Cancer focus image data processing method and processing system, electronic equipment and computer readable storage medium

Abstract

The invention provides a cancer lesion image data processing method and system, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining a cancer lesion mask corresponding to a cancer lesion; performing calculation processing on the cancer focus mask to obtain a focus datum point corresponding to the cancer focus mask; performing transformation processing on the cancer focus mask based on the focus reference point to obtain a K view of the cancer focus mask; and performing calculation processing based on the K view and the cancer focus mask to obtain an approximate value of the long diameter of the surface curve of the cancer focus. The technical problems that the error of measuring the long diameter of the cancer lesion through a manual measurement method or a clinician on the basis of computed tomography angiography images is large, and the long diameter of the cancer lesion cannot be accurately obtained without an operation or before the operation are solved.

Description

Technical Field

[0001] The present invention relates to the field of medical data processing, and in particular to a cancer lesion image data processing method, a processing system, an electronic device and a computer-readable storage medium. Background Art

[0002] In the current ultrasound lesion scanning process, medical personnel need to know the size of the largest cross-section of the lesion in order to understand the lesion in detail and accurately. In medicine, the long diameter and short diameter of the lesion are usually used to characterize the size of the lesion, while in related technologies, obtaining the long diameter, short diameter, and size of the lesion requires manual work by medical personnel, and related systems can only provide the distance measurement function between two points on the image.

[0003] When clinicians measure the long diameter of cancer lesions based on computed tomography angiography (CTA) images, they are not concerned with the length of the straight line through the lesion, but the length of the curve along the outer surface of the lesion. The current measurement method is to cut the lesion open and physically unfold it based on experience after surgical resection of the lesion, and then use a measuring tool to measure it. Disadvantages: It is impossible to measure without surgery or before surgery, and the approximate long diameter direction is sought based on experience. When unfolding, the cell tissue will be squeezed or stretched to a certain extent, resulting in large errors. In addition, different doctors have different experiences, and the measurement varies greatly.

[0004] The above manual measurement methods or the measurement of the long diameter of cancer lesions by clinicians based on computed tomography angiography (CTA) images are greatly affected by personal subjective factors and measurement tool errors. Especially when the lesion morphology is complex, it is impossible to effectively perform accurate measurements through the coronal, sagittal, and axial views.

[0005] Based on the above defects of the prior art, a technical solution is proposed that can quickly and accurately measure the long diameter of the surface curve of the cancer lesion based on the lesion mask without surgery. Summary of the invention

[0006] The purpose of the embodiments of the present invention is to solve the technical problems in the prior art that the error in measuring the long diameter of cancer lesions is large and the long diameter of cancer lesions cannot be accurately obtained without surgery or before surgery. To this end, the present invention proposes a cancer lesion image data processing method, processing system, electronic device and computer-readable storage medium.

[0007] A first aspect of the present invention provides a method for processing cancer lesion image data, the method comprising: obtaining a cancer lesion mask corresponding to the cancer lesion;

[0008] Perform computational processing on the cancer lesion mask to obtain the lesion reference points corresponding to the cancer lesion mask;

[0009] Perform transformation processing on the cancer lesion mask based on the lesion reference points to obtain the K views of the cancer lesion mask; where the K views include K first views, and K > 3, K is an integer;

[0010] Perform computational processing based on the K views and the cancer lesion mask to obtain an approximate value of the major axis of the cancer lesion surface curve.

[0011] Preferably, performing transformation processing on the cancer lesion mask based on the lesion reference points to obtain the K views of the cancer lesion mask specifically includes:

[0012] Taking the lesion reference point as the coordinate origin, rotate the cancer lesion mask 90 / M - 1 times by M degrees successively along the Z-axis, Y-axis, and Z-axis to obtain K - 3 first views of the cancer lesion mask; where 0 < M < 90, and M can divide 90 evenly;

[0013] The 2×(90 / M - 1)×3 first views of the cancer lesion mask, the coronal view, sagittal view, and axial view of the cancer lesion mask form the K views of the lesion mask.

