Method, system and device for automatically measuring lesion image

Abstract

The invention provides an automatic measurement method, system and device for a lesion image, and relates to the technical field of lesion image measurement, and the method mainly comprises the steps: carrying out the medical segmentation of an image based on a measured organ, including lesion segmentation and tissue segmentation obtained according to a medical original image; lesion segmentation is used as a target, and tissue segmentation adjacent to the target is selected as a reference; performing boundary extraction on the reference to generate a tissue distance field; marking the target in the tissue distance field to obtain a target distance field; in the target distance field, a first reference point and a second reference point are extracted, and a connecting line of the two is used as a measuring line. According to the scheme, the size of the lesion area can be accurately measured according to different scenes and different requirements, so that the efficiency and effect of subsequent treatment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of lesion image measurement, and in particular to an automated lesion image measurement method, system and device. Background Art

[0002] At present, artificial intelligence technology in medical imaging is gradually maturing. For the measurement of medical lesion images, rough size measurements are often performed using the major and minor axes or circumscribed rectangles of the target lesion area. However, the measurement results are not accurate and cannot meet the precise measurement requirements of lesion images in some medical sub-fields, thus affecting the subsequent treatment effects.

[0003] For example, in the dental field, Figure 1 As shown, existing lesions (such as Figure 1 The image measurement method can only measure the straight-line distance between adjacent tooth surfaces (such as Figure 1 The straight-line distance between the two purple points on the red line is taken as the maximum width of the lesion, and the maximum width of the lesion in this view direction should be Figure 1 The straight-line distance between the two purple points on the middle blue line will result in measurement errors and cannot meet actual usage needs. Summary of the Invention

[0004] The object of the present invention is to provide a method, system and device for automatically measuring lesion images to solve at least one of the above-mentioned technical problems existing in the prior art.

[0005] In a first aspect, to solve the above technical problems, the present invention provides an automated lesion image measurement method, comprising the following steps:

[0006] Step 1: Based on the medical segmentation image of the organ under test, including lesion segmentation and tissue segmentation obtained according to the original medical image; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark.

[0007] In a feasible embodiment, the detected organ is a tooth.

[0008] In a feasible implementation, the original medical image is a tomographic image such as CT (Computed Tomography), CBCT (CT improved for dentistry), or MRI (Magnetic Resonance Imaging).

[0009] In a feasible embodiment, the tissue segmentation is tooth bone segmentation.

[0010] In a feasible embodiment, the specific method for selecting the benchmark in step 1 includes: obtaining the adjacent area of ​​lesion segmentation and tissue segmentation through morphological operations; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark based on the selection strategy.

[0011] In a feasible implementation, the morphological operation specifically includes the following steps:

[0012] Step 11: In the medical segmentation image, a region of interest is captured with the lesion segmented as the center, in a preset box and in a preset direction;

[0013] Step 12: In the preset layer of the region of interest, conventional dilation operations are performed on the lesion segmentation and the tissue segmentation according to the preset dilation assignment to obtain the expanded overlapping area;

[0014] Step 13: Take the largest overlapping area in each preset layer as the adjacent area;

[0015] In this way, according to the actual three-dimensional morphology of the lesion segmentation, the adjacent area images with the best measurement position can be obtained from different directions and different layers for subsequent processing.

[0016] In a feasible implementation, the dilation value is assigned to 10 pixels, which is applicable to most application scenarios.

[0017] In a feasible implementation, the selection strategy is: within a preset range, select several tissue monomers with the longest adjacency line lengths as a benchmark.

[0018] Step 2: Extract the boundary of the reference and generate a tissue distance field.

[0019] In a feasible implementation manner, the step 2 specifically includes:

[0020] Step 21: extracting the boundary of the reference to obtain a reference boundary area;

[0021] Step 22: Generate a tissue distance field of the full radiation image using the contour line of the reference boundary area (which may be the entire contour line or a portion of the contour line, depending on the specific medical anatomical structure) as a starting line.

[0022] In a feasible implementation manner, the step 22 specifically includes:

[0023] Step 221: Using the pixel points on the starting line as starting points, obtain the expansion area by using the four-neighborhood translation comparison method;

[0024] Step 222: Calculate the distance between each point in the extended area and the corresponding starting point;

[0025] Step 223: assigning contour values ​​according to the distance;

[0026] Step 224: Convert the distance into an isoline value, where a larger value indicates a longer distance.

