A method, device and intelligent device for measuring skin growths

Abstract

The present application discloses a method, device and intelligent device for measuring skin warts, the method comprising: projecting a light source of a preset light mode onto the surface of the skin warts; synchronously collecting deformed images captured from different angles by an imaging device; performing data fusion and point cloud registration on the deformed images, constructing a three-dimensional model of the skin warts, and converting the three-dimensional model into a triangular mesh model; obtaining the volume, surface area, perimeter, circularity, maximum height and surface roughness of the skin warts according to the triangular mesh model, generating a measurement report of the skin warts, extracting the structural features of the skin warts from the triangular mesh model, performing disease identification on the skin warts based on the structural features and a preset disease feature library, and outputting the corresponding disease type and its confidence. In the embodiment of the present application, the three-dimensional model of the skin warts can be calculated to achieve accurate measurement of the skin warts.

Description

Technical Field

[0001] The present invention relates to the field of biological measurement, and in particular to a method, a device and an intelligent device for measuring skin growths. Background Art

[0002] Skin tags, as a common dermatological disease, widely affect people of different ages and different living backgrounds. From congenital skin tags that may appear in infants and young children to various types of tags that are highly prevalent in the elderly due to skin aging and functional decline, the onset of skin tags runs throughout the entire life cycle of a person. The presence of skin tags poses a potential threat to the patient's physical health. Some skin tags, such as warts, are caused by human papillomavirus (HPV) infection and are contagious. They can be transmitted through direct or indirect contact. Not only will they spread themselves, but they may also be transmitted to others, seriously affecting the patient's social and daily life. At the same time, some skin tags are at risk of malignant transformation, such as certain moles and seborrheic keratosis. If not treated and monitored in time, they may develop into skin cancer and threaten the patient's life and health. Therefore, accurate typing and diagnosis are key prerequisites for formulating an effective treatment plan.

[0003] At present, in clinical practice, skin tags require an experienced dermatologist to make a physical examination and make a diagnosis. If necessary, dermoscopy can be used to observe the structure below the skin surface to assist in diagnosis. However, in primary medical institutions, due to limited equipment and technical level, the accuracy and timeliness of diagnosis still need to be improved. At the same time, in the process of remote diagnosis and treatment over the Internet, the two-dimensional photos and manual measurement data provided by patients and primary physicians cannot provide complete three-dimensional morphological data and the precise size of skin tags, which can lead to misjudgment of morphological characteristics and may lead to misdiagnosis. Summary of the invention

[0004] The embodiments of the present application disclose a skin growth measurement method, apparatus and intelligent device, which can achieve accurate measurement of skin growths by calculating a three-dimensional model of the skin growths.

[0005] In a first aspect, an embodiment of the present application discloses a method for measuring skin tags, comprising: projecting a light source of a preset light pattern onto the surface of the skin tag; synchronously collecting deformed images captured from different angles through an imaging device, the deformed images being obtained based on the light source projected onto the surface of the skin tag; performing data fusion and point cloud registration on the deformed images to construct a three-dimensional model of the skin tag, and converting the three-dimensional model into a triangular mesh model; calculating the volume of each triangular facet in the triangular mesh model; accumulating the volumes of all triangular facets in the triangular mesh model to obtain the volume of the skin tag; traversing the triangular facets in the triangular mesh model, determining the three edges of each triangular facet, determining the edges that appear only in one triangular facet as boundary edges; connecting the boundary edges A closed contour is formed to obtain the perimeter of the skin tag; the triangular facets containing the boundary edges are determined as external facets; the area of ​​each external facet is calculated; the areas of the external facets are accumulated to obtain the surface area of ​​the skin tag; the circularity of the skin tag is determined based on the surface area and the perimeter; the bottom of the skin tag is determined as the reference plane, and the height of each vertex in the triangular mesh model relative to the reference plane is calculated to obtain the maximum height of the skin tag; the normal vector of each external facet is calculated; the principal curvature of the triangular mesh model is determined by analyzing the changes in adjacent normal vectors; the surface roughness of the skin tag is determined based on the principal curvature of the triangular mesh model; a measurement report of the skin tag is generated based on the volume, surface area, perimeter, circularity, maximum height and surface roughness.

[0006] In a second aspect, an embodiment of the present application discloses a skin tag measuring device, comprising: a projection unit, used to project a light source of a preset light mode onto the surface of the skin tag; a collection unit, used to synchronously collect deformed images captured from different angles through an imaging device, the deformed images being obtained based on the light source projected onto the surface of the skin tag; a construction unit, used to perform data fusion and point cloud registration on the deformed images, construct a three-dimensional model of the skin tag, and convert the three-dimensional model into a triangular mesh model; a volume unit, used to calculate the volume of each triangular facet in the triangular mesh model; the volume unit is also used to accumulate the volumes of all triangular facets in the triangular mesh model to obtain the volume of the skin tag; a perimeter unit, used to traverse the triangular facets in the triangular mesh model, determine the three edges of each triangular facet, and determine the edges that appear only in one triangular facet as boundary edges; the perimeter unit is also used to connect the boundary edges to form a closed contour to obtain the skin tag. The perimeter of the growth; a surface area unit, used to determine the triangular facet containing the boundary edge as the external facet; the surface area unit, also used to calculate the area of ​​each external facet; the surface area unit, also used to accumulate the area of ​​the external facets to obtain the surface area of ​​the skin growth; a circularity unit, used to determine the circularity of the skin growth according to the surface area and the perimeter; a maximum height unit, used to determine the bottom of the skin growth as the reference plane, calculate the height of each vertex in the triangular mesh model relative to the reference plane, and obtain the maximum height of the skin growth; a roughness unit, used to calculate the normal vector of each external facet; the roughness unit, also used to determine the principal curvature of the triangular mesh model by analyzing the changes in adjacent normal vectors; the roughness unit, also used to determine the surface roughness of the skin growth according to the principal curvature of the triangular mesh model; a generation unit, used to generate a measurement report of the skin growth according to the volume, surface area, perimeter, circularity, maximum height and surface roughness.

