagram of a blood vessel segment model in an alternative embodiment, Picture 9 Is a schematic diagram of the distribution of the center points of the blood vessel segment in an alternative embodiment, Picture 10 It is a threshold pair based on the number of center points of the peripheral branches on the vessel segment in an alternative embodiment Figure 7 The schematic diagram of the blood vessel tree obtained after the image shown is deleted, Picture 11 Is an optional embodiment based on the vascular caliber parameter threshold pair at the center of the vessel segment Figure 7 The schematic diagram of the blood vessel tree obtained after the image shown is deleted, Picture 12 It is a schematic diagram of using the Strahler classification method to classify the portal vein tree in an optional embodiment, Figure 13a A schematic diagram of an operation interface for user setting richness in an optional embodiment, Figure 13b A schematic diagram of an operation interface for user setting richness in an optional embodiment, Figure 14a A schematic diagram of an operation interface for a user to set richness based on image information and parameter information in an optional embodiment, Figure 14b A schematic diagram of an operation interface for a user to set richness based on image information and parameter information in an optional embodiment.

[0143] The following is a detailed description of the processing operations such as segmentation, noise reduction and display of the liver blood vessel image in conjunction with the accompanying drawings, that is, Figure 6-12 As shown, a processing method for the vascular tree in the vascular image can be applied to Figure 7 Performing processing operations such as segmentation and display on the blood vessel tree 11 in the liver blood vessel image may specifically include the following steps:

[0144] Step S41: Obtain image information of the blood vessel image and parameter information of each blood vessel segment on the blood vessel tree.

[0145] Specific, such as Figure 7~8 As shown, the image information of the blood vessel image and the parameter information of each blood vessel segment on the blood vessel tree 11 can be extracted based on the blood vessel tracking model. Picture 8 The blood vessel segment model shown in 21 center point x 0 , Obtain the extension direction v of the center line, and calculate the radius r and other information data. The above-mentioned image information may include at least one parameter information among the contrast parameter of the blood vessel segment, the matching degree parameter of the blood vessel segment, and the gray average value of the surrounding area of ​​the blood vessel segment, and the above-mentioned parameter information may include the attribute parameter of the blood vessel segment, the current blood vessel segment and At least one of the angle parameter between the parent branch vessel segments, the center position parameter of the blood vessel segment, the extension direction parameter of the blood vessel segment, the blood vessel caliber parameter at the center of the blood vessel segment and the maximum blood vessel caliber parameter of the blood vessel segment, and the above parameters The blood vessel segment attribute parameters include root main branches, intermediate branches and peripheral branches, and the parameter information may also include the number of center points of the peripheral branches of the current blood vessel segment.

[0146] It should be noted that, in the embodiments of the present application, it is not limited to the method of using a blood vessel tracking model to obtain the information of each blood vessel segment. For example, the central line segment can be extracted by performing morphological skeletonization on the binary mask image of the segmentation result of the blood vessel tree, and then by calculating the extension direction of the central line segment and the radius and length of the blood vessel segment at the location of the center point of the center line And so on to obtain the above geometric parameters. At the same time, the distance field can also be obtained through the distance transformation method or rapid advance, and the center line is extracted based on the distance field, and then the radius of the blood vessel segment at the position of the center point of the center line segment is calculated by calculating the extension direction of the center line segment , Length, etc., to obtain the above geometric parameters.

[0147] In addition, Figure 7 After the vascular tree 11 on the middle blood vessel image is segmented, the parameter information and image information obtained above can be stored so as to be recalled during subsequent operations such as noise reduction and interactive display. At the same time, based on the above parameters of the center position of the vessel segment, Picture 9 The topological structure 31 of the vascular tree is shown.

[0148] Step S42, preset a first noise reduction rule.

[0149] Specifically, the preset first noise reduction rule may include setting a first parameter threshold and setting a second parameter threshold, that is, based on specific noise reduction requirements, setting a first threshold corresponding to at least one parameter in the parameter information, and The setting includes a second threshold corresponding to at least one of a parameter included in the parameter information and a parameter included in the image information.