[0027] Step 3: Mark the target in the tissue distance field to obtain a target distance field.

[0028] Step 4: In the target distance field, extract the first reference point and the second reference point, and use the line connecting the two as the measurement line.

[0029] In a feasible implementation manner, step 4 may specifically include:

[0030] Step 4a1: In the target distance field, select the maximum value point as the first reference point, i.e., point A;

[0031] Step 4a2: On the contour line of the reference boundary area, select the point closest to point A as the second reference point, i.e., point B;

[0032] Step 4a3: Use the line AB between point A and point B as the measurement line of the maximum width of the lesion, and use the distance difference between point A and point B as the measurement value.

[0033] In a feasible implementation manner, the step 4 may further include:

[0034] Step 4b1: In the target distance field, select a number of maximum value points (e.g., the top 2%) as the first reference point candidate set, and then take the average of the points in the first reference point candidate set as the first reference point, i.e., point A;

[0035] Step 4b2: The second reference point is extracted by selecting a number of points (e.g., the top 2%) closest to point A on the contour line of the reference boundary area as a second reference point candidate set, and then taking the average of the points in the second reference point candidate set as the second reference point, i.e., point B.

[0036] Step 4b3: Use the line AB between point A and point B as the measurement line of the maximum width of the lesion, and use the distance difference between point A and point B as the measurement value;

[0037] In this way, even if the segmentation edge of the medical segmentation image is relatively fuzzy, the optimal measurement line and measurement value can be approximately obtained.

[0038] In a second aspect, based on the same inventive concept, the present application also provides an automated lesion image measurement system, comprising a data receiving module, a data processing module, and a result generating module;

[0039] The data receiving module is used to receive the medical segmentation image of the organ under test, including the lesion segmentation and tissue segmentation obtained based on the original medical image; taking the lesion segmentation as the target and selecting the tissue segmentation adjacent to the target as the reference;

[0040] The data processing module includes a tissue distance field unit, a target distance field unit and a measurement unit;

[0041] The tissue distance field unit extracts the boundary of the reference to generate a tissue distance field;

[0042] The target distance field unit marks the target in the tissue distance field to obtain a target distance field;

[0043] The measuring unit extracts a first reference point and a second reference point in the target distance field and uses a line connecting the first reference point and the second reference point as a measuring line;

[0044] The result generating module is used to send the measurement result of the measurement unit to external devices.

[0045] In a third aspect, based on the same inventive concept, the present application also provides an automated lesion image measurement device, comprising a processor, a memory, and a bus, wherein the memory stores instructions and data read by the processor, and the processor is used to call the instructions and data in the memory to execute the automated lesion image measurement method as described above, and the bus connects the functional components for transmitting information.

[0046] By adopting the above technical solution, the present invention has the following beneficial effects:

[0047] The present invention provides an automated lesion image measurement method, system, and device that can accurately measure the size of lesion areas according to different scenarios and needs, thereby improving the efficiency and effectiveness of subsequent treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 This is a diagram illustrating the problem of lesion image measurement in the background technology;

[0050] Figure 2 A flow chart of an automated lesion image measurement method provided by an embodiment of the present invention;

[0051] Figure 3 An example diagram of an overlapping area provided in an embodiment of the present invention;

[0052] Figure 4 An example diagram of a benchmark provided for an embodiment of the present invention;

[0053] Figure 5 An example diagram of boundary extraction provided by an embodiment of the present invention;

[0054] Figure 6 An example diagram of a tissue distance field provided by an embodiment of the present invention;

[0055] Figure 7 An example diagram of a target distance field provided by an embodiment of the present invention;

[0056] Figure 8 An example diagram of point A and point B provided in an embodiment of the present invention;

[0057] Figure 9 An example diagram of a measurement line provided by an embodiment of the present invention;

[0058] Figure 10 An example diagram of a complex lesion provided by an embodiment of the present invention;

[0059] Figure 11 An example diagram of two benchmarks in a complex lesion provided by an embodiment of the present invention;

[0060] Figure 12 An example diagram of point A and point B in a complex lesion provided by an embodiment of the present invention;

[0061] Figure 13 An example diagram of existing measurement lines for complex lesions provided by an embodiment of the present invention;

[0062] Figure 14 A diagram of an automated lesion image measurement system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0064] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0065] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0066] The present invention will be further explained below with reference to specific embodiments.