[0014] Further preferably, taking the lesion reference point as the coordinate origin, rotating the cancer lesion mask 90 / M - 1 times by M degrees successively along the Z-axis, Y-axis, and Z-axis specifically means:

[0015] Taking the lesion reference point as the coordinate origin, using the Z-axis as the rotation axis, rotate the cancer lesion mask 90 / M - 1 times, each time rotating M degrees, to obtain 2×(90 / M - 1) first views;

[0016] Taking the lesion reference point as the coordinate origin, using the Y-axis as the rotation axis, rotate the cancer lesion mask 90 / M - 1 times by M degrees successively to obtain 2×(90 / M - 1) first views;

[0017] Taking the lesion reference point as the coordinate origin, using the X-axis as the rotation axis, rotate the cancer lesion mask 90 / M - 1 times by M degrees successively to obtain 2×(90 / M - 1) first views.

[0018] Further preferably, M and K satisfy the following relationship: K = 2×(90 / M - 1)×3 + 3.

[0019] Preferably, performing computational processing based on the K views and the cancer lesion mask to obtain an approximate value of the major axis of the cancer lesion surface curve specifically means:

[0020] Based on the K intersection surfaces formed by each first view in the K views intersecting with the cancer lesion mask respectively, obtain K first cross-sections;

[0021] Perform calculation processing according to the first cross section to obtain the first major axis R1;

[0022] Determine the intersection points A' and B' formed by the first major axis R1 and the boundary of the first cross section where it is located;

[0023] Calculate and process the intersection points A' and B', and the lengths of the two half curves are L1 and L2;

[0024] Determine the larger value of L1 and L2 as the first curve major diameter L of the first cross section;

[0025] The maximum value among the K first curve long axes L is determined as the approximate value of the curve long axis of the cancer lesion surface.

[0026] Preferably, the lesion reference point is the center of gravity of the cancer lesion mask.

[0027] Preferably, the lesion reference point is the centroid of the cancer lesion mask.

[0028] In a second aspect of the present invention, a cancer lesion image data processing system is provided, the system comprising: a data acquisition unit, a lesion reference point calculation unit, a view transformation processing unit and a data calculation unit; wherein the data acquisition unit is used to acquire a cancer lesion mask corresponding to the cancer lesion;

[0029] The lesion reference point calculation unit is used to perform calculation processing based on the cancer lesion mask to obtain the lesion reference point corresponding to the cancer lesion mask; the view transformation processing unit is used to perform view transformation processing on the cancer lesion mask based on the lesion reference point to obtain K views of the cancer lesion mask; wherein the K views include K first views, and K>3, K is an integer; the data calculation unit is used to perform calculation processing based on the K views and the cancer lesion mask to obtain the approximate value of the long diameter of the cancer lesion surface curve.

[0030] The third aspect of the present invention proposes an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute a cancer lesion image data processing method proposed in the first aspect of the present invention.

[0031] A fourth aspect of the present invention provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implements a cancer lesion image data processing method proposed in the first aspect of the present invention.

[0032] The present invention provides a cancer lesion image data processing method, processing system, electronic device and computer-readable storage medium, which calculates the long diameter of the surface curve of the cancer lesion based on the lesion mask. The method can eliminate the influence of human factors in measuring the long diameter of the surface curve of the cancer lesion in the prior art solution, and solve the problem that the long diameter of the surface curve of the cancer lesion cannot be effectively and accurately measured through the coronal, sagittal and axial three views in the prior art. Through the cancer lesion image data processing method and processing system based on the lesion mask provided by the present invention, the measured long diameter of the surface curve of the cancer lesion is highly accurate and consistent, which can provide a reliable and powerful reference for clinical medical diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A flowchart of a method for processing cancer lesion image data provided by an embodiment of the present invention;

[0034] Figure 2 A flowchart of a view transformation processing method provided by an embodiment of the present invention;

[0035] Figure 3 A flowchart of an example of a view transformation processing method provided by an embodiment of the present invention;

[0036] Figure 4 A method for calculating the approximate value of the long diameter of a cancer lesion surface curve provided in an embodiment of the present invention;

[0037] Figure 5 A schematic diagram of a first cross section provided by an embodiment of the present invention;

[0038] Figure 6 A system block diagram of a cancer lesion image data processing system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0040] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0041] The first and second in the following text are only for distinction and have no other meanings.

[0042] Figure 1 A flowchart of a cancer lesion image data processing method provided by an embodiment of the present invention is shown in FIG. Figure 1As shown, a cancer lesion image data processing method provided by an embodiment of the present invention comprises the following steps:

[0043] Step 110, obtaining a cancer lesion mask corresponding to the cancer lesion.

[0044] Specifically, in the embodiment of the present invention, a cancer lesion mask corresponding to the cancer lesion is acquired by a data acquisition unit.