[0007] In a third aspect, an embodiment of the present application discloses an intelligent device, including a processor and a memory, wherein the processor calls a computer program stored in the memory to implement the method disclosed in the first aspect above.

[0008] In a fourth aspect, an embodiment of the present application discloses a computer-readable storage medium, on which a computer program or computer instructions are stored. When the computer program or computer instructions are executed by a processor, the method disclosed in the first aspect above is implemented.

[0009] In a fifth aspect, an embodiment of the present application discloses a computer program product, which includes a computer program code. When the computer program code is executed by a processor, the above method is executed.

[0010] From the above, it can be seen that the present application provides a skin tag measurement method, device and intelligent device, which projects a light source of a preset light mode onto the surface of the skin tag, collects a deformed image through an imaging device, constructs a three-dimensional model and converts it into a triangular mesh model, and then calculates parameters such as volume, surface area, circumference, circularity, maximum height and surface roughness, and generates a measurement report. By adopting three-dimensional reconstruction technology and mathematical calculations, the three-dimensional morphological characteristics of the skin tag can be accurately measured, providing objective and comprehensive evaluation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 It is a flow chart of a method for measuring skin tags disclosed in an embodiment of the present application;

[0013] Figure 2 It is a flow chart of another skin tag measurement method disclosed in an embodiment of the present application;

[0014] Figure 3 It is a flow chart of another skin tag measurement method disclosed in an embodiment of the present application;

[0015] Figure 4 It is a structural schematic diagram of a skin tag measuring device disclosed in an embodiment of the present application;

[0016] Figure 5 It is a structural diagram of a smart device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0018] The embodiments of the present application disclose a skin tag measurement method, device and intelligent device, which can accurately measure the skin tag by calculating the three-dimensional model of the skin tag.

[0019] In order to better understand the embodiments of the present application, the related technology is described below.

[0020] Skin tags are common pathological changes in the skin, which are usually not accompanied by serious symptoms, but may affect the patient's appearance and may even have a potential risk of malignant transformation in some cases. Therefore, accurate measurement of the morphological characteristics of skin tags is of great significance for diagnosis, monitoring and treatment. Traditional methods for measuring skin tags mainly rely on visual observation and ruler measurement. The above methods are not only difficult to achieve high-precision measurements, but also difficult to cope with complex morphological changes and three-dimensional structures of tags, which can easily lead to measurement errors.

[0021] With the development of medical imaging technology and computer graphics, there are some studies and products that use two-dimensional imaging combined with computer graphics technology to measure and analyze skin tumors to a certain extent. However, these methods usually have problems such as low measurement accuracy and inability to fully reflect the morphological characteristics of tumors, especially for complex three-dimensional tumors, which cannot provide sufficient morphological information.

[0022] In order to overcome the above technical difficulties, in an embodiment of the present application, a light source of a preset light mode is projected onto the surface of the skin tag; deformed images captured from different angles are synchronously collected by an imaging device, and the deformed images are obtained based on the light source projected onto the surface of the skin tag; data fusion and point cloud registration are performed on the deformed images to construct a three-dimensional model of the skin tag, and the three-dimensional model is converted into a triangular mesh model; the volume of each triangular facet in the triangular mesh model is calculated; the volume of all triangular facets in the triangular mesh model is accumulated to obtain the volume of the skin tag; the triangular facets in the triangular mesh model are traversed to determine the three edges of each triangular facet, and the edges that appear only in one triangular facet are determined as boundary edges; the boundary edges are connected to form a closed The invention discloses a method for measuring the 3D morphology of skin tags by measuring the contour of the skin tag, determining the triangular facet containing the boundary edge as the external facet, calculating the area of ​​each external facet, accumulating the area of ​​the external facets, and obtaining the surface area of ​​the skin tag, determining the circularity of the skin tag according to the surface area and the perimeter, determining the bottom of the skin tag as the reference plane, calculating the height of each vertex in the triangular mesh model relative to the reference plane, and obtaining the maximum height of the skin tag, calculating the normal vector of each external facet, determining the principal curvature of the triangular mesh model by analyzing the change of adjacent normal vectors, determining the surface roughness of the skin tag according to the principal curvature of the triangular mesh model, and generating a measurement report of the skin tag according to the volume, surface area, perimeter, circularity, maximum height and surface roughness. The present application can accurately measure the three-dimensional morphological characteristics of skin tags, overcome the problems of large subjectivity, insufficient precision, large error and inability to comprehensively evaluate complex morphology in the prior art, provide complete three-dimensional morphological data, and provide strong support for the early diagnosis of the disease, the formulation of treatment plans and the evaluation of postoperative effects.

