Navigation path planning method, system and readable storage medium

Abstract

The present invention provides a navigation path planning method, system and readable storage medium, the method comprising: acquiring a target organ volume data image; segmenting the natural cavity in the target organ volume data image to obtain a natural cavity region volume data image; performing 3D distance transformation on each natural cavity region pixel point in the natural cavity region volume data image to obtain natural cavity distance transformation volume data; according to the natural cavity distance transformation volume data, based on minimization of a cost function, iterating on the natural cavity region volume data image, searching for a target path from a starting point to a target point, and obtaining a planned navigation path. The present invention uses a minimum optimization iteration method based on a cost function to search for the best path from a starting point to a target point, obtains the best navigation path based on the center line of the natural cavity, and improves the accuracy of navigation path planning.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a preoperative navigation path planning method, system and readable storage medium. Background Art

[0002] The diagnosis of lung lesions is becoming increasingly important for the early detection and treatment of lung cancer. The conventional lung lesion biopsy method is to insert the puncture needle directly into the lungs to obtain the biopsy sample by percutaneous puncture. This puncture method has the risk of pneumothorax. Studies have shown that percutaneous lung puncture has a 20-40% incidence of pneumothorax, which is a serious adverse reaction that can be fatal.

[0003] A safer way is to use a bronchoscope to biopsy lung lesions through the natural airway, which can reduce the risk of adverse reactions. However, the bronchial tree is relatively complex and requires experienced doctors to operate, and the skills required of bronchoscope operators are relatively strict. Therefore, it is important to formulate the best path from the main airway to the target lesion before surgery and guide the doctor to operate the endoscope to the target lesion during surgery.

[0004] Therefore, it is necessary to provide a method that can accurately plan the navigation path of a pulmonary bronchoscope to better assist doctors and improve diagnostic accuracy. Summary of the invention

[0005] The object of the present invention is to provide a navigation path planning method, system and readable storage medium to improve the accuracy of natural cavity navigation path planning in an organ.

[0006] To achieve the above object, the present invention provides a navigation path planning method, comprising:

[0007] Acquiring a target organ volume data image, wherein the target organ has a natural cavity;

[0008] Segmenting the natural cavity in the target organ volume data image to obtain the natural cavity region volume data image;

[0009] Performing 3D distance transformation on each natural cavity region pixel point in the natural cavity region volume data image to obtain natural cavity region distance transformation volume data;

[0010] According to the natural cavity distance transformation volume data, based on minimization of the cost function, iterate on the natural cavity area volume data image, search for a target path from a starting point to a target point, and obtain a planned navigation path;

[0011] The starting point is a pixel point at the starting position of the natural cavity, and the target point is a pixel point in the natural cavity region adjacent to the lesion position in the volume data image of the natural cavity region.

[0012] Optionally, in the navigation path planning method, the method further includes: correcting the target path in the following manner:

[0013] For each path pixel point of a single cavity located in the natural cavity in the target path, obtain a connected domain corresponding to a slice of the path pixel point on the natural cavity distance transformation volume data, find a peak pixel point corresponding to a peak value in the connected domain, and use the peak pixel point as a correction point for the path pixel point;

[0014] For each path pixel point in the target path located near the bifurcation position in the natural cavity, a connected domain corresponding to the slice of the path pixel point on the natural cavity distance transformation volume data is obtained, and the connected domain is filtered to remove abnormal pixels, and then peak pixel points corresponding to at least two peaks in the connected domain are found, and the target peak pixel point is used as the correction point of the path pixel point; wherein the target peak pixel point is the peak pixel point with the smallest distance from the path pixel point.

[0015] Optionally, the navigation path planning method further includes:

[0016] For each path pixel point of a single cavity in the target path located in the natural cavity, after obtaining a connected domain corresponding to a slice of the path pixel point on the natural cavity distance transformation volume data, the area of ​​the connected domain is calculated to obtain cross-sectional feature information of the path pixel point;

[0017] For each path pixel point in the target path located near the bifurcation position in the natural cavity, after determining the target peak pixel point, the target peak pixel point is used as a seed point for watershed segmentation to obtain the cavity area where the target peak pixel point is located, and the area of ​​the cavity area is calculated to obtain the cross-sectional feature information of the path pixel point.