[0045] The cancer lesion in the present invention refers to a lesion that is pre-generated by a medical imaging method or system in the prior art, and needs to be further calculated using the present invention for the long diameter of the lesion surface curve. For example, a cancer lesion generated by manual delineation or model prediction methods in the prior art, or obtained by an artificial intelligence detection method. The cancer lesion mask in the present invention refers to the cancer lesion mask data obtained by further processing the cancer lesion by an image processing scheme in the prior art.

[0046] The cancer lesion in the embodiment of the present invention can be various lesions called cancer or tumor in existing medical terms or general terms. In a specific embodiment of the present invention, the cancer lesion refers to a gastric cancer lesion.

[0047] Step 120 , calculating and processing the cancer lesion mask to obtain the lesion reference point corresponding to the cancer lesion mask.

[0048] Specifically, in the embodiments provided by the present invention, the lesion reference point is the center of gravity (mean (x, y, z)), centroid (3D bounding box center) or other forms of geometric points of the lesion, which can be located inside or outside the lesion body.

[0049] In the optional embodiment provided by the present invention, the lesion reference point calculation unit calculates the reference point, and the lesion reference point calculation unit provides a variety of reference point calculation schemes, including a center of gravity calculation scheme, a centroid calculation scheme, and other forms of geometric point calculation methods that can be obtained by the prior art. Further, the present invention also provides a method for configuring the selection of reference points, specifically including providing a reference point configuration interface, listing the reference point configuration options that can be selected individually in the reference point configuration section, and the options include the center of gravity, the centroid, and any other geometric points that can be selected in the prior art. The user can modify the reference point according to the size of the cancer lesion to be measured, after multiple modification attempts or relying on experience, and select the center of gravity, the centroid or one of other geometric points from the listed optional options. The embodiment provided by the present invention will automatically match the reference point calculation method according to the selected method to calculate the reference point.

[0050] In the preferred solution of the embodiment provided by the present invention, the centroid (mean(x,y,z)) of the cancer lesion is calculated based on the lesion mask, and the calculated centroid of the cancer lesion is used as the lesion reference point corresponding to the cancer lesion mask.

[0051] In the embodiment provided by the present invention, when calculating the centroid of the cancer lesion, a calculation method based on the volume of the cancer lesion is adopted to complete the calculation, so as to reduce the total calculation amount and improve the calculation efficiency.

[0052] Step 130: Based on the lesion reference point, perform transformation processing on the cancer lesion mask to obtain K views of the cancer lesion mask.

[0053] Specifically, the present invention executes the transformation processing steps of this step through a view transformation processing unit. Among them, the K views include K first views, and the K first views intersect with the cancer lesion mask respectively to form K first cross-sections, and K is a positive integer.

[0054] In the embodiment provided by the present invention, with the lesion reference point as the coordinate origin, that is, with the lesion reference point calculated and processed in step 120 as the coordinate origin, the lesion is rotated M degrees along the Z axis. At this time, it is equivalent to a -M degree rotation of both the X and Y axes, thus adding 2 more views. Among them, 0 < M < 90, and M can divide 90 evenly. This is because the included angle between any two views is 90 degrees, so the value of M should preferably be a degree that can be divided evenly by 90, that is, the value of the rotation angle M satisfies 0 < M < 90 and M can be divided evenly by 90. In the embodiment of the present invention, for example: the corresponding relationship between the value of M and the number of rotation times T is as follows:

[0055] M = 45, T = 1; that is, rotate 45 degrees, then rotate 1 time, and a total of 2 first views are added.

[0056] M = 30, T = 2; that is, rotate 30 degrees, then rotate 2 times, and a total of 4 first views are added.

[0057] M = 22.5, T = 3; that is, rotate 22.5 degrees, then rotate 3 times, and a total of 6 first views are added.

[0058] And so on, M = 1, T = 89; that is, rotate 1 degree, then rotate 89 times, and a total of 178 first views are added.

[0059] Similarly, when rotating along the X axis and along the Y axis, 2 first views will be added each time a rotation is made.

[0060] From the above analysis, it can be seen that the relationship between the number K of the added views and the value of M is as follows: K = 2×(90 / M - 1)×3 + 3.