[0023] See also Figure 1 , Figure 1FIG. 1 is a flow chart of a method for measuring skin tags disclosed in an embodiment of the present application. Figure 1 As shown, the skin tag measurement method may include the following steps.

[0024] 101. Project a light source of a preset light pattern onto the surface of the skin tag.

[0025] The light source may be a laser or a light emitting diode that can generate stable and uniform light. The preset light mode may be a stripe light mode, a dot matrix light mode, a grid pattern, etc. The specific mode may be selected according to the characteristics of the skin tag to be measured.

[0026] When a light source with a preset light pattern is projected onto the surface of a skin tag, the morphology of the skin tag surface (e.g., ups and downs) will cause the light pattern to deform. Different light patterns have different geometric characteristics, and the measurement accuracy can be improved by selecting the appropriate light pattern. For example, the stripe light pattern is suitable for skin tags with relatively smooth surface changes, and the grid light pattern is suitable for capturing skin tags with rich surface details. The density of the light pattern (e.g., the spacing of the stripes, the distribution of the light spots) can determine the resolution of the pattern. A high-resolution light pattern can capture finer surface changes.

[0027] 102. Synchronously collect deformed images captured from different angles through an imaging device, where the deformed images are obtained based on a light source projected onto the surface of the skin tag.

[0028] Skin tags may change in a few milliseconds due to slight muscle contraction or tiny movement of the epidermis. To avoid slight changes in the surface of skin tags due to tiny movements, deformations or breathing, etc., imaging devices (such as high-definition cameras) can be used to synchronously capture deformed images captured from different angles to ensure that images taken from different angles capture the same surface information at the same time, avoiding errors or inconsistencies caused by time differences, thereby improving the accuracy of point cloud registration and ensuring the accuracy of the three-dimensional model.

[0029] 103. The deformed images are subjected to data fusion and point cloud registration to construct a three-dimensional model of skin growths, and the three-dimensional model is converted into a triangular mesh model.

[0030] It can be understood that data fusion of deformed images refers to integrating deformed image information from different angles into a unified and complete 3D data set. Point cloud registration of deformed images refers to converting image data acquired from different angles into 3D point clouds and aligning them into a unified coordinate system through registration to ensure that data from multiple angles can be accurately fused.

[0031] It should be noted that, by performing data fusion and point cloud registration on the deformed image, a three-dimensional model of the skin tag is obtained. In order to facilitate subsequent morphological measurement and data analysis, the three-dimensional model can be converted into a triangular mesh model.

[0032] The triangular mesh model is composed of a group of connected triangular patches. Each triangular patch consists of three vertices (usually three-dimensional coordinates) and the corresponding boundary, representing a surface element. The triangular mesh model can subdivide the surface of the skin tag into triangles and assign spatial coordinates to each triangular patch.

[0033] 104. Calculate the volume of each triangle patch in the triangular mesh model.

[0034] 105. The volumes of all triangular facets in the triangular mesh model are accumulated to obtain the volume of the skin growth.

[0035] In some embodiments, step 104 may include:

[0036] Extract the three vertex coordinates of each triangle patch from the triangular mesh model and calculate the volume of each triangle patch. Specifically, if the triangular mesh model includes triangular patches, the triangular mesh model can be calculated based on the first formula The first The volume of a triangular face, the first formula is: ;

[0037] in, Indicates The volume of a triangular face, , , Indicates The coordinates of the three vertices of a triangle face.

[0038] In some embodiments, step 105 may include: accumulating the triangular mesh model based on the second formula The volume of a triangular face, the second formula is: ;

[0039] in, Indicates the volume of skin tags, Indicates the number of triangle patches in the triangular mesh model.

[0040] 106. Traverse the triangular patches in the triangular mesh model, determine the three edges of each triangular patch, and determine the edges that appear only in one triangular patch as boundary edges.

[0041] 107. Connect the boundary edges to form a closed contour and obtain the perimeter of the skin growth.

[0042] The perimeter of a skin tag indicates the boundary between the skin tag and normal skin, and the boundary edge of a triangular mesh model can reflect the boundary between the skin tag and normal skin. In a triangular mesh model, some edges of a face are inner boundaries (shared by adjacent faces) and some are outer boundaries (edges of only one face).

[0043] Edges bounded by only one triangle patch may be determined as boundary edges, and by identifying these boundary edges, a closed contour, ie, the boundary (perimeter) of the skin tag, may be formed by connecting these boundary edges.

[0044] The boundary (perimeter) is composed of a series of edges, each of which is a two-dimensional line segment. The length of each edge can be obtained by calculating the Euclidean distance between the two endpoints. , if the coordinates of the two endpoints are and , the length of each side can be calculated based on the eighth formula , the eighth formula is: ;

[0045] The boundary (perimeter) of the skin tag can then be calculated based on the ninth formula, which is: ;

[0046] in, Indicates the circumference of the skin tag, Indicates the number of boundary edges.