[0018] Optionally, in the navigation path planning method, the iterative process based on minimizing the cost function on the volume data image of the natural cavity region to search for a target path from a starting point to a target point includes:

[0019] Step A1: providing a first set and a second set, wherein the first set includes the starting point, and the second set is empty;

[0020] Step B1: Select the pixel point with the smallest cost function value in the first set as the current point, put the current point into the second set, and process each pixel point in the neighborhood around the current point in the natural cavity area volume data image as follows:

[0021] a. If the pixel is in the second set, no processing is performed;

[0022] b. If the pixel point is not in the first set, add the pixel point to the first set, calculate the cost function value of the pixel point, and set the current point as the parent node of the pixel point;

[0023] c. If the pixel point is in the first set, recalculate the cost function value of the pixel point; if the recalculated cost function value is less than the cost function value of the pixel point calculated last time, modify the parent node of the pixel point to the current point;

[0024] Step C1: determining whether the first set is empty, and if not empty, returning to execute step B1 until the target point exists in the first set;

[0025] Step D1, backtracking the target point, searching for the parent node and recording it until the starting point is found, and the recorded series of pixel points constitute the target path from the starting point to the target point.

[0026] Optionally, in the navigation path planning method, performing 3D distance transformation on each natural cavity region volume pixel point in the natural cavity region volume data image includes:

[0027] For each natural cavity region pixel point in the natural cavity region volume data image, the surrounding neighborhood is traversed from near to far to find the nearest non-natural cavity region pixel point, and the straight-line distance between the natural cavity region volume pixel point and the non-natural cavity region pixel point is calculated as the pixel value corresponding to the natural cavity region pixel point.

[0028] Optionally, in the navigation path planning method, segmenting the natural cavity in the target organ volume data image to obtain the natural cavity region volume data image includes:

[0029] Detecting the starting position of the cavity in the target organ volume data image and determining a regional growth seed point;

[0030] The natural cavity is segmented by region growing to obtain a volume data image of the natural cavity region.

[0031] Optionally, in the navigation path planning method, detecting the starting position of the cavity in the target organ volume data image and determining the region growth seed point includes:

[0032] Step A2, taking a layer of two-dimensional image located at the upper end of the target organ volume data image as an initial image;

[0033] Step B2, binarizing the initial image, and calculating the areas and ellipticity of different connected domains in the initial image, and taking the connected domains whose areas and ellipticity meet preset conditions as potential cavity areas;

[0034] Step C2, respectively calculating N layers of two-dimensional images adjacent to the initial image to obtain N potential cavity regions, where N≥3;

[0035] Step D2, respectively calculate the centroid coordinates of the N+1 potential cavity areas. If the deviations between two adjacent centroid coordinates are less than a preset threshold, the centroid of the initial image is determined as the region growth seed point. Otherwise, the next layer of the two-dimensional image of the initial image is used as the initial image and return to execute step B2.

[0036] Optionally, in the navigation path planning method, before performing 3D distance transformation on each natural cavity region volume pixel point in the natural cavity region volume data image, the method further includes:

[0037] The holes inside the natural cavity in the natural cavity region volume data image are filled using morphological operations.

[0038] Optionally, in the navigation path planning method, before segmenting the natural cavity in the target organ volume data image, the method further includes:

[0039] The target organ volume data image is subjected to edge enhancement processing.

[0040] Optionally, in the navigation path planning method, the cost function is:

[0041] F(n)=w1*G(n)+w2*H(n),

[0042] G(n)=(P start -P cur ) 2 ,

[0043] H(n)=(P cur -P goal ) 2

[0044] w1=β*e ΔVolume_Dist

[0045] w2=σlog abs(Z_cur-Z_goal)

[0046] Where F(n) represents the estimated cost of the starting point to reach the target point via the current point n, G(n) represents the actual cost of the path from the starting point to the current point n, H(n) represents the estimated cost of the best path from the current point n to the target point, and P start represents the position coordinates of the starting point, P cur Represents the position coordinates of the current point, P goal represents the position coordinates of the target point, w1 and w2 represent weight values, Z_cur represents the z coordinate in the position coordinates of the current point, Z_goal represents the z coordinate in the position coordinates of the target point, and ΔVolume_Dist represents the absolute value of the difference between the maximum value in the natural cavity distance transformation volume data and the data value corresponding to the current point.

[0047] Optionally, in the navigation path planning method, it is characterized in that the target organ is the lung.

[0048] To achieve the above-mentioned purpose, the present invention also provides a navigation path planning system, including a processor and a memory, wherein the memory stores instructions, and when the instructions are executed by the processor, the navigation path planning method described above is implemented.

[0049] To achieve the above-mentioned object, the present invention further provides a readable storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by a processor, the navigation path planning method described above is implemented.