[0061] It should be noted that in the transformation process provided in the embodiment of the present invention, the rotation along the X, Y, and Z axes is not performed simultaneously, but the original data of the cancer lesion mask is rotated 90 / M-1 M degrees along the X axis to obtain 2×(90 / M-1) first views; the original data of the cancer lesion mask is rotated 90 / M-1 M degrees along the Y axis to obtain 2×(90 / M-1) first views; the original data of the cancer lesion mask is rotated 90 / M-1 M degrees along the Z axis to obtain 2×(90 / M-1) first views; a total of 2×(90 / M-1)×3 first views are obtained. Together with the coronal view, sagittal view, and axial view of the original cancer lesion mask, a total of 2×(90 / M-1)×3+3 views are obtained, that is, the K view of the cancer lesion mask is obtained.

[0062] In the embodiment provided by the present invention, the view transformation processing unit uses the following method to perform transformation processing to obtain a K view of the cancer lesion mask: Figure 2 A flowchart of a view transformation processing method provided by an embodiment of the present invention is shown in the figure. The K view is obtained by calculating with the reference point as the coordinate origin, including the following steps:

[0063] Step 210, taking the lesion reference point as the coordinate origin and the Z axis as the rotation axis, the cancer lesion mask is rotated 90 / M-1 times, M degrees respectively, to obtain 2×(90 / M-1) first views;

[0064] Step 220, taking the lesion reference point as the coordinate origin and the Y axis as the rotation axis, the cancer lesion mask is rotated 90 / M-1 times, M degrees respectively, to obtain 2×(90 / M-1) first views;

[0065] Step 230, taking the lesion reference point as the coordinate origin and the X axis as the rotation axis, the cancer lesion mask is rotated 90 / M-1 times, M degrees respectively, to obtain 2×(90 / M-1) first views;

[0066] Step 240 , compose a K view of the lesion mask according to the 2×(90 / M-1)×3 first views of the cancer lesion mask, the coronal view, the sagittal view and the axial view of the cancer lesion mask.

[0067] The above steps can be summarized as follows: taking the lesion reference point as the coordinate origin, the cancer lesion mask is rotated 90 / M-1 M degrees along the Z axis, Y axis and Z axis respectively, to obtain 2×(90 / M-1)×3 first views of the cancer lesion mask; the K view of the lesion mask is composed of the 2×(90 / M-1)×3 first views of the cancer lesion mask, the coronal view, the sagittal view and the axial view of the cancer lesion mask, where K=2×(90 / M-1)×3+3.

[0068] The above transformation processing steps are further explained below with a specific example. Take M=30. Therefore, the cancer lesion mask needs to be rotated twice along the Z axis, the Y axis and the X axis respectively, and finally 15 views of the cancer lesion mask are obtained. Figure 3 A flowchart of an example of a view transformation processing method provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, when M=30, according to the embodiment provided by the present invention, the cancer lesion mask is subjected to view transformation processing to obtain K=15 views. The specific steps of the transformation processing are as follows:

[0069] Step 310, taking the lesion reference point as the coordinate origin and the Z axis as the rotation axis, the cancer lesion mask is rotated twice by 30 degrees, so that the number of first views of the lesion mask added is 4;

[0070] Specifically, the cancer lesion mask is rotated 30 degrees for the first time with the lesion reference point as the coordinate origin and the Z axis as the rotation axis to obtain two first views. Then, it is rotated 30 degrees for the second time to obtain two more first views. At this time, four first views are obtained.

[0071] Step 320, taking the lesion reference point as the coordinate origin and the Y axis as the rotation axis, the cancer lesion mask is rotated twice by 30 degrees, so that the number of first views of the lesion mask added is 8;

[0072] Specifically, the cancer lesion mask is rotated 30 degrees for the first time with the lesion reference point as the coordinate origin and the Y axis as the rotation axis to obtain two first views. Then, it is rotated 30 degrees for the second time to obtain two more first views. When the entire transformation process is executed to this point, a total of 8 first views are obtained.

[0073] Step 330, taking the lesion reference point as the coordinate origin and the X axis as the rotation axis, the cancer lesion mask is rotated twice by 30 degrees, so that the number of first views of the lesion mask added is 12;

[0074] Specifically, the cancer lesion mask is rotated 30 degrees for the first time with the lesion reference point as the coordinate origin and the X-axis as the rotation axis to obtain two first views. Then, it is rotated 30 degrees for the second time to obtain two more first views. When the entire transformation process is executed to this point, a total of 12 first views are obtained.