[0047] 108. Determine the triangular patch containing the boundary edge as the external patch.

[0048] In a triangular mesh model, the edges of external patches are only in contact with the outside, and the edges of internal patches are shared with other patches. Therefore, the external patches can be determined by checking whether the edges of each triangular patch are shared by other patches.

[0049] In some embodiments, the determination of external facets can be further assisted by calculating the normal of the triangular facet, and if the triangular facet includes a boundary edge and the normal faces outward, the triangular facet is confirmed to be an external facet. It can be understood that the normal faces outward refer to the normal vector pointing to the outside of the mesh.

[0050] 109. Calculate the area of ​​each external patch.

[0051] 110. Add up the areas of the external patches to get the surface area of ​​the skin growth.

[0052] In some embodiments, step 109 may include:

[0053] Based on the third formula, determine the The third formula for the area of ​​a triangle face is: ;

[0054] in, Indicates The area of ​​a triangle face, , , , , Respectively represent The three vertices of a triangle face , , The coordinates of .

[0055] In some embodiments, step 109 may include:

[0056] Based on the fourth formula, the external patch is accumulated The area of ​​the outer patch, the fourth formula is: ;

[0057] in, represents the surface area of ​​the skin tag, Indicates the number of triangle patches in the triangular mesh model.

[0058] 111. Determine the circularity of skin tags based on surface area and circumference.

[0059] In some embodiments, step 111 may include:

[0060] The circularity of the skin tag is determined based on the fifth formula, which is: ;

[0061] in, represents the circularity of the skin tag, represents the surface area of ​​the skin tag, Indicates the circumference of the skin tag.

[0062] The roundness can reflect whether the overall outline of the skin growth is regular, and can help determine whether the growth has a malignant tendency.

[0063] 112. The bottom of the skin tag is determined as a reference plane, and the height of each vertex in the triangular mesh model relative to the reference plane is calculated to obtain the maximum height of the skin tag.

[0064] The maximum height of a skin tag refers to the maximum vertical distance from the most protruding point on the tag surface to the surrounding normal skin surface (or reference plane), and is an important indicator of a skin tag. In the embodiment of the present application, the bottom of the skin tag can be determined as the reference plane. If the shape of the skin tag is complex, a reference plane can be obtained by fitting the point cloud data of the entire area, or the minimum plane in contact with the normal skin can be selected as the reference plane.

[0065] When calculating the height of each vertex in the triangular mesh model relative to the reference plane, you can first use the least squares method of the point cloud to fit a plane equation: ;

[0066] .in, are the components of the normal vector, is a plane constant.

[0067] After obtaining the above coefficients through point cloud data fitting, the reference surface equation can be obtained.

[0068] In some embodiments, step 112 may include:

[0069] Calculate each vertex based on the sixth formula The sixth formula is:

[0070] ;

[0071] in, Represents each vertex The height relative to the reference surface, Represents a vertex The coordinates of The coefficients of the plane equation representing the reference plane;

[0072] The maximum height can be determined as the maximum height of the skin tag according to the tenth formula, which is:

[0073] ;

[0074] in, Represents the vertical distance from each vertex in the triangular mesh model to the reference plane. represents the number of vertices in the mesh, Indicates the maximum height.

[0075] 113. Calculate the normal vector of each external face.

[0076] In some embodiments, the normal vector may be calculated by the cross product formula:

[0077] ;

[0078] in, and Represents the two edge vectors of the external face.

[0079] The principal curvature of the triangular mesh model is determined by analyzing the changes in adjacent normal vectors.

[0080] In some embodiments, the change of adjacent normal vectors can be analyzed by comparing the difference of normal vectors of each face with those of its adjacent faces. and its adjacent patches , the angle between two vectors can be calculated using the following formula: ;

[0081] in, and is the normal vector between two adjacent patches, and is the modulus of the normal vector. This angle can reflect the difference in normals of adjacent patches.

[0082] For each pair of adjacent facets, by calculating the change between the normal vectors, we can get a local curvature information. The angle information is converted into a curvature tensor, which can describe the local change of the normal vector. The principal curvature is the eigenvalue of the curvature tensor, which can describe the curvature degree of the surface at that point.

[0083] The principal curvature includes the maximum principal curvature and the minimum principal curvature, which correspond to the strongest and weakest curvatures of the surface. The maximum principal curvature is the strongest curvature of the surface (i.e., the maximum curvature value of the surface in this direction), which can reflect the intensity of local changes in the surface morphology, especially in areas where the surface changes rapidly (such as bulges, edges, etc.). The minimum principal curvature is the weakest curvature of the surface (i.e., the minimum curvature value of the surface in this direction), which can reflect the flatness of the surface.

[0084] 115. Determine the surface roughness of skin tumors based on the principal curvature of the triangular mesh model.

[0085] The maximum principal curvature identifies areas of a surface that are "convex" or "concave" and is used to reflect the most prominent curved parts of a surface. The minimum principal curvature is used to reflect areas of a surface that are relatively flat or less curved. For example, if the surface is convex, the maximum principal curvature corresponds to the direction of the convexity, while the minimum principal curvature describes how smooth the surface is in the lateral or parallel direction.

