Microneedle patch size determination method and system and medium

Abstract

The invention provides a microneedle patch size determination method and system and a medium, and relates to the technical field of microneedle patch size determination, and the method comprises the steps: obtaining body surface three-dimensional image data of a to-be-applied part, and segmenting the data into a plurality of sub-regions after preprocessing; performing image analysis on each sub-region, extracting local point cloud data, performing curved surface fitting, determining a main curvature coefficient of each sub-region, and further determining a fitting degree index of each sub-region; performing clustering analysis on all the sub-regions according to the fitness indexes, dividing the adjacent sub-regions of which the fitness indexes are located in the same preset interval into the same cluster, extracting local point cloud data of each cluster, performing fitting to generate a region boundary contour line, and incorporating the region boundary contour line into an initial microneedle patch contour line set; and screening and optimizing the initial microneedle patch contour line set to obtain candidate contour lines, smoothing the candidate contour lines to generate a target microneedle patch contour line, and determining an area enclosed by the target microneedle patch contour line as the substrate size of the microneedle patch.

Description

Technical Field

[0001] This invention relates to the field of microneedle patch size determination technology, specifically to a method, system, and medium for determining the size of microneedle patches. Background Technology

[0002] Microneedle patches, as a novel transdermal drug delivery carrier, consist of a base and several fine needle-like structures arranged on the base. They can penetrate the stratum corneum of the skin to form microchannels, achieving highly efficient transdermal drug delivery and showing broad application prospects in fields such as cosmetic skincare, vaccination, and chronic disease treatment. However, poor adhesion between the microneedle patch and the skin can lead to the microneedles failing to effectively penetrate the skin, reducing drug penetration, and even causing wrinkles, curling, or detachment at the application site. Therefore, determining the size of the microneedle patch based on individual skin surface morphology to ensure a close fit to the application site is a key issue that needs to be addressed in the design and application of microneedle patches.

[0003] In the prior art, the method, system, electronic device, and medium for determining the size of a facial mask (publication number CN117765586A) collects facial size data of a sample group, uses the elbow method to determine the optimal number of clusters, clusters the sample data into multiple clusters, and determines multiple sets of target facial size data based on the cluster center points of each cluster, thereby designing multiple mask sizes. This method can generate various mask sizes based on the distribution of facial features of a group, improving the adaptability of masks to different face shapes to a certain extent. However, this method only performs cluster analysis based on the geometric dimensions of facial feature data on a two-dimensional plane, without considering the curvature and local morphological changes of the skin surface in three-dimensional space. It cannot assess the risk of wrinkles that may occur when the mask or microneedle patch is applied to curved skin. For microneedle patches, the application area is not an ideal plane, but a three-dimensional curved surface with complex curvature changes. Relying solely on two-dimensional size features is insufficient to accurately reflect the actual adhesion between the microneedle patch and the skin, leading to potential problems with poor adhesion even when the determined microneedle patch size is used in practice.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, and medium for determining the size of microneedle patches, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method and system for determining the size of microneedle patches, the specific steps of which include: Step 1: Obtain the three-dimensional image data of the target object's surface area to be covered, preprocess the three-dimensional image data, and divide the preprocessed three-dimensional image data into several sub-regions, each sub-region corresponding to a local surface unit of the surface area to be covered. Step 2: Perform image analysis on each sub-region, extract the local point cloud data corresponding to each sub-region, and perform surface fitting based on the local point cloud data to determine the principal curvature coefficient of each sub-region. The principal curvature coefficient is used to characterize the curvature of the corresponding local surface unit; then, determine the fitting index of each sub-region based on the principal curvature coefficient. Step 3: Based on the fit index of each sub-region, perform cluster analysis on all sub-regions, and divide adjacent sub-regions whose fit indices are within the same preset index range into the same cluster to obtain several clusters; extract the local point cloud data of all sub-regions contained in each cluster, and fit the extracted local point cloud data to generate the region boundary contour line of each cluster, and incorporate it into the initial micro-needle patch contour line set. Step 4: Filter and optimize the initial set of microneedle patch outlines to obtain candidate outlines. Smooth the candidate outlines to generate the target microneedle patch outline, and determine the area enclosed by the target microneedle patch outline as the base size of the microneedle patch.

[0007] Furthermore, the three-dimensional image data of the body surface is acquired through a three-dimensional scanning device; the preprocessing includes denoising filtering of the three-dimensional image data of the body surface to eliminate isolated noise points, and smoothing to eliminate irregular protrusions on the surface. The preprocessed three-dimensional image data of the body surface is divided into several sub-regions. Specifically, the three-dimensional image data of the body surface is divided into a uniform grid. The three-dimensional image data of the body surface is divided into several sub-regions according to a preset grid size. Each sub-region has the same size and each sub-region corresponds to a local surface unit of the part to be covered.