[0075] Step 340, 15 views of the lesion mask are composed of the 12 first views of the cancer lesion mask added after the processing of steps 310 to 330 and the coronal view, sagittal view and axial view of the cancer lesion mask.

[0076] In the embodiment provided by the present invention, in the transformation processing step, the value setting of the rotation angle M of the lesion mask will affect the calculation speed. Therefore, in the optional scheme of the embodiment provided by the present invention, the balance between speed and accuracy can be sought through multiple tests in use, and the following settings can be made based on experience: when the lesion volume is greater than the first volume preset threshold, only the center of gravity is used as the coordinate origin; when the lesion volume is greater than the second volume preset threshold, the rotation angle M is gradually increased, etc. to configure the values ​​of the lesion reference point and the rotation angle M. The number of coordinate origins and the rotation angle M can also be flexibly configured as needed according to the hardware environment conditions of the implementation of the invention embodiment, or experience. The first volume preset threshold and the second volume preset threshold are determined by relevant personnel through multiple tests based on the hardware conditions of the computer equipment that implements the method and the experience of relevant personnel.

[0077] In the preferred solution of the embodiment provided by the present invention, the view transformation processing unit performs view transformation processing in a parallel processing manner, and sets the value of M to 1, so that the dimension of the view can be increased. When M=1, K=2×(90÷1-1)×3+3=537, and therefore, the cancer lesion mask is transformed into 537 views of the cancer lesion mask. This can effectively improve the accuracy of the approximate value of the long diameter of the cancer lesion surface curve measured by the present invention. Of course, M can also be set to a smaller value. The larger the value of M, the greater the overhead for the computing system. During use, it is necessary to adjust the size of M according to actual experience so that the computing efficiency and accuracy can be balanced.

[0078] Step 140 , performing calculation processing based on the K view and the cancer lesion mask to obtain an approximate value of the long diameter of the cancer lesion surface curve.

[0079] Specifically, in the embodiment provided by the present invention, K first cross sections formed by the intersection of K first views of the K view and the cancer lesion mask are respectively calculated and processed to obtain the approximate value of the long diameter of the cancer lesion surface curve. More specifically, in the embodiment provided by the present invention, K first views are respectively intersected with the cancer lesion mask to form K first cross sections, and the data calculation unit performs calculation processing based on the K first cross sections to obtain the approximate value of the long diameter of the cancer lesion surface curve. Figure 4 A method for calculating the approximate value of the long diameter of the surface curve of a cancer lesion provided by an embodiment of the present invention, the specific calculation steps are as follows:

[0080] Step 410: Obtain K first cross sections according to K intersection surfaces formed by the intersection of each first view in the K views and the cancer lesion mask.

[0081] Specifically, in the embodiment provided by the present invention, K first views intersect with the cancer lesion mask respectively to form K first cross sections. Figure 5A schematic diagram of a first cross section provided by an embodiment of the present invention, such as Figure 5 As shown, the graphic is a first cross section formed by the intersection of the cancer lesion mask and one of the K first views.

[0082] Step 420 , performing calculation processing based on the first cross section to obtain the first major axis R1 .

[0083] Specifically, the long axis and short axis of the first cross section are calculated by the 2D graphics calculation method in the prior art, and are recorded as the first long axis R1 and the first short axis R2. Figure 5 As shown, the figure is the intersection of the cancer lesion mask and the first view at a closed intersection line (the bold line in the figure is a closed boundary curve), which is recorded as the first section boundary, and the surface formed by the closed intersection line is the first section. All points on the first section are in the same plane, that is, the first section is a plane, not a curved surface. The embodiment of the present invention adopts the prior art graphic data processing method to calculate and process the long axis R1 and the short axis R2 of the cancer lesion.

[0084] Step 430 , determining intersection points A′ and B′ formed by the first major axis R1 and the boundary of the first cross section where the first major axis R1 is located.

[0085] Specifically, Figure 5 As shown, the embodiment of the present invention adopts the prior art graphic data processing method to calculate and process to obtain the intersection points A' and B' formed by the long axis R1 of the cancer lesion and the boundary of the first section.

[0086] Step 440 , performing calculation processing based on the intersection point A′, the intersection point B′ and the first cross-section boundary to obtain two half-curve lengths L1 and L2 .

[0087] Specifically, Figure 5 As shown, the embodiment of the present invention adopts the prior art graphic data processing method to calculate and process to obtain two half curves L1 and L2 along the intersection line of the cancer lesion, and calculates the values ​​of L1 and L2 by a graphic calculation method.