[0050] Compared with the prior art, the navigation path planning method, system and storage medium provided by the present invention have the following advantages: after acquiring the target organ volume data image, the present invention first segments the natural cavity in the target organ volume data image to obtain the natural cavity area volume data image, and performs 3D distance transformation on each natural cavity area pixel point in the natural cavity area volume data image to obtain the natural cavity distance transformation volume data; then, according to the natural cavity distance transformation volume data, based on the minimization of the cost function, iterates on the natural cavity area volume data image to search for the target path from the starting point to the target point, and obtains the planned navigation path. The present invention uses the method of minimum optimization iteration based on the cost function to search for the target path from the starting point to the target point, and obtains the target navigation path based on the cavity centerline of the target organ, which improves the accuracy of the natural cavity navigation path planning in the target organ; further, it can also provide cross-sectional feature information on the navigation path, providing more comprehensive path planning information, which can better assist doctors in improving the accuracy of diagnosis. In addition, the present invention can automatically perform navigation path planning, reduce the tedious operation of human-computer interaction, and improve the efficiency of diagnosis. In addition, the navigation path planning algorithm of the present invention has strong versatility, realizes an end-to-end algorithm flow, and further improves the diagnostic efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 is a flow chart of a navigation path planning method in one embodiment of the present invention;

[0053] Figure 2a It is a schematic diagram of a slice image of the main airway in a lung volume data image;

[0054] Figure 2b An example of a segmented three-dimensional bronchial tree;

[0055] Figure 3a This is a schematic diagram of the bifurcated road area;

[0056] Figure 3b This is a schematic diagram of the bifurcated road area after watershed segmentation;

[0057] Figure 4 This is a schematic diagram of navigation path planning based on the bronchial centerline;

[0058] Figure 5It is a structural schematic diagram of a navigation path planning system in one embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following is a further detailed description of the navigation path planning method, system and storage medium proposed in the present invention in conjunction with the accompanying drawings and specific embodiments. According to the following description, the advantages and features of the present invention will be clearer. It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In order to make the purposes, features and advantages of the present invention more obvious and easy to understand, please refer to the accompanying drawings. It should be noted that the structures, proportions, sizes, etc. illustrated in the accompanying drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention, so they have no technical substantive significance. Any structural modification, change in proportional relationship or adjustment of size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope of the technical content disclosed by the present invention.

[0060] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0061] The core idea of ​​the present invention is to provide a navigation path planning method, system and storage medium to improve the accuracy of natural cavity navigation path planning in organs.

[0062] It should be noted that the navigation path planning method of the embodiment of the present invention can be applied to the navigation path planning system of the embodiment of the present invention, which can be a personal computer, a mobile terminal, etc., and the mobile terminal can be a mobile phone, a tablet computer, or other hardware device with various operating systems.

[0063] The following is a detailed description of a navigation path planning method provided by the present invention, taking the target organ as the lung organ as an example, wherein the pulmonary bronchial tree is a natural cavity in the pulmonary organ. It can be understood by those skilled in the art that, in addition to the navigation path planning of the pulmonary bronchial tree, the present invention can also be used for the navigation path planning of natural cavities in other organs, such as natural cavities such as the intestine.

[0064] Please refer to Figure 1 When the present invention is applied to the navigation path planning of the pulmonary bronchial tree, it specifically includes the following steps:

[0065] Step S100: Acquire a lung volume data image.

[0066] In this embodiment, the lung volume data image may be a computed tomography (CT) volume data image of a patient, or may be other types of medical images. It should be noted that the size of the lung volume data image may be set according to specific circumstances, and the present invention does not limit this. For example, the three-dimensional data size of the lung volume data image may be 512×512×327.

[0067] Step S200: segmenting the bronchial tree from the lung volume data image to obtain a bronchial region volume data image.

[0068] The bronchial region volume data image is the natural cavity region volume data image of the lung. In this embodiment, the region growing method can be used to realize the three-dimensional bronchial tree segmentation. The region growing method is to combine pixels with similar properties in the image to form a region. Specifically, a seed pixel is first found as the starting point of growth for the area to be segmented, and then the pixels in the neighborhood around the seed pixel that have the same or similar properties as the seed pixel (the same or similar properties can be determined according to a predetermined growth or similarity criterion) are merged into the area where the seed pixel is located. These new pixels are treated as new seed pixels and the above process is continued until no pixels that meet the conditions can be included, and then a region is grown.

[0069] Preferably, before executing step S200, the method further includes: performing edge enhancement processing on the lung volume data image to enhance the bronchial wall and improve the image quality of the organ image of the lung volume data image. Specifically, Laplace sharpening processing can be used to enhance the image edge. The basic idea of ​​Laplace sharpening processing is: when the grayscale of the central pixel of the neighborhood is lower than the average grayscale of other pixels in its neighborhood, this central pixel should be further reduced; when the grayscale of the central pixel in the neighborhood is higher than the average grayscale of other pixels in its neighborhood, this central pixel should be further increased. By convolving the lung volume data image with a Laplace second-order template operator, image enhancement is achieved to improve the edges and details of the image. In other embodiments, other image enhancement methods can also be used.