[0067] It should also be noted that the following specific embodiments or specific implementations are a series of optimized settings listed in the present invention to further explain the specific content of the invention, and these settings can be combined or used in association with each other.

[0068] Example 1:

[0069] like Figure 2 As shown, this embodiment provides an automatic lesion image measurement method, comprising the following steps:

[0070] Step 1: Based on the medical segmentation image of the organ under test, including lesion segmentation and tissue segmentation obtained according to the original medical image; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark.

[0071] Furthermore, the detected organ is a tooth.

[0072] Furthermore, the original medical image is a CT (Computed Tomography) or CBCT (CT improved for dentistry) or MRI (Magnetic Resonance Imaging) image.

[0073] Furthermore, the tissue segmentation is tooth-bone segmentation.

[0074] Furthermore, the specific method for selecting the benchmark in step 1 includes: obtaining adjacent areas of the lesion segmentation and the tissue segmentation through morphological operations; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark based on the selection strategy.

[0075] Furthermore, the morphological operation specifically includes the following steps:

[0076] Step 11: In the medical segmentation image, a region of interest is captured with the lesion segmented as the center, in a preset box and in a preset direction;

[0077] Step 12: In the preset layer of the region of interest, the lesion segmentation and the tooth bone segmentation are expanded according to the preset expansion assignment to obtain the expanded overlapping area, such as Figure 3 As shown by the yellow line in the figure, the horizontal coordinate is the x-coordinate and the vertical coordinate is the y-coordinate;

[0078] Step 13: Take the largest overlapping area in each preset layer as the adjacent area;

[0079] In this way, according to the actual three-dimensional morphology of the lesion segmentation, the adjacent area images with the best measurement position can be obtained from different directions and different layers for subsequent processing.

[0080] Furthermore, the dilation value is assigned to 10 pixels.

[0081] Furthermore, the selection strategy is to select a tissue monomer with the longest adjacent line length as a benchmark, such as Figure 4 Shown in yellow crown.

[0082] Step 2: Extract the boundary of the reference and generate a tissue distance field.

[0083] Furthermore, the step 2 specifically includes:

[0084] Step 21: Figure 5 As shown by the yellow circle in FIG, the boundary of the benchmark is extracted to obtain a benchmark boundary area;

[0085] Step 22: Using the outline of the reference boundary area as the starting line, generate the tissue distance field of the radiation full image, such as Figure 6 shown.

[0086] Furthermore, the step 22 specifically includes:

[0087] Step 221: Using the pixel points on the starting line as starting points, obtain the expansion area by using the four-neighborhood translation comparison method;

[0088] Step 222: Calculate the distance between each point in the extended area and the corresponding starting point;

[0089] Step 223: assigning contour values ​​according to the distance;

[0090] Step 224: Convert the distance into an isoline value, where a larger value indicates a longer distance.

[0091] Furthermore, the four-neighborhood translation comparison method specifically includes: in a Python environment, performing a four-direction (up, down, left, and right) translation operation on the original image to create a four-neighborhood translation matrix; then taking the maximum value of the four-neighborhood translation matrix pixel by pixel to obtain the expansion area in four directions.

[0092] Step 3: In the tissue distance field, the target is marked by mask operation to obtain the target distance field, such as Figure 7 As shown in the gradient area in .

[0093] Step 4: Figure 8 As shown, in the target distance field, a first reference point and a second reference point are extracted, and the line connecting the two is used as a measurement line.

[0094] Furthermore, the step 4 specifically includes:

[0095] Step 4a1: In the target distance field, select the maximum value point as the first reference point, i.e., point A;

[0096] Step 4a2: On the contour line of the reference boundary area, select the point closest to point A as the second reference point, i.e., point B;

[0097] Step 4a3: Figure 9 As shown, the line AB between point A and point B is used as the measurement line of the maximum width of the lesion, and the distance difference between point A and point B is used as the measurement value.