[0150] For example, the aforementioned first parameter threshold may include a blood vessel segment attribute parameter threshold, an angle parameter threshold between the current blood vessel segment and the parent branch blood vessel segment, a blood vessel segment center position parameter threshold, and a peripheral branch center point number threshold. The second parameter threshold can include the blood vessel segment attribute parameter threshold, the angle parameter threshold between the current blood vessel segment and the parent branch blood vessel segment, the blood vessel center position parameter threshold, the center point number threshold of the peripheral branch, the blood vessel segment contrast parameter threshold, and the blood vessel segment. The segment matching degree parameter threshold and the gray average threshold of the area around the blood vessel segment, etc. In addition, the first parameter threshold and the second parameter threshold may include a single type of parameter threshold, or may include multiple types of parameter thresholds, and the first parameter threshold and the second parameter threshold include parameter types and/or parameter thresholds The size of is different, that is, the first parameter threshold is different from the second parameter threshold.

[0151] Step S43, according to the first parameter threshold, remove the terminal branches on the vascular tree.

[0152] Specifically, any blood vessel segment whose corresponding parameter value is less than or equal to the first parameter threshold can be deleted to remove the terminal spurs and/or short branches of the blood vessel tree, thereby realizing the noise reduction processing of the blood vessel tree. Wherein, the aforementioned first parameter threshold may also include multiple, and different first parameter thresholds include parameter types and/or parameter thresholds with different sizes, so that the multiple first parameter thresholds are used to gradually delete the vascular tree Defects such as spurs and short branches on the top.

[0153] For example, the first parameter threshold includes the threshold of the number of central points of the peripheral branch, and the threshold of the number of central points of the peripheral branch is 3. Based on the geometric parameters of each blood vessel segment obtained above, in the subsequent noise reduction processing, the number of central points of the peripheral branch is less than A vessel segment equal to 3 (e.g. Picture 9 The branch shown in 32) is deleted.

[0154] For another example, the first parameter threshold includes a blood vessel caliber parameter threshold at the center of the blood vessel segment, and the blood vessel caliber parameter threshold at the center of the blood vessel segment may be a radius threshold or a diameter threshold. Assuming that the first parameter threshold is the radius threshold at the center of the blood vessel segment, and the radius threshold at the center of the blood vessel segment is 0.5mm, the radius parameter at the center of the blood vessel segment can be less than or equal to 0.5 during the noise reduction process. The blood vessel segment of mm can be deleted, or the blood vessel segment whose maximum radius parameter is less than or equal to 0.5mm in the blood vessel segment can be deleted.

[0155] For another example, the first parameter threshold may also include the angle parameter threshold between the current blood vessel segment and the parent branch blood vessel segment, and the angle parameter threshold between the current blood vessel segment and the parent branch blood vessel segment is 60°, then it is decreasing. During the noise processing, the blood vessel segments whose angle between the current vessel segment and the parent vessel segment is less than or equal to 60° can be deleted.

[0156] In addition, in order to further improve the accuracy of noise reduction, the above-mentioned deletion operation based on the first parameter threshold can be performed only on the blood vessel segment whose attribute parameter of the blood vessel segment is the middle branch and/or the peripheral branch, so as to ensure that the deleted blood vessel segments are all blood vessels. The sub-branch in the tree.

[0157] In an optional embodiment, based on one or more first parameter thresholds similar to the above-mentioned deletion operation, after the above-mentioned deletion of the blood vessel segment, noise reduction processing is performed on other blood vessel trees, or by changing the first parameter threshold. The size and/or type of each parameter threshold in the parameter threshold can realize the noise reduction processing according to the corresponding requirements of different vascular trees, or the operation of deleting sub-branch iteratively for the same vascular tree multiple times.

[0158] For example, first use the first parameter threshold including the threshold of the number of center points of the peripheral branch to be 3, Figure 7 The vascular tree 11 shown performs the first deletion operation, and then obtains Picture 10 The vascular tree 41 shown; then, using another first parameter threshold with a radius threshold of 0.5 mm at the center of the vessel segment, right Picture 10 The vascular tree 41 shown performs the second deletion operation, and then obtains Picture 11 The blood vessel tree 51 shown can be deleted by sequentially deleting Picture 11 The vascular tree 51 shown is compared to the obtained Picture 10 The illustrated vascular tree 41 has fewer defects such as peripheral spurs and short branches.

[0159] Step S44, according to the second parameter threshold, remove part of the blood vessel segment.

[0160] Specifically, for the blood vessel segment whose attribute parameter of the blood vessel segment is the middle branch, the blood vessel segment whose value of any corresponding parameter is less than or equal to the second parameter threshold can be deleted to remove part of the middle blood vessel segment; because the second parameter threshold is combined Since the parameter information and image information are included, the effectiveness and accuracy of noise reduction can be further improved on the basis of step S43. Wherein, the noise reduction processing operation on the blood vessel tree may include step S43 and/or step S44.