[0086] In some embodiments, step 115 may include:

[0087] The surface roughness of the skin tag is determined based on the seventh formula and the principal curvature. The seventh formula is: ;

[0088] in, Indicates the surface roughness of skin tags, Indicates the reference value of surface roughness. represents the strength of the relationship between the principal curvature and the surface roughness, represents the principal curvature.

[0089] when When is the maximum principal curvature, the surface roughness refers to the roughness of the most prominent part of the surface. When it is the minimum principal curvature, the surface roughness represents the roughness in the direction of smoothness or flatness.

[0090] Generate a measurement report of skin growths based on volume, surface area, circumference, circularity, maximum height, and surface roughness.

[0091] By comprehensively analyzing the above-mentioned measurement indicators of skin tags (volume, surface area, circumference, circularity, maximum height and surface roughness), a skin tag measurement report with detailed data and graphical display can be generated. The above report can not only provide basic geometric data of skin tags, but also provide doctors or researchers with accurate morphological evaluation basis.

[0092] exist Figure 1 In the described skin tag measurement method, deformed images captured from different angles are synchronously collected by an imaging device, which can reduce data loss or errors caused by different shooting angles and improve the accuracy of capturing surface features of skin tags. After data fusion and point cloud registration processing, a three-dimensional model is constructed, which can truly reflect the morphology of the skin tags and convert the three-dimensional model into a triangular mesh model to facilitate subsequent calculations and analysis. Through refined measurement methods, the various morphological characteristics of the skin tags can be comprehensively and accurately evaluated, providing a solid foundation for subsequent treatment or research. At the same time, generation of a measurement report can improve data processing efficiency and facilitate further analysis and use.

[0093] It should be understood that the same or corresponding information in the above different embodiments can be referenced to each other.

[0094] Skin tags are common abnormal growths on the surface of the skin, including various types, such as benign lesions such as skin tags, seborrheic keratosis, flat warts, and nevi, as well as precancerous lesions and malignant tumors such as actinic keratosis, Bowen's disease, basal cell carcinoma, and squamous cell carcinoma. Accurate identification of the type of skin tags is of great significance for early diagnosis of the disease, formulation of treatment plans, and prognosis assessment.

[0095] At present, the preliminary identification of skin tags mostly relies on the doctor's clinical experience and naked eye observation. The judgment is made by observing the color, shape, size, surface characteristics, etc. of the tags. However, there are differences in experience and visual perception between different people, resulting in inconsistent diagnostic results. For skin tags that are difficult to diagnose by naked eye observation, it is usually necessary to perform a biopsy or remove the tags for pathological analysis. Biopsy or removal of the tags will cause certain trauma and pain to the patient, and may lead to complications such as bleeding and scar formation.

[0096] In order to overcome the above technical difficulties, Figure 2 As shown, the following steps are also disclosed in the embodiments of the present application.

[0097] 201. Establish a disease feature database. The disease feature database is a database containing disease features of various skin growths, wherein each disease has its own unique structural features.

[0098] For example, the disease feature database includes but is not limited to the following diseases and their features:

[0099] Benign skin tags can include the following:

[0100] Skin tag: Small in size ( ), smooth surface ( ), with clear boundaries and regular shapes, commonly found in skin folds.

[0101] Seborrheic keratosis: Larger surface area ( ), surface roughness ( ), with clear borders, commonly seen in exposed areas.

[0102] Mole: clear border, regular shape, smooth surface, no obvious height change ( ).

[0103] Precancerous and malignant skin growths can include the following:

[0104] Solar keratosis: ill-defined borders (circularity ), the surface has scales (surface roughness ), commonly seen in exposed areas.

[0105] Bowen disease: clear borders but with scaling or crusting on the surface, high surface roughness ( ), grows slowly.

[0106] Basal cell carcinoma: The surface is shiny, with ulcers in the middle and raised pearly edges ( ).

[0107] It should be noted that the disease feature library can be dynamically updated based on clinical data and new research results to improve the accuracy and reliability of identification.

[0108] 202. Compare the extracted three-dimensional structural features with the data in the disease feature library.

[0109] Before the comparison, in order to ensure that the data format of the extracted three-dimensional structural features and the disease feature library is consistent, all features can be normalized, such as using minimum-maximum normalization, to improve the accuracy and computational efficiency of the comparison.

[0110] Specifically, each feature is normalized using the following formula: ;

[0111] in, represents the original eigenvalue, Indicates the minimum value of the feature in the entire database. Indicates the maximum value of the feature in the entire database. Indicates the normalized value.

[0112] For missing feature values, methods such as mean filling, median filling or interpolation can be used to ensure the integrity of all feature data.

[0113] 203. Calculate the similarity between the structural features and each disease in the preset disease feature library.

[0114] Specifically, we can first traverse the disease feature library: for each disease in the disease feature library , get its feature vector .

[0115] Then the feature vector of the skin tag to be identified can be calculated by Euclidean distance and disease feature vector The similarity between .

[0116]

[0117] in, Indicates skin growths and diseases to be identified The Euclidean distance of represents the structural feature vector of the skin tag to be identified, Represents the disease in the disease feature database The structural feature vector of Indicates the first structural eigenvalues, Indicates disease No. structural eigenvalues, Represents the total number of structural features.