[0070] Preferably, the main airway (the main airway is the starting position of the natural cavity of the lung) can be detected in the lung volume data image first, the regional growth seed point can be determined, and then the bronchial tree can be segmented by regional growth to obtain the bronchial regional volume data image.

[0071] Specifically, the main airway may be detected in the lung volume data image and the region growth seed point may be determined by the following iterative loop:

[0072] Step A2, taking a layer of two-dimensional image located at the upper end of the lung volume data image as an initial image;

[0073] Step B2, binarizing the initial image, and calculating the areas and ellipticity of different connected domains in the initial image, and taking the connected domains whose areas and ellipticity meet preset conditions as potential main airway regions;

[0074] Step C2, respectively calculating N layers of two-dimensional images adjacent to the initial image to obtain N potential main airway regions, where N≥3;

[0075] Step D2, respectively calculate the centroid coordinates of the N+1 potential main airway areas. If the deviation between two adjacent centroid coordinates is less than a preset threshold, the centroid of the initial image is determined as the region growth seed point. Otherwise, the next layer of two-dimensional image of the initial image is used as the initial image and return to execute step B2.

[0076] Figure 2a A slice image of the main airway region of the lung is shown. Figure 2aThe small black circular area in the middle is the cross section of the main bronchus, and the larger black areas on both sides are the cross sections of the lung areas. Specifically, since the main airway is usually located at the upper end of the lung volume data image, the main airway can be detected starting from a certain axial slice image located at the upper end of the lung volume data image, such as the 10th axial slice image, and the slice image is used as the initial image. Preferably, in the lung volume data image, the main airway is usually located in the middle area of ​​the slice image, so a fixed search area can be defined in the initial image, for example, a certain distance (e.g., 50 pixels) is translated upward from the center of the initial image as the center of the search area, the length and width of the search area can be set to 50 pixels, and a certain threshold (e.g., -800HU) is used to binarize the search area. The area and ellipticity of different connected domains in the search area are calculated respectively (for binary images, the area is the number of pixels in the connected domain, and the ellipticity is the aspect ratio of the minimum circumscribed rectangle). When the area and ellipticity of a connected domain meet the preset conditions, for example, the second preset condition can be: the area is within the range of [150, 450] and the ellipticity is greater than 0.8, then the connected domain is considered to be a potential main airway area. Then, the three layers of slice images adjacent to the initial image are calculated respectively, and a total of four potential areas are obtained. The centroid coordinates of each potential area are calculated respectively. When the deviation between the centroid coordinates of the potential areas in two adjacent slice images is less than the preset threshold (for example, 3 pixels), it means that the centroid in the initial image is the center of the tracheal interface, which can be used as the seed point for regional growth; otherwise, the potential area is left blank, and the next layer of slice image of the current initial image is used as the initial image, and it is iterated again until the regional growth seed point is found.

[0077] When performing regional growth to obtain a three-dimensional bronchial tree, an initial threshold value T1 (for example, -915HU) can be set, and a 3D 26-neighborhood can be used to judge the adjacent pixels around the regional growth seed point: if the pixel value of the adjacent pixel T < T1, it is considered that the adjacent pixel is located in the bronchial region, and the pixel value of the pixel is set to 1; if the pixel value of the adjacent pixel T > T1, it is considered that the adjacent pixel is located in the non-bronchial region, and the pixel value of the adjacent pixel is set to 0. In this way, by performing a loop iteration in the lung volume data image, the bronchial region can be extracted from the lung volume data image, and finally a bronchial region volume data image is obtained, that is, the following is obtained. Figure 2b The three-dimensional bronchial tree is shown.

[0078] Step S300: performing 3D distance transformation on each bronchial region pixel point in the bronchial region volume data image to obtain bronchial distance transformation volume data.

[0079] It should be noted that step S300 performs 3D distance transformation on each pixel point of the bronchial area in the bronchial area volume data image to obtain bronchial distance transformation volume data. It has two purposes: one is to calculate the parameter value of the weight value w2 of the cost function in step S400; the other is to correct the centerline of the target path when some path points in the target path searched in step S400 are not on the centerline of the cavity.

[0080] Among them, the process of 3D distance transformation is specifically as follows: for each bronchial area pixel point in the bronchial area volume data image, traverse the surrounding neighborhood from near to far to find the nearest non-bronchial area pixel point, and calculate the straight-line distance between the bronchial area pixel point and the non-bronchial area pixel point as the pixel value corresponding to the bronchial area pixel point.