[0161] Step S45: Obtain the first classification result of each blood vessel segment of the blood vessel tree after noise reduction processing.

[0162] Specifically, a classification operation may be performed on the blood vessel tree after the noise reduction process to obtain the above-mentioned first classification result. In addition, before performing step S43, or between step S43 and step S44, the first grading result can be obtained by performing a grading operation on each blood vessel segment and obtaining the grading information of each blood vessel segment; The above-mentioned step S43 and/or step S44 is performed based on the classification information. At the same time, the deletion operation of step S43 can also be performed based on the classification information.

[0163] For example, by marking the progression from the peripheral branch and backtracking, the parent branch can be determined at the bifurcation according to the progression of the child branch (such as the Strahler classification) to classify each blood vessel segment. Such as Picture 12 As shown in the portal vein tree 61; or, start from the root and mark the progression along the extension direction of the vascular tree. Each time a bifurcation is encountered, a higher progression is marked (for example, the main branch of the hepatic portal vein is marked as the first At the first level, the left main branch and the right main branch are marked as the second level, and so on, until the peripheral branch is marked) to perform grading operations on each blood vessel segment. Then, obtain and perform the subsequent deletion operation of the blood vessel segment according to the classification information of each blood vessel segment.

[0164] Further, the blood vessel segment can be deleted in sequence according to the level of the blood vessel segment; for example, for the classification information obtained by the classification method of marking the level from the root, each blood vessel can be deleted in order from largest to smallest. Segments; and for the grading information obtained by the grading method of marking the number of stages starting from the peripheral branch, each blood vessel segment can be deleted in order from small to large.

[0165] In addition, for blood vessel segments of the same grade, sorting can be performed according to their parameter information or other set rules to improve the accuracy of subsequent deletion of blood vessel segments.

[0166] For example, all blood vessel segments can be sorted according to the size of the hierarchical level. For example, the high-level blood vessel segment can be ranked before the low-level blood vessel segment. For the blood vessel segment with parent-child branch relationship, the child branch can be ranked in the parent branch. Before; At the same time, the blood vessel segments of the same level can also be arranged in the order of radius from small to large, length from short to long, and blood vessel image brightness from dark to bright; subsequent noise reduction processing and display can be based on the above The sequence of adding or deleting blood vessel segments is performed in sequence.

[0167] Step S46, based on the first classification result, perform a hierarchical display of each blood vessel segment on the blood vessel tree after the noise reduction process.

[0168] Specifically, the blood vessel segment included in the blood vessel tree can be adjusted according to the blood vessel segment display classification level and the first classification result, and the blood vessel tree after the adjustment operation can be displayed in a hierarchical manner. If one or more third parameter thresholds can be preset based on specific application requirements, the third parameter thresholds can include the threshold corresponding to at least one parameter in the aforementioned parameter information and/or image information and the threshold corresponding to the branch level parameter; , By displaying the blood vessel segments whose values ​​of the corresponding parameters are all greater than the thresholds of the third parameters step by step; wherein the thresholds included in the multiple third parameter thresholds are of the same type and different values. In addition, the corresponding third threshold parameter can also be set based on the display richness, and the type of parameter in the third threshold parameter and the size of the parameter threshold can also be adjusted in real time through the input device.

[0169] It should be noted that in the embodiments of the present application, one or more parameter thresholds can be set based on specific requirements, and the parameter types and parameter values ​​of different parameter thresholds are different, so as to perform at least one denoising of the vascular tree. Processing can also achieve multiple iterations of noise reduction processing on the vascular tree, so as to accurately and effectively remove various noise defects on the vascular tree. At the same time, the above-mentioned vascular segment display hierarchical level can be the acquired level number set by the user according to the application scenario, so that the user can select the blood vessel extraction result to be displayed by himself, and realize the interactive display of the vascular tree.

[0170] Step S47: Sort the first classification result to obtain a second classification result.

[0171] Specifically, in order to increase the richness of the display of the blood vessel tree, the first classification result can be ordered based on the obtained image information and/or parameter information according to preset rules, so as to obtain the second classification result. For example, you can combine the results of the above noise reduction Figure 7 The blood vessel segment contained in the blood vessel tree 11 is set to the first level of richness, and the obtained in step S43 Picture 10 The blood vessel segment contained in the blood vessel tree 41 shown in the figure is set to the second level of richness, and the obtained in step S44 Picture 11 The blood vessel segments contained in the blood vessel tree 51 shown in the figure are set to the third level of richness; then, based on the above-mentioned first grading result, and can be combined with parameter information and/or image information, in descending order of the displayed number of blood vessel segments , Set the fourth-level richness, fifth-level richness...Nth-level richness in sequence (N is a positive integer greater than or equal to four), and then obtain the above-mentioned second classification result.