[0088] Step 450, determining the larger value of L1 and L2 as the first curve major diameter L of the first cross section.

[0089] Specifically, the values ​​of L1 and L2 are compared, and the one with a larger value is determined as the first curve major diameter L of the first cross section.

[0090] Step 460: determine the maximum value among the K first curve long axes L as the approximate value of the cancer lesion surface curve long axis.

[0091] Specifically, after the K first cross sections are processed according to step 420 to step 450, K first curve long diameters L are obtained. The K first curve long diameters L are compared in size, and the maximum value L is confirmed as the approximate value of the curve long diameter of the cancer lesion surface.

[0092] In an optional solution provided by an embodiment of the present invention, the following method is further used to perform calculation processing based on K views and cancer lesion masks to obtain an approximate value of the long axis of the cancer lesion surface curve: performing calculation processing based on K first cross sections to obtain K first long axes R1;

[0093] Based on the obtained K first major axes R1, a comparison is performed, and the R1 with the largest value among the K first major axes R1 is determined as the cancer lesion major axis R' corresponding to the cancer lesion mask; and the first section where the cancer lesion major axis R' is located is determined as the cancer lesion major axis section corresponding to the cancer lesion mask; the intersection point A" and the intersection point B" of the cancer lesion major axis R' and the boundary of the cancer lesion major axis section are determined; the curve length is calculated based on the intersection point A" and the intersection point B", and the two half curve lengths L1" and L2" of the cancer lesion major axis section are obtained; the larger value of L1" and L2" is determined as the approximate value of the major axis of the cancer lesion surface curve.

[0094] The above introduces a cancer lesion image data processing method provided by the present invention in the form of an embodiment. The method calculates the long axis of the cancer lesion surface curve based on the lesion mask. The method can eliminate the influence of human factors in measuring the long axis of the cancer lesion surface curve in the prior art solution, and solves the problem that the prior art cannot effectively and accurately measure the long axis of the cancer lesion surface curve through the coronal, sagittal and axial three views.

[0095] The embodiment of the present invention further provides a cancer lesion image data processing system, which is used to execute a cancer lesion image data processing method provided by the embodiment of the present invention. Figure 6 A system block diagram of a cancer lesion image data processing system provided by an embodiment of the present invention, such as Figure 6As shown, the cancer lesion image data processing system 600 includes a data acquisition unit 601, a lesion reference point calculation unit 602, a view transformation processing unit 603 and a data calculation unit 604. The data acquisition unit is used to acquire a cancer lesion mask corresponding to the cancer lesion; the lesion reference point calculation unit is used to perform calculation processing based on the cancer lesion mask to obtain the lesion reference point corresponding to the cancer lesion mask; the view transformation processing unit is used to perform view transformation processing on the cancer lesion mask based on the lesion reference point to obtain K views of the cancer lesion mask; wherein the K view includes K first views, and K>3, K is an integer; the data calculation unit is used to perform calculation processing based on the K view and the cancer lesion mask to obtain the approximate value of the long diameter of the cancer lesion surface curve. The specific manner and steps of the system to perform a cancer lesion image data processing method provided by an embodiment of the present invention have been introduced in detail when introducing a cancer lesion image data processing method provided by an embodiment of the present invention, and will not be repeated again.

[0096] An embodiment of the present invention further provides an electronic device, comprising a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute a method for processing cancer lesion image data provided by the present invention.

[0097] An embodiment of the present invention further provides a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, a method for processing cancer lesion image data provided by the present invention is implemented.

[0098] The embodiment of the present invention provides a cancer lesion image data processing method, processing system, electronic device and computer-readable storage medium, which can calculate the long diameter of the surface curve of the cancer lesion based on the lesion mask. The method eliminates the influence of human factors in measuring the long diameter of the surface curve of the cancer lesion in the prior art solution, and solves the problem that the long diameter of the surface curve of the cancer lesion cannot be effectively and accurately measured through the three views of coronal, sagittal and axial in the prior art. Through the cancer lesion image data processing method and processing system based on the lesion mask provided by the present invention, the measured long diameter of the surface curve of the cancer lesion is highly accurate and consistent, which can provide a reliable and powerful reference for clinical medical diagnosis.