[0081] As mentioned above, in the bronchial region volume data image obtained in step S200, the pixel values ​​of the bronchial region are all 1, and the pixel values ​​of the non-bronchial region are 0. Therefore, for all the pixels in the bronchial region, for each pixel point, the pixel points in its surrounding neighborhood are traversed from near to far, the nearest pixel point with a pixel value of 0 is found, and the straight-line distance between the two is calculated as the pixel value corresponding to the pixel point in the bronchial region. By traversing all the pixels in the bronchial region in a loop in the above manner, the pixel values ​​corresponding to all the pixels in the bronchial region can be finally obtained, which is the bronchial distance transformation volume data Volume_Dist. In the bronchial distance transformation volume data Volume_Dist, the closer the pixel point is to the boundary of the bronchial region, the smaller its data value (i.e., pixel value), the closer the pixel point is to the inside of the bronchial cavity, the larger its data value (i.e., pixel value), and the data values ​​(i.e., pixel values) of the pixel points on both sides of the pixel point on the center line of the bronchial tree (also called the bronchial tree skeleton) are approximately symmetrical about the skeleton.

[0082] Preferably, before performing the 3D distance transformation, the holes inside the bronchial tree in the bronchial region volume data image may be filled using a morphological operation. Since there may be holes inside the segmented bronchial tree, that is, there are pixels with a pixel value of 0 inside the bronchial tree, they need to be filled, specifically by using the hole filling operation in the morphological operation to change the pixel values ​​of these pixels from 0 to 1.

[0083] Step S400: According to the bronchial distance transformation volume data, based on minimization of the cost function, iterate on the bronchial region volume data image to search for a target path from the starting point to the target point, and obtain a planned navigation path.

[0084] The starting point is a pixel point at the main airway position in the bronchial tree, and the target point is a pixel point in the bronchial region adjacent to the lesion position in the bronchial region volume data image.

[0085] Specifically, in this embodiment, in step S200, the regional growth seed point determined by detecting the main airway in the lung volume data image can be used as the starting point. It can be understood that the starting point of the navigation path is usually set at the main airway position of the bronchial tree, so in other embodiments, other pixel points at the main airway position in the bronchial region volume data image can also be determined as the starting point, and the present invention does not limit this.

[0086] In this embodiment, the coordinates [x, y, z] of the location of the external bronchial lesion can be manually input by the doctor, where z is the number of slice layers in the lung volume data image. According to the coordinates of the lesion location, the bronchial region pixel point adjacent to the lesion location in the bronchial region volume data image can be determined, and the pixel point is the target point. For example, the pixel point can be the bronchial region pixel point closest to the lesion location in the bronchial region volume data image. The specific determination method is as follows: intercept all bronchial region pixel points in the bronchial region with a slice number in the range of [z-30, z+30], calculate the Euclidean distance between the lesion location and each bronchial region pixel point, and use the bronchial region pixel point corresponding to the minimum distance as the navigation target point, and also as a potential point for puncture during surgery.

[0087] Before searching, the cost function is constructed based on the bronchial distance transformation volume data as follows:

[0088] F(n)=w1*G(n)+w2*H(n)

[0089] G(n)=(P start -P cur ) 2

[0090] H(n)=(P cur -P goal ) 2

[0091] w1=β*e ΔVolume_Dist

[0092] w2=σlog abs(Z_cur-Z_goal)

[0093] Where F(n) represents the estimated cost of the starting point to reach the target point through the current point n (also called state n), G(n) represents the actual cost of the path from the starting point to the current point n (state n), H(n) represents the estimated cost of the best path from the current point n (state n) to the target point, Pstart represents the position coordinates of the starting point, P cur Represents the position coordinates of the current point, P goal represents the position coordinates of the target point, w1 and w2 represent weight values, which are respectively used to adjust the influence of the actual cost G(n) and the estimated cost H(n) on the overall loss cost F(n), Z_cur represents the z coordinate in the position coordinates of the current point, Z_goal represents the z coordinate in the position coordinates of the target point, ΔVolume_Dist represents the absolute value of the difference between the maximum value in the bronchial distance transformation volume data and the data value corresponding to the current point.

[0094] In this embodiment, based on minimization of the cost function, iteration is performed on the bronchial region volume data image to search for the best path from the starting point to the target point. The specific iteration process is as follows:

[0095] Step A1: providing a first set and a second set, wherein the first set includes the starting point, and the second set is empty;

[0096] Step B1: Select the pixel point with the smallest cost function value in the first set as the current point, put the current point into the second set, and process each pixel point in the surrounding neighborhood (e.g., the surrounding 26 neighborhoods) of the current point in the bronchial region volume data image as follows:

[0097] a. If the pixel is in the second set, no processing is performed;

[0098] b. If the pixel point is not in the first set, add the pixel point to the first set, calculate the cost function value of the pixel point, and set the current point as the parent node of the pixel point;

[0099] c. If the pixel point is in the first set, the cost function value of the pixel point is recalculated. If the recalculated cost function value is less than the cost function value of the pixel point calculated last time, the parent node of the pixel point is changed to the current point. It can be understood that the cost function value of the pixel point is recalculated.