[0008] Furthermore, image analysis is performed on each sub-region to extract the corresponding local point cloud data. The specific logic is as follows: A three-dimensional coordinate system is established to map each pixel in the three-dimensional image data of the body surface to a three-dimensional space. Image pixel analysis is performed on each sub-region to extract the image coordinates and depth information corresponding to all pixels within the sub-region. The X-axis and Y-axis coordinates of each pixel in the three-dimensional space are determined based on the image coordinates, and the Z-axis coordinates of each pixel in the three-dimensional space are determined based on the depth information. The X-axis, Y-axis, and Z-axis coordinates together constitute the three-dimensional space point corresponding to the pixel. All three-dimensional space points constitute the local point cloud data corresponding to the sub-region, and each three-dimensional space point uniquely corresponds to a pixel within the sub-region. Using the geometric center point of each sub-region as the query point, search the local point cloud data for the nearest point to the query point. A three-dimensional spatial point, which will be searched Three-dimensional spatial points constitute a neighborhood point set for a corresponding sub-region, and the neighborhood point set is used to characterize the local surface morphology around the center point of the sub-region; Based on the neighborhood point set, a local quadratic surface is fitted using the least squares method. The local quadratic surface is a quadratic polynomial function of local planar coordinates with the center point of the sub-region as the origin. The function consists of quadratic coefficients, linear coefficients, and a constant term. The local planar coordinates are obtained by coordinate transformation of the three-dimensional spatial coordinates of each three-dimensional point in the neighborhood point set. The function value of the quadratic polynomial function is the height value on the surface at the corresponding point position. Based on the fitted local quadratic surface parameters, calculate the first and second fundamental quantities of the surface in this sub-region. The first fundamental quantity of the surface includes... , , Used to describe the arc length and angle measurement on a curved surface, the second fundamental quantity of the curved surface includes , , , used to characterize the degree of curvature of a surface in space; Then, the principal curvature coefficients of the sub-region are obtained by solving the characteristic equation composed of the first and second fundamental quantities. The characteristic equation is a quadratic equation in one variable with the principal curvature coefficients as unknowns. The three coefficients of the equation are obtained by the combination operation of the first and second fundamental quantities, respectively. Solving the quadratic equation yields two roots, and the root with the largest absolute value is taken as the principal curvature coefficient of the subregion.

[0009] Furthermore, based on the principal curvature coefficient, the fitting index of each sub-region is determined. Specifically, for each sub-region, the principal curvature coefficient of that sub-region is obtained, and the maximum and minimum values ​​of the principal curvature coefficients of all sub-regions are obtained; the difference between the maximum and minimum values ​​is used as the denominator, and the difference between the absolute value and the minimum value of the principal curvature coefficient of that sub-region is used as the numerator. The quotient of the two is then calculated, and 1 is subtracted from the quotient. The result is used as the fitting index of that sub-region. The fit index ranges from 0 to 1. A higher fit index indicates a higher degree of fit between the microneedle patch and the body surface in that sub-region, while a lower fit index indicates a lower degree of fit.

[0010] Furthermore, cluster analysis is performed on all sub-regions based on their fit index, specifically as follows: S31, Create an empty cluster set and initialize all sub-regions to an unvisited state; S32, select any unvisited sub-region as the initial seed point for the current cluster, mark it as visited, and use the initial seed point as the current seed point; S33, Search for unvisited sub-regions adjacent to the current seed point; For each adjacent sub-region found, determine whether its fit index falls within the corresponding preset index range, and the upper and lower limits of the preset index range are respectively the fit index of the initial seed point in the current cluster plus a preset amplitude value and minus a preset amplitude value. S34, If the judgment result of an adjacent sub-region is yes, then the adjacent sub-region is marked as visited and included in the current cluster. At the same time, the adjacent sub-region is used as the new current seed point, and the search and judgment process in S33 is recursively repeated until no new adjacent sub-region that meets the conditions and has not been visited can be found in the current cluster. S35, save the current cluster to the cluster set and return to S32, until all sub-regions are marked as visited, thus obtaining several clusters composed of adjacent sub-regions whose fit index is located in the same preset index range, and calculate the mean fit index of the sub-regions in each cluster as the comprehensive fit index of the cluster. For each cluster, local point cloud data of all sub-regions contained in the cluster are extracted, and all extracted local point cloud data are merged into the overall point cloud data of the cluster. Based on the overall point cloud data, the edge points of the cluster are extracted using the convex hull algorithm, and curve fitting is performed on the edge points to generate the region boundary contour line of the cluster as an initial micro-needle patch contour line. All initial micro-needle patch contour lines are summarized to construct an initial micro-needle patch contour line set.