[0099] The professionals should further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0100] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0101] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for processing cancer lesion image data, characterized in that the method Comprising: Obtaining a cancer lesion mask corresponding to a cancer lesion; Performing calculation processing on the cancer lesion mask to obtain a lesion reference point corresponding to the cancer lesion mask; Performing transformation processing on the cancer lesion mask based on the lesion reference point to obtain K views of the cancer lesion mask; wherein, the K views include K first views, and K > 3, K is an integer; Performing calculation processing based on the K views and the cancer lesion mask to obtain an approximate value of the major axis length of the cancer lesion surface curve.

2. The method for processing cancer lesion image data according to claim 1, characterized in that: Performing transformation processing on the cancer lesion mask based on the lesion reference point to obtain K views of the cancer lesion mask specifically includes: Taking the lesion reference point as the coordinate origin, rotating the cancer lesion mask 90 / M - 1 times by M degrees successively along the Z-axis, Y-axis, and Z-axis to obtain K - 3 first views of the cancer lesion mask; wherein, 0 < M < 90, and M can divide 90 evenly; The 2×(90 / M - 1)×3 first views of the cancer lesion mask, the coronal view, sagittal view, and axial view of the cancer lesion mask form the K views of the lesion mask.

3. The method for processing cancer lesion image data according to claim 2, characterized in that: Taking the lesion reference point as the coordinate origin, rotating the cancer lesion mask 90 / M - 1 times by M degrees successively along the Z-axis, Y-axis, and Z-axis specifically means: Taking the lesion reference point as the coordinate origin, taking the Z-axis as the rotation axis, rotating the cancer lesion mask 90 / M - 1 times, each time rotating by M degrees, to obtain 2×(90 / M - 1) first views; Taking the lesion reference point as the coordinate origin, taking the Y-axis as the rotation axis, rotating the cancer lesion mask successively by 90 / M - 1 times by M degrees respectively to obtain 2×(90 / M - 1) first views; Taking the lesion reference point as the coordinate origin, taking the X-axis as the rotation axis, rotating the cancer lesion mask successively by 90 / M - 1 times by M degrees respectively to obtain 2×(90 / M - 1) first views.

4. The method for processing cancer lesion image data according to claim 2 or 3, characterized in that: M and K satisfy the following relationship: K = 2×(90 / M - 1)×3 + 3.

5. The method for processing cancer lesion image data according to claim 1, characterized in that: Performing calculation processing based on the K views and the cancer lesion mask to obtain an approximate value of the major axis length of the cancer lesion surface curve specifically includes: Obtaining K first cross-sections according to the K intersection surfaces formed by each first view in the K views intersecting with the cancer lesion mask respectively; Performing calculation processing according to the first cross-section to obtain a first major axis R1; Determining the intersection points A' and B' formed by the intersection of the first major axis R1 and the boundary of its所在 first cross-section; Performing calculation processing according to the intersection points A', B' and the boundary of the first cross-section to obtain two half-curve lengths L1 and L2; Determining the larger value of L1 and L2 as the first curve major axis length L of the first cross-section; Determining the maximum value among the K first curve major axis lengths L as the approximate value of the major axis length of the cancer lesion surface curve.

6. The method for processing cancer lesion image data according to claim 1, characterized in that: The lesion reference point is the centroid of the cancer lesion mask.

7. The method for processing cancer lesion image data according to claim 1, characterized in that: The lesion reference point is the centroid of the cancer lesion mask.

8. A cancer lesion image data processing system, characterized in that: The system includes a data acquisition unit, a lesion reference point calculation unit, a view transformation processing unit, and a data calculation unit; wherein, The data acquisition unit is used to obtain a cancer lesion mask corresponding to a cancer lesion; The lesion reference point calculation unit is used to perform calculation processing according to the cancer lesion mask to obtain a lesion reference point corresponding to the cancer lesion mask; A view transformation processing unit, configured to perform view transformation processing on the cancer lesion mask based on the lesion reference point to obtain K views of the cancer lesion mask; wherein the K views include K first views, and K>3, K is an integer; The data calculation unit is used to perform calculation processing based on the K view and the cancer lesion mask to obtain an approximate value of the long diameter of the cancer lesion surface curve.

9. An electronic device, comprising: processor; A memory for storing processor executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute a cancer lesion image data processing method according to claims 1-7.

10. A computer-readable storage medium having computer program instructions stored thereon, wherein when the computer program instructions are executed by a processor, a method for processing cancer lesion image data according to claims 1-7 is implemented.

Images