[0100] Step C1: determining whether the first set is empty, and if not, returning to step B1 until the target point exists in the first set; it can be understood that if the first set is empty, it means that the path does not exist and navigation fails;

[0101] Step D1, backtracking the target point, searching for the parent node and recording it until the starting point is found, and the recorded series of pixel points constitute the best path from the starting point to the target point.

[0102] It can be understood that the starting point is located in the main airway, and the target point is the lesion area, which is usually located in the multi-level bronchial area of ​​the lung. The weight value w2 is dynamically combined with the number of slice layers, and abs(Z_cur-Z_goal) is the absolute value of the z coordinate of the current point from the z coordinate of the target point. The farther the current point is from the target point, the larger the weight value w2. In the weight value w1, ΔVolume_Dist is the absolute value of the difference between the maximum value in the bronchial distance transformation volume data Volume_Dist and the data value corresponding to the current point. If the data value of the current point in the bronchial distance transformation volume data Volume_Dist is larger, the weight value w1 is smaller, so that the iterated path can be made to go as far as possible on the center line of the lung bronchial cavity. The cost function F(n) is dynamically adjusted during the iteration process through the above-mentioned weight values ​​w1 and w2, which is conducive to more efficient search for the optimal path.

[0103] Furthermore, after the target path from the starting point to the target point is obtained through iteration, some path points on the target path may not be located on the center line of the bronchial cavity of the lung. Therefore, the center line of the target path may be corrected so that the corrected path is located on the center line of the bronchial cavity. The specific correction method is as follows:

[0104] For each path pixel point of a single lumen in the target path located in the bronchial tree, a connected domain corresponding to a slice of the path pixel point on the bronchial distance transformation volume data is obtained, a peak pixel point corresponding to a peak value in the connected domain is found, and the peak pixel point is used as a correction point for the path pixel point;

[0105] As can be seen from the above description, the data value corresponding to each bronchial region pixel point in the bronchial distance transformation volume data is the straight-line distance between the bronchial region pixel point and the nearest non-bronchial region pixel point. Therefore, the peak pixel point corresponding to the peak value in the connected domain is the center point of the connected domain. In other words, the peak pixel point is located on the center line of the bronchial cavity. Therefore, by using the peak pixel point as the correction point of the path pixel point, the path pixel point on the target path can be corrected to be located on the center line of the bronchial cavity.

[0106] For each path pixel point in the target path located near the bifurcation position in the bronchial tree, obtain the connected domain corresponding to the slice of the path pixel point on the bronchial distance transformation volume data, filter the connected domain to remove abnormal pixels, and then find the peak pixel points corresponding to at least two peaks in the connected domain, and use the target peak pixel point as the correction point of the path pixel point; wherein the target peak pixel point is the peak pixel point with the smallest distance from the path pixel point.

[0107] It is understood that when the target path passes through the bifurcation of the bronchus, such as Figure 3a As shown in , the connected domain actually includes multiple bronchial cavities, and the peak pixel points corresponding to the peak value in the connected domain also include multiple peak pixel points. Therefore, it is necessary to determine the peak pixel point with the smallest distance from the path pixel point as the correction point, and correct the path pixel point on the target path to be located on the center line of a bifurcated cavity. Please refer to Figure 3a , which schematically shows the connected domain where the path pixel point a is located at a bifurcation position, and the connected domain has two bronchial cavities, that is, includes two peak pixel points b1 and b2. It can be understood that the two peak pixel points are respectively located on the center lines of the two bronchial cavities. Since the distance between the path pixel point a and the peak pixel point b1 is the shortest, it means that the path pixel point a and the peak pixel point b1 are in the same bronchial cavity. Therefore, the peak pixel point b1 is used as the correction point of the path pixel point a.

[0108] Figure 4 The final correction result of the navigation path in the bronchial tree is shown. The gray circle C in the figure represents the lesion location, and the dotted line represents the target path from the starting point to the target point. Figure 4 It can be seen that the corrected target path follows the center line of the cavity from the starting point to the target point, which improves the accuracy of the navigation path planning of the pulmonary bronchial tree.

[0109] Furthermore, during the correction process, the cross-sectional feature information of each path pixel point in the target path may be calculated in the following manner:

[0110] For each path pixel point of a single lumen in the target path located in the bronchial tree, after obtaining a connected domain corresponding to a slice of the path pixel point on the natural lumen distance transformation volume data, the area of ​​the connected domain is calculated to obtain cross-sectional feature information of the path pixel point;

[0111] For each path pixel point in the target path located near the bifurcation position in the bronchial tree, after determining the target peak pixel point, the target peak pixel point is used as a seed point for watershed segmentation to obtain the tracheal area where the target peak pixel point is located, and the area of ​​the tracheal area is calculated to obtain the cross-sectional feature information of the path pixel point. Specifically, the basic idea of ​​watershed segmentation is: based on the seed point, the surrounding pixel points are traversed, and when the pixel value is 1 or there is no pixel to be expanded, it is exited. Figure 3a The effect of watershed segmentation on the bifurcated road area shown in the figure is as follows Figure 3b shown.