[0011] Furthermore, the initial microneedle patch outline set is screened and optimized, specifically including the following steps: S41, traverse each initial microneedle patch outline in the initial microneedle patch outline set. For any initial microneedle patch outline, calculate the area of ​​the region it encloses and use it as the area value of the corresponding initial microneedle patch outline. The area calculation method is as follows: project the initial microneedle patch outline along the normal of the area to be applied onto a two-dimensional plane to obtain a two-dimensional closed curve. Calculate the area of ​​the region enclosed by this two-dimensional closed curve using the polygon area calculation formula. Sort each initial microneedle patch outline according to the comprehensive adhesion index from largest to smallest, select the initial microneedle patch outline at the top of the sort as the current preferred object, and perform the following steps: S42: Determine whether the area value of the current preferred object is greater than or equal to the preset minimum effective area threshold. If yes, then use the current preferred object as a candidate contour line and end the optimization process; otherwise, proceed to S43. S43: Using the current preferred object as the reference fusion object, traverse other initial micro-needle patch outlines adjacent to it in descending order of area value. For each traversed initial micro-needle patch outline, determine whether it meets the fusion conditions with the reference fusion object. The fusion conditions include: the clusters corresponding to the traversed initial micro-needle patch outline and the reference fusion object are adjacent in spatial position, and the absolute difference of their comprehensive fit index is less than a preset difference threshold. If an initial micro-needle patch outline that is traversed satisfies the fusion condition, its corresponding cluster is merged into the cluster corresponding to the benchmark fusion object to obtain a fusion cluster. The comprehensive fit index of the fusion cluster is calculated, and the region boundary outline of the fusion cluster is generated as the updated current preferred object. S44, return to execute S42 until the optimization process ends or fusion cannot continue. When fusion cannot continue, select the next initial microneedle patch outline in the sorted order as the current preferred object, return to execute S42 until the optimization process ends or all initial microneedle patch outlines have been selected as the current preferred objects. If all initial microneedle patch outlines have been selected as the current preferred targets, but candidate outlines have not yet been determined, then other locations of the target objects will be selected as the application sites.

[0012] Furthermore, a cubic B-spline curve is fitted to the discrete point sequence on the candidate contour line, and the resulting cubic B-spline curve is used as the target microneedle patch contour line. Extract all image pixel areas enclosed by the outline of the target microneedle patch in the three-dimensional image data of the body surface, and determine the surface area corresponding to the image pixel area in three-dimensional space as the base size of the microneedle patch to be applied.

[0013] The present invention also provides a microneedle patch size determination system, which is used to perform the above-described microneedle patch size determination method, including: The image acquisition and segmentation module is used to acquire three-dimensional image data of the body surface of the target object to be applied, preprocess the three-dimensional image data of the body surface, and segment the preprocessed three-dimensional image data of the body surface into several sub-regions, each sub-region corresponding to a local surface unit of the body surface to be applied. The curvature and fit calculation module is used to perform image analysis on each sub-region, extract the local point cloud data corresponding to each sub-region, and perform surface fitting based on the local point cloud data to determine the principal curvature coefficient of each sub-region. The principal curvature coefficient is used to characterize the curvature of the corresponding local surface unit; and then, the fit index of each sub-region is determined based on the principal curvature coefficient. The clustering module is used to perform cluster analysis on all sub-regions based on the fit index of each sub-region. It divides adjacent sub-regions whose fit indices are within the same preset index range into the same cluster, thereby obtaining several clusters. It extracts the local point cloud data of all sub-regions contained in each cluster, and uses the extracted local point cloud data to fit and generate the region boundary contour line of each cluster, which is then incorporated into the initial micro-needle patch contour line set. The size determination module is used to filter and optimize the initial set of micro-needle patch outlines to obtain candidate outlines, smooth the candidate outlines to generate the target micro-needle patch outline, and determine the area enclosed by the target micro-needle patch outline as the base size of the micro-needle patch.

[0014] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for determining the size of a micropatch.

[0015] Compared with the prior art, the beneficial effects of the present invention are: First, this invention acquires three-dimensional image data of the body surface of the area to be covered, divides the image data into several sub-regions by uniform grid division, extracts local point cloud data of each sub-region and performs surface fitting, calculates the principal curvature coefficient of each sub-region, and then determines the adhesion index of each sub-region based on the principal curvature coefficient. Compared with the prior art, which only performs cluster analysis based on two-dimensional geometric dimensions, this invention introduces curvature features in three-dimensional space, which can quantitatively characterize the curvature of each local unit on the skin surface, providing a more accurate data basis for evaluating the adhesion effect of microneedle patches to the skin. Second, this invention performs cluster analysis on all sub-regions based on the fit index of each sub-region, and divides adjacent sub-regions whose fit indices are within the same preset index range into the same cluster. The local point cloud data of each cluster is extracted and fitted to generate the region boundary contour line, which is then incorporated into the initial microneedle patch contour line set. By clustering and merging adjacent sub-regions with similar fit indices, continuous regions with similar curvature characteristics can be identified, so that the generated initial contour line can reflect the curvature distribution law of the skin surface, avoiding forcibly including regions with excessive curvature differences into the same microneedle patch range, thereby reducing the risk of wrinkles during application. Third, this invention filters the initial set of microneedle patch outlines, selects outlines that meet the requirements as candidate outlines, and performs cubic B-spline curve smoothing on the candidate outlines to generate the target microneedle patch outline. The area enclosed by the target outline is determined as the base size of the microneedle patch. This allows the effective working area of ​​the microneedle patch to be maximized while ensuring adhesion. The smoothing process eliminates local abrupt changes and jagged edges in the outline, making the generated microneedle patch base