[0098] Example 2:

[0099] For complex lesions mixed between multiple crowns, such as Figure 10 As shown in the light green area in FIG, this embodiment provides an automatic measurement method for lesion images, including the following steps:

[0100] Step 1: Based on the medical segmentation image of the teeth, including lesion segmentation and tissue segmentation obtained according to the original medical image; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark.

[0101] Furthermore, the original medical image is a CT, CBCT or MRI image.

[0102] Furthermore, the tissue segmentation is tooth-bone segmentation.

[0103] Furthermore, the specific method for selecting the benchmark in step 1 includes: obtaining adjacent areas of the lesion segmentation and the tissue segmentation through morphological operations; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark based on the selection strategy.

[0104] Furthermore, the morphological operation specifically includes the following steps:

[0105] Step 11: In the medical segmentation image, a region of interest is captured with the lesion segmented as the center, in a preset box and in a preset direction;

[0106] Step 12: In the preset layer of the region of interest, the lesion segmentation and the tooth bone segmentation are respectively expanded according to the preset expansion assignment to obtain the expanded overlapping area, where the horizontal coordinate is the x-coordinate and the vertical coordinate is the y-coordinate;

[0107] Step 13: Take the largest overlapping area in each preset layer as the adjacent area;

[0108] In this way, according to the actual three-dimensional morphology of the lesion segmentation, the adjacent area images with the best measurement position can be obtained from different directions and different layers for subsequent processing.

[0109] Furthermore, the dilation value is assigned to 10 pixels.

[0110] Furthermore, the selection strategy is to select the two tissue monomers with the longest adjacency line length within a preset range as a benchmark, such as Figure 11 shown.

[0111] Step 2: Extract the boundaries of each benchmark and generate their own tissue distance fields.

[0112] Furthermore, the step 2 specifically includes:

[0113] Step 21: Extract the boundary of each benchmark to obtain the benchmark boundary area;

[0114] Step 22: Using the outline of the reference boundary area as a starting line, generate a tissue distance field of the radiation full image.

[0115] Furthermore, the step 22 specifically includes:

[0116] Step 221: Using the pixel points on the starting line as starting points, obtain the expansion area by using the four-neighborhood translation comparison method;

[0117] Step 222: Calculate the distance between each point in the extended area and the corresponding starting point;

[0118] Step 223: assigning contour values ​​according to the distance;

[0119] Step 224: Convert the distance into an isoline value, where a larger value indicates a longer distance.

[0120] Step 3: In the tissue distance field, mark the target through a mask operation to obtain a corresponding target distance field.

[0121] Step 4: In the target distance field, extract the first reference point and the second reference point, and use the line connecting the two as the measurement line.

[0122] Furthermore, the step 4 specifically includes: selecting the two closest points on the contour lines of the two reference boundary areas in the target distance field as point A and point B, respectively, such as Figure 12 As shown; the line AB between point A and point B is used as the measurement line of the maximum width of the lesion between the two crowns (references), and the length of AB is used as the measurement value; in this way, the limitation of the existing technology that can only calculate the overall width of the tooth lesion can be avoided, such as Figure 13 As shown in the figure, the measurement line runs through the entire tooth lesion.

[0123] Of course, in other embodiments, more references may be selected as needed to perform lesion image measurement so as to obtain a local width value of the lesion.

[0124] Example 3:

[0125] like Figure 14 As shown, this embodiment provides an automatic lesion image measurement system, including a data receiving module, a data processing module and a result generating module;

[0126] The data receiving module is used to receive the medical segmentation image of the organ under test, including the lesion segmentation and tissue segmentation obtained based on the original medical image; taking the lesion segmentation as the target and selecting the tissue segmentation adjacent to the target as the reference;

[0127] The data processing module includes a tissue distance field unit, a target distance field unit and a measurement unit;

[0128] The tissue distance field unit extracts the boundary of the reference to generate a tissue distance field;

[0129] The target distance field unit marks the target in the tissue distance field to obtain a target distance field;

[0130] The measuring unit extracts a first reference point and a second reference point in the target distance field and uses a line connecting the first reference point and the second reference point as a measuring line;

[0131] The result generating module is used to send the measurement result of the measurement unit to external devices.