[0112] The cross-sectional feature information of each path pixel point is used to characterize the radius of the bronchial cavity at the path pixel point. Providing the cross-sectional feature information can help doctors understand the changes in the thickness of the cavity on the navigation path from the starting point to the target point.

[0113] In summary, the navigation path planning method provided in this embodiment, after acquiring the lung volume data image, firstly segments the bronchial tree from the lung volume data image to obtain the bronchial region volume data image, and performs 3D distance transformation on each bronchial region pixel point in the bronchial region volume data image to obtain bronchial distance transformation volume data, and then iterates on the bronchial region volume data image based on the minimization of the cost function according to the bronchial distance transformation volume data, searches for the target path from the starting point to the target point, and obtains the planned navigation path. This embodiment uses the method of minimum optimization iteration based on the cost function to search for the target path from the starting point to the target point, obtains the target navigation path based on the center line of the cavity of the pulmonary trachea, and improves the accuracy of the pulmonary bronchial navigation path planning; further, it can also provide cross-sectional feature information on the navigation path, providing more comprehensive path planning information, which can better assist doctors in improving the accuracy of diagnosis. In addition, this embodiment can automatically perform navigation path planning, reduce the cumbersome operation of human-computer interaction, and improve the efficiency of diagnosis. In addition, the navigation path planning algorithm of this embodiment has strong versatility, realizes an end-to-end algorithm flow, and further improves the diagnostic efficiency and accuracy.

[0114] Based on the above inventive concept, the present invention also provides a navigation path planning system, such as Figure 5 As shown, the navigation path planning system 200 may include a processor 210 and a memory 220. The memory 220 stores instructions. When the instructions are executed by the processor 210, the steps in the navigation path planning method described above may be implemented.

[0115] Among them, the processor 210 can perform various actions and processes according to the instructions stored in the memory 220. Specifically, the processor 210 can be an integrated circuit chip with signal processing capabilities. The above-mentioned processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The various methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can be an X86 architecture or an ARM architecture, etc.

[0116] The memory 220 stores executable instructions, which are executed by the processor 210 to perform the navigation path planning method described above. The memory 220 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of exemplary but not limiting explanation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous connection dynamic random access memory (SLDRAM) and direct memory bus random access memory (DR RAM). It should be noted that the memory of the method described herein is intended to include but is not limited to these and any other suitable types of memory.

[0117] Based on the same inventive concept, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the navigation path planning method described above can be implemented.

[0118] Similarly, the computer-readable storage medium in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. It should be noted that the computer-readable storage medium described herein is intended to include but is not limited to these and any other suitable types of memory.

[0119] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0120] In general, various example embodiments of the present invention may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of embodiments of the present invention are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0121] It should be noted that the various embodiments in this specification are described in a related manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system and computer-readable storage medium, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0122] The above description is only a description of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any changes or modifications made by a person skilled in the art in the field of the present invention based on the above disclosure shall fall within the scope of protection of the claims.

Claims

1. A navigation path planning method, characterized in that: include: Acquiring a volume data image of a target organ, wherein the target organ has a natural cavity; Segmenting the natural cavity in the target organ volume data image to obtain the natural cavity region volume data image; Performing 3D distance transformation on each natural cavity region pixel point in the natural cavity region volume data image to obtain natural cavity region distance transformation volume data; According to the natural cavity distance transformation volume data, based on minimization of the cost function, iterate on the natural cavity area volume data image, search for a target path from a starting point to a target point, and obtain a planned navigation path; Wherein, the starting point is a pixel point at the starting position of the natural cavity, and the target point is a pixel point in the natural cavity area adjacent to the lesion position in the volume data image of the natural cavity area; The performing 3D distance transformation on each natural cavity region volume pixel point in the natural cavity region volume data image includes: for each natural cavity region pixel point in the natural cavity region volume data image, traversing the surrounding neighborhood from near to far to find the nearest non-natural cavity region pixel point, and calculating the straight-line distance between the natural cavity region volume pixel point and the non-natural cavity region pixel point as the pixel value corresponding to the natural cavity region pixel point; The cost function is: ， ， ， ， ， in, represents the estimated cost of the starting point to reach the target point through the current point n, represents the actual cost of the path from the starting point to the current point n, represents the estimated cost of the best path from the current point n to the target point, represents the position coordinates of the starting point, represents the position coordinates of the current point, represents the position coordinates of the target point, , represents the weight value, represents the z coordinate in the position coordinates of the current point, represents the z coordinate in the position coordinates of the target point, Represents the absolute value of the difference between the maximum value in the natural cavity distance transformation volume data and the data value corresponding to the current point.