[0118] It should be noted that the structural feature vector of the skin tumor to be identified It may include multiple geometric and morphological features of the skin tumor, such as volume, surface area, perimeter, circularity, maximum height, and surface roughness. For example, , It can express volume, It can represent the surface area. The structural feature vector of Can include diseases The typical values ​​of each structural feature are used to compare with the characteristics of the vegetation to be identified. , Can indicate disease The typical volume of Can mean: disease Typical surface area. Total number of structural features represents the number of different structural features contained in the feature vector. For example, if the features include volume, surface area, perimeter, circularity, maximum height, and surface roughness, then .

[0119] 204. Set a threshold for each disease. When the similarity is less than the threshold, output the corresponding disease type and its confidence.

[0120] Confidence indicates the degree of trust the system has in the matching results. For the Euclidean distance, a threshold is set. , when the similarity Less than Skin tags are considered a disease , output the corresponding disease type and its confidence.

[0121] In order to avoid misjudgment of a single feature, multiple features can be combined for comprehensive evaluation. For example, when the circularity is low and the surface roughness is high, it may indicate a malignant lesion. Based on the comparison results, the disease type of the skin tag and its confidence are output. For example, "This skin tag may be basal cell carcinoma with a confidence of 85%."

[0122] For example, the feature vector of skin tags is shown in Table 1.

[0123]

[0124] The sorted eigenvectors:

[0125] The disease feature library contains diseases and their corresponding structural features as shown in Table 2.

[0126]

[0127] The sorted disease feature vector:

[0128]

[0129]

[0130]

[0131]

[0132]

[0133] Calculate the feature vector of the skin tag to be identified by Euclidean distance and disease feature vector The similarity between .

[0134] Benign skin tags ; .

[0135] Seborrheic keratosis ; .

[0136] mole ; .

[0137] Basal cell carcinoma ; .

[0138] Squamous cell carcinoma ; .

[0139] According to the above calculation results, the similarity with the mole is 5.92. Assuming that the threshold of the mole is 10, that is, when the similarity is less than 10, the skin tumor is considered to be a mole.

[0140] Please refer to the figure, Figure 3 FIG. 2 is a flow chart of another method for measuring skin tags disclosed in an embodiment of the present application. Figure 3 As shown, the skin tag measurement method may include the following steps.

[0141] 301. Measure the skin tags multiple times and obtain measurement reports of the skin tags at different periods.

[0142] By measuring skin tags multiple times, it is possible to track their changes over time, such as changes in volume, surface area, morphology, etc., which can provide the growth trend and development pattern of skin tags. If skin tags show significant abnormal changes (such as rapid increase or mutation) over a period of time, they can be discovered in time, which helps to take treatment measures early. At the same time, during the treatment process, multiple measurement reports of skin tags can be regularly compared to determine whether the treatment plan is effective and whether the treatment strategy needs to be adjusted.

[0143] 302. Generate an evaluation report of skin tags based on multiple measurement reports.

[0144] The evaluation report generated based on multiple measurement reports can quantify the changes in skin tags and provide more scientific and objective analysis results.

[0145] See also Figure 4 , Figure 4 Schematic diagram of the structure of a skin tag measuring device disclosed in an embodiment of the present application. Figure 4 As shown, the skin tag measuring device may include.

[0146] In some embodiments, the skin tumor measuring device may also include a projection unit 401, a collection unit 402, a construction unit 403, a volume unit 404, a perimeter unit 405, a surface area unit 406, a circularity unit 407, a maximum height unit 408, a roughness unit 409 and a generation unit 410.

[0147] The projection unit 401 may project a light source of a preset light pattern onto the surface of the skin tag.

[0148] The acquisition unit 402 may synchronously acquire deformed images captured from different angles through an imaging device, where the deformed images are obtained based on a light source projected onto the surface of the skin tag.

[0149] The construction unit 403 can perform data fusion and point cloud registration on the deformed image, construct a three-dimensional model of the skin tag, and convert the three-dimensional model into a triangular mesh model.

[0150] The volume unit 404 may calculate the volume of each triangular facet in the triangular mesh model. The volume unit 404 may also accumulate the volumes of all triangular facets in the triangular mesh model to obtain the volume of the skin growth.

[0151] The perimeter unit 405 can traverse the triangular patches in the triangular mesh model, determine the three edges of each triangular patch, and determine the edges that appear only in one triangular patch as boundary edges; the perimeter unit 405 can also connect the boundary edges to form a closed contour to obtain the perimeter of the skin growth.

[0152] The surface area unit 406 may determine the triangular facets containing the boundary edges as external facets; the surface area unit 406 may also calculate the area of ​​each external facet; the surface area unit 406 may also accumulate the areas of the external facets to obtain the surface area of ​​the skin growth.

[0153] The circularity unit 407 may determine the circularity of the skin tag based on the surface area and the perimeter.

[0154] The maximum height unit 408 may determine the bottom of the skin tag as a reference plane, calculate the height of each vertex in the triangular mesh model relative to the reference plane, and obtain the maximum height of the skin tag.