[0132] Example 4:

[0133] This embodiment provides an automated lesion image measurement device, including a processor, a memory, and a bus. The memory stores instructions and data read by the processor. The processor is used to call the instructions and data in the memory to execute the automated lesion image measurement method described above. The bus connects the functional components to transmit information.

[0134] In another embodiment, this solution can also be implemented in the form of an integrated device, which can include corresponding modules for performing each or several steps in each of the above embodiments. The modules can be one or more hardware modules specifically configured to perform the corresponding steps, or implemented by a processor configured to perform the corresponding steps, or stored in a computer-readable medium for implementation by the processor, or implemented by some combination.

[0135] The processor performs the various methods and processes described above. For example, the method implementation in this solution can be implemented as a software program, which is tangibly contained in a machine-readable medium, such as a memory. In some embodiments, part or all of the software program can be loaded and / or installed via a memory and / or a communication interface. When the software program is loaded into the memory and executed by the processor, one or more steps in the method described above can be performed. Alternatively, in other embodiments, the processor can be configured to perform one of the above methods by any other appropriate means (e.g., by means of firmware).

[0136] The device can be implemented using a bus architecture. The bus architecture can include any number of interconnecting buses and bridges, depending on the specific application and overall design constraints of the hardware. The bus connects various circuits including one or more processors, memories, and / or hardware modules. The bus can also connect various other circuits such as peripherals, voltage regulators, power management circuits, external antennas, etc.

[0137] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Component (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically measuring lesion images, characterized in that: include: Step 1: Medical segmentation image based on the organ under test, including lesion segmentation and tissue segmentation obtained from the original medical image; Taking lesion segmentation as a target, selecting tissue segmentation adjacent to the target as a benchmark; Step 2: extracting the boundary of the reference to generate a tissue distance field; Step 3: Mark the target in the tissue distance field to obtain a target distance field; Step 4: In the target distance field, extract the first reference point and the second reference point, and use the line connecting the two as the measurement line.

2. The method according to claim 1, characterized in that The organ to be detected is a tooth.

3. The method according to claim 2, characterized in that The original medical image is a CT, CBCT or MRI image.

4. The method according to claim 3, characterized in that The tissue segmentation is tooth bone segmentation.

5. The method according to claim 1, wherein The specific method for selecting the benchmark in step 1 includes: obtaining the adjacent area of ​​the lesion segmentation and the tissue segmentation through morphological operations; taking the lesion segmentation as the target, and selecting the tissue segmentation adjacent to the target as the benchmark based on the selection strategy.

6. The method according to claim 5, characterized in that The morphological operations specifically include: Step 11: In the medical segmentation image, a region of interest is captured with the lesion segmented as the center, in a preset box and in a preset direction; Step 12: In the preset layer of the region of interest, dilation operations are performed on the lesion segmentation and the tissue segmentation according to the preset dilation values ​​to obtain the expanded overlapping area; Step 13: Take the largest overlapping area in each preset layer as the adjacent area.

7. The method according to claim 5, characterized in that The selection strategy is: within a preset range, select several tissue monomers with the longest adjacency line length as a benchmark.

8. The method according to claim 1, characterized in that The step 2 specifically includes: Step 21: extracting the boundary of the reference to obtain a reference boundary area; Step 22: Generate a tissue distance field of the full radiation image using the contour line of the reference boundary area as a starting line.

9. An automated lesion image measurement system, characterized in that: It includes a data receiving module, a data processing module and a result generating module; The data receiving module is used to receive the medical segmentation image of the organ under test, including the lesion segmentation and tissue segmentation obtained based on the original medical image; taking the lesion segmentation as the target and selecting the tissue segmentation adjacent to the target as the reference; The data processing module includes a tissue distance field unit, a target distance field unit and a measurement unit; The tissue distance field unit extracts the boundary of the reference to generate a tissue distance field; The target distance field unit marks the target in the tissue distance field to obtain a target distance field; The measuring unit extracts a first reference point and a second reference point in the target distance field and uses a line connecting the first reference point and the second reference point as a measuring line; The result generating module is used to send the measurement result of the measurement unit to external devices.

10. An automated lesion image measurement device, characterized in that: It includes a processor, a memory and a bus, the memory stores instructions and data read by the processor, the processor is used to call the instructions and data in the memory to execute the method according to any one of claims 1 to 8, and the bus connects the functional components to transmit information.

Images