2. The navigation path planning method according to claim 1, characterized in that: Also includes: The target path is modified as follows: For each path pixel point of a single cavity located in the natural cavity in the target path, obtain a connected domain corresponding to a slice of the path pixel point on the natural cavity distance transformation volume data, find a peak pixel point corresponding to a peak value in the connected domain, and use the peak pixel point as a correction point for the path pixel point; For each path pixel point in the target path located near the bifurcation position in the natural cavity, a connected domain corresponding to the slice of the path pixel point on the natural cavity distance transformation volume data is obtained, and the connected domain is filtered to remove abnormal pixels, and then the peak pixel points corresponding to at least two peaks in the connected domain are found, and the target peak pixel point is used as the correction point of the path pixel point; wherein the target peak pixel point is the peak pixel point with the smallest distance from the path pixel point.

3. The navigation path planning method according to claim 2, characterized in that: Also includes: For each path pixel point of a single cavity in the target path located in the natural cavity, after obtaining a connected domain corresponding to a slice of the path pixel point on the natural cavity distance transformation volume data, the area of ​​the connected domain is calculated to obtain cross-sectional feature information of the path pixel point; For each path pixel point in the target path located near the bifurcation position in the natural cavity, after determining the target peak pixel point, the target peak pixel point is used as a seed point for watershed segmentation to obtain the cavity area where the target peak pixel point is located, and the area of ​​the cavity area is calculated to obtain the cross-sectional feature information of the path pixel point.

4. The navigation path planning method according to claim 1, characterized in that: The method of iterating on the volume data image of the natural cavity region based on minimization of the cost function to search for a target path from a starting point to a target point includes: Step A1: providing a first set and a second set, wherein the first set includes the starting point, and the second set is empty; Step B1: Select the pixel point with the smallest cost function value in the first set as the current point, put the current point into the second set, and process each pixel point in the neighborhood around the current point in the natural cavity area volume data image as follows: a. If the pixel is in the second set, no processing is performed; b. If the pixel point is not in the first set, add the pixel point to the first set, calculate the cost function value of the pixel point, and set the current point as the parent node of the pixel point; c. If the pixel point is in the first set, recalculate the cost function value of the pixel point; if the recalculated cost function value is less than the cost function value of the pixel point calculated last time, modify the parent node of the pixel point to the current point; Step C1: determining whether the first set is empty, and if not empty, returning to execute step B1 until the target point exists in the first set; Step D1, backtracking the target point, searching for the parent node and recording it until the starting point is found, and the recorded series of pixel points constitute the target path from the starting point to the target point.

5. The navigation path planning method according to claim 1, characterized in that: The segmenting of the natural cavity in the target organ volume data image to obtain the natural cavity region volume data image includes: Detecting the starting position of the cavity in the target organ volume data image and determining a regional growth seed point; The natural cavity is segmented by region growing to obtain a volume data image of the natural cavity region.

6. The navigation path planning method according to claim 5, characterized in that: The step of detecting the starting position of the cavity in the target organ volume data image and determining the region growth seed point comprises: Step A2, taking a layer of two-dimensional image located at the upper end of the target organ volume data image as an initial image; Step B2, binarizing the initial image, and calculating the areas and ellipticity of different connected domains in the initial image, and taking the connected domains whose areas and ellipticity meet preset conditions as potential cavity areas; Step C2, respectively calculating N layers of two-dimensional images adjacent to the initial image to obtain N potential cavity regions, where N≥3; Step D2, respectively calculate the centroid coordinates of the N+1 potential cavity areas. If the deviations between two adjacent centroid coordinates are less than a preset threshold, the centroid of the initial image is determined as the region growth seed point. Otherwise, the next layer of the two-dimensional image of the initial image is used as the initial image and return to execute step B2.

7. The navigation path planning method according to claim 1, characterized in that: Before performing 3D distance transformation on each natural cavity region volume pixel point in the natural cavity region volume data image, the method further includes: The holes inside the natural cavity in the natural cavity region volume data image are filled using morphological operations.

8. The navigation path planning method according to claim 1, characterized in that: Before segmenting the natural cavity in the target organ volume data image, the method further includes: The target organ volume data image is subjected to edge enhancement processing.

9. The navigation path planning method according to any one of claims 1 to 8, characterized in that: The target organ is the lung.

10. A navigation path planning system, characterized in that: The method comprises a processor and a memory, wherein the memory stores instructions, and when the instructions are executed by the processor, the method according to any one of claims 1 to 9 is implemented.

11. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images