[0155] The roughness unit 409 can calculate the normal vector of each external facet; the roughness unit 409 can also determine the principal curvature of the triangular mesh model by analyzing the changes in adjacent normal vectors; the roughness unit 409 can also determine the surface roughness of the skin growth based on the principal curvature of the triangular mesh model.

[0156] The generating unit 410 may generate a measurement report of the skin tumor according to the volume, surface area, perimeter, circularity, maximum height and surface roughness.

[0157] In some embodiments, the apparatus further comprises:

[0158] An extraction unit, used for extracting structural features of skin tumors from the triangular mesh model, wherein the structural features include volume, surface area, perimeter, circularity, maximum height and surface roughness;

[0159] The identification unit is used to perform disease identification on skin growths based on structural features and a preset disease feature library.

[0160] In some embodiments, the authentication unit is further configured to:

[0161] Establish a disease signature database;

[0162] Compare the extracted three-dimensional structural features with the data in the disease feature library;

[0163] Calculate the similarity between the structural features and each disease in the preset disease feature library;

[0164] A threshold is set for each disease. When the similarity is less than the threshold, the corresponding disease type and its confidence are output.

[0165] This application proposes a device that can use imaging equipment to three-dimensionally scan the morphology of skin tags on the surface of a patient's skin, accurately collect data such as the relative area, height, roughness, light transmittance, and overall spatial structure of the skin tag, and build a dedicated database for comparison and assist in diagnosis. This device is intended to improve the diagnostic accuracy of primary medical institutions, greatly improve the accuracy of data during Internet remote diagnosis and treatment, and reduce the risk of malignant transformation caused by misdiagnosis of skin tags.

[0166] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the skin tumor measuring device, projection unit 401, acquisition unit 402, construction unit 403, volume unit 404, perimeter unit 405, surface area unit 406, circularity unit 407, maximum height unit 408, roughness unit 409 and generation unit 410 described above can refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0167] In several embodiments provided in the present application, the coupling between the units may be electrical, mechanical or other forms of coupling.

[0168] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0169] Please refer to Figure 5 , Figure 5 Schematic diagram of the structure of a smart device disclosed in the embodiment of the present application. Figure 5 As shown, the smart device may include a processor 501 and a memory 502. The memory 502 may store one or more computer programs. The one or more computer programs are configured to execute the method described in the foregoing method embodiment. The memory 502 may exist independently or may be integrated with the processor 501.

[0170] The processor 501 may include one or more processing cores. The processor 501 may use various interfaces and lines to connect various parts of the entire terminal device, and may execute various functions of the terminal device and process data by running or executing instructions, programs, code sets or instruction sets stored in the memory 502, and calling data stored in the memory 502. Optionally, the processor 501 may be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor 501 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 501, but may be implemented separately through a communication chip.

[0171] The memory 502 may include a random access memory (RAM) or a read-only memory (ROM). The memory 502 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. The data storage area may also store data created by the terminal device during use (such as a phone book, audio and video data, chat record data), etc.

[0172] When the computer program instructions stored in the memory 502 are executed, the processor 501 can be used to execute various operations performed by the terminal device in the above method embodiment. The specific implementation of these operations can refer to the previous embodiments and will not be repeated here.

[0173] The present application also discloses a computer-readable storage medium, in which a computer program code is stored, and the computer program code can be called by a processor to perform various operations in the above method embodiment. The specific implementation of each of the above operations can be referred to the previous embodiment, and will not be repeated here.

[0174] The computer-readable storage medium may be an electronic memory such as a flash memory, an electrically erasable programmable read only memory (EEPROM), an erasable programmable read only memory (EPROM), a hard disk or a ROM. Alternatively, the computer-readable storage medium may include a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program codes for executing any method steps in the above method. These computer program codes may be read from or written to one or more computer program products. The computer program codes may be compressed, for example, in an appropriate form.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for measuring skin tumors, characterized in that: include: projecting a light source of a preset light pattern onto the surface of the skin tag; Synchronously collecting, by an imaging device, deformed images captured from different angles, wherein the deformed images are obtained based on a light source projected onto the surface of the skin tag; Performing data fusion and point cloud registration on the deformed image to construct a three-dimensional model of the skin tag, and converting the three-dimensional model into a triangular mesh model; Calculating the volume of each triangular facet in the triangular mesh model; Accumulating the volumes of all triangular facets in the triangular mesh model to obtain the volume of the skin tag; Traversing the triangular face patches in the triangular mesh model, determining three edges of each of the triangular face patches, and determining an edge that only appears in one triangular face patch as a boundary edge; Connecting the boundary edges to form a closed contour to obtain the perimeter of the skin tag; Determine the triangular face patch containing the boundary edge as an external face patch; Calculating the area of ​​each of the external patches; Accumulating the areas of the external face sheets to obtain the surface area of ​​the skin tag; determining a circularity of the skin tag based on the surface area and the perimeter; Determine the bottom of the skin tag as a reference plane, calculate the height of each vertex in the triangular mesh model relative to the reference plane, and obtain the maximum height of the skin tag; Calculating a normal vector for each of the exterior patches; Determining the principal curvature of the triangular mesh model by analyzing changes in adjacent normal vectors; Determining the surface roughness of the skin tag according to the principal curvature of the triangular mesh model; generating a measurement report of the skin tag based on the volume, surface area, perimeter, circularity, maximum height and surface roughness; Extracting structural features of the skin tag from the triangular mesh model, the structural features including the volume, the surface area, the perimeter, the circularity, the maximum height and the surface roughness; Based on the structural features and a preset disease feature library, disease identification is performed on the skin growths.