[0172] Step S48: Obtain the completeness of the blood vessel tree display set by the user, and selectively display the blood vessel segments included in the blood vessel tree based on the second classification result.

[0173] Specific, such as Figure 13a-13b As shown, based on the second classification result, a limited number of richness adjustment gears can be set on the operation interface, and the user can touch Figure 13a High and low adjustment buttons shown or move Figure 13b The richness adjustment bar shown to set the required richness, you can use the Figure 13a with Figure 13b The operation interface to obtain the richness set by the user (that is, the completeness of the blood vessel tree display), that is, based on the obtained richness set by the user and the above-mentioned second classification result, the blood vessel tree containing the corresponding blood vessel segment is displayed , So that the user can select the desired blood vessel extraction results by himself, and realize the adjustment of interactive richness.

[0174] In another alternative embodiment, such as Figure 14a-14b As shown, the user can also based on image information and parameter information, the user can touch Figure 14a High and low adjustment buttons shown or move Figure 14b The richness adjustment bar shown to accurately set the required displayed blood vessel segment can be used alone as Figure 14a with Figure 14b Interface, or combined with Figure 13a with Figure 13b The user interface is used to obtain the more precise display requirements set by the user, so that the user can select the precise blood vessel extraction results required by himself, and realize the precise adjustment of the interactive richness.

[0175] Figure 15 It is a schematic diagram of the module structure of an apparatus for processing a blood vessel tree in a blood vessel image in an optional embodiment. Such as Figure 15 As shown, the embodiment of the present application also provides a processing device for a vascular tree in a vascular image, which may include a medical imaging device body 71, a processing device 72, and a computer-readable storage medium 73. The medical imaging device body 71 can be used to collect The blood vessel image of the target organ, and the blood vessel image has the blood vessel tree of the above-mentioned target organ; the computer-readable storage medium 73 can store a computer program; the processing device 72 can be respectively connected with the medical imaging equipment body 71 and the computer-readable storage medium The 73 connection is used to retrieve and execute the computer program to implement the steps of the blood vessel image processing method described in any embodiment of the present application based on the aforementioned blood vessel image and the blood vessel tree included therein.

[0176] Figure 16 It is a schematic diagram of the module structure of a device for processing a blood vessel tree in a blood vessel image in another optional embodiment. Such as Figure 16 As shown, in an optional embodiment, the present application also provides another device for processing a vascular tree in a blood vessel image. The processing device may include an information acquisition device 81, a denoising device 82, a display device 83, etc., which are sequentially connected. And the information acquisition device 81 can be used to acquire the parameter information of each blood vessel segment on the vascular tree, the image information of the blood vessel image and the classification information of each blood vessel segment, etc.; the denoising device 82 can be used to obtain the parameter information and information based on the information obtained by the information acquisition device. / Or image information performs denoising processing on the above-mentioned vascular tree; and the display device 83 can be used to perform denoising processing on the vascular tree based on the above-mentioned parameter information and/or image information. The included vessel segments are displayed hierarchically or selectively; among them, the parameter information can include the attributes of the vessel segment, the angle between the current vessel segment and the parent vessel segment, the center position of the vessel segment, the extension direction of the vessel segment, and the center of the vessel segment The diameter of the blood vessel at the location and the number of center points of the peripheral branches of the current blood vessel segment. The attributes of the blood vessel segment can include the root main branch, the middle branch and the peripheral branch; the image information can include the contrast of the blood vessel segment, the matching degree of the blood vessel segment and the area around the blood vessel segment. Gray value, etc.

[0177] Figure 17 It is a schematic diagram of the module structure of the medical imaging equipment in an optional embodiment. Such as Figure 17 As shown, the embodiment of the present application also provides a medical imaging device, which may include a medical imaging device body 91 and a processor 92 that are connected to each other; the medical imaging device body 91 may be used to collect a blood vessel image of a target organ, and the blood vessel image includes There is a vascular tree with the above target organ, and the processor 92 can be the vascular image processing device described in any of the above embodiments, which is used to denoise the vascular tree or perform processing on each vascular segment contained in the vascular tree. Interactive outline display.

[0178] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, All should be considered as the scope of this specification.

[0179] The above-mentioned embodiments only express several embodiments of the present invention, and the descriptions are more specific and detailed, but they should not be understood as limiting the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.