2. A method for measuring skin tags as claimed in claim 1, characterized in that: The disease identification of the skin tumor based on the structural features and the preset disease feature library includes: Establish a disease signature database; Compare the extracted three-dimensional structural features with the data in the disease feature library; Calculating the similarity between the structural feature and each disease in a preset disease feature library; A threshold is set for each disease, and when the similarity is less than the threshold, the corresponding disease type and its confidence level are output.

3. A method for measuring skin tags as claimed in claim 1, characterized in that: The calculating the volume of each triangular facet in the triangular mesh model comprises: The triangular mesh model is calculated based on the first formula The first The first formula for the volume of a triangular face is: ; in, Indicates The volume of a triangular face, , , Indicates The coordinates of the three vertices of a triangle face; The step of accumulating the volumes of all triangular facets in the triangular mesh model to obtain the volume of the skin tag comprises: Based on the second formula, the triangular mesh model is accumulated The second formula for the volume of a triangular face is: ; in, represents the volume of the skin tag, Indicates the number of triangle patches in the triangular mesh model.

4. A method for measuring skin tags as claimed in claim 1, characterized in that: Determining, according to the normal vector of each triangular facet in the triangular mesh model, an external facet in the triangular mesh model comprises: The normal vector of each triangular facet in the triangular mesh model is calculated, and the triangular facet with the normal vector pointing outward is determined as the external facet.

5. A method for measuring skin tags as claimed in claim 1, characterized in that: The calculating the area of ​​each of the external patches comprises: Determine the first The area of ​​a triangular face, the third formula is: ; in, Indicates The area of ​​a triangle face, , , , , Respectively represent The three vertices of a triangle face , , The coordinates of The accumulating the areas of the external face sheets to obtain the surface area of ​​the skin tag comprises: Based on the fourth formula, the external patch is accumulated The area of ​​the external patch, the fourth formula is: ; in, represents the surface area of ​​the skin tag, Indicates the number of triangle patches in the triangular mesh model.

6. A method for measuring skin tags as claimed in claim 1, characterized in that: Determining the circularity of the skin tag based on the surface area and the perimeter includes: The circularity of the skin tag is determined based on a fifth formula, which is: ; in, represents the circularity of the skin tag, represents the surface area of ​​the skin tag, Indicates the circumference of the skin tag.

7. A method for measuring skin tags as claimed in claim 1, characterized in that: The method further comprises: Measuring the skin tags multiple times to obtain measurement reports at different times; An evaluation report of the skin tag is generated based on the multiple measurement reports.

8. A skin tag measuring device, characterized in that: include: a projection unit, configured to project a light source of a preset light pattern onto the surface of the skin tag; A collection unit, configured to synchronously collect, by an imaging device, deformed images captured from different angles, wherein the deformed images are obtained based on a light source projected onto a surface of the skin tag; A construction unit, configured to perform data fusion and point cloud registration on the deformed image, construct a three-dimensional model of the skin tag, and convert the three-dimensional model into a triangular mesh model; A volume unit, used to calculate the volume of each triangular face in the triangular mesh model; The volume unit is further used to accumulate the volumes of all triangular facets in the triangular mesh model to obtain the volume of the skin tag; A perimeter unit, used for traversing the triangular facets in the triangular mesh model, determining three edges of each triangular facet, and determining an edge that only appears in one triangular facet as a boundary edge; The perimeter unit is further used to connect the boundary edges to form a closed contour to obtain the perimeter of the skin tag; A surface area unit, used to determine the triangular facet containing the boundary edge as an external facet; The surface area unit is also used to calculate the area of ​​each of the external facets; The surface area unit is further used to accumulate the areas of the external face sheets to obtain the surface area of ​​the skin tag; a circularity unit, for determining the circularity of the skin tag based on the surface area and the perimeter; A maximum height unit, used for determining the bottom of the skin tag as a reference plane, calculating the height of each vertex in the triangular mesh model relative to the reference plane, and obtaining the maximum height of the skin tag; A roughness unit for calculating a normal vector of each of said external patches; The roughness unit is further used to determine the principal curvature of the triangular mesh model by analyzing changes in adjacent normal vectors; The roughness unit is further used to determine the surface roughness of the skin tag according to the principal curvature of the triangular mesh model; a generating unit, configured to generate a measurement report of the skin tag according to the volume, surface area, perimeter, circularity, maximum height and surface roughness; an extraction unit, configured to extract structural features of the skin tag from the triangular mesh model, wherein the structural features include the volume, the surface area, the perimeter, the circularity, the maximum height and the surface roughness; An identification unit is used to perform disease identification on the skin growth based on the structural features and a preset disease feature library.

9. A smart device, characterized in that: The method comprises a processor and a memory, wherein the processor calls a computer program stored in the memory to implement the method according to any one of claims 1 to 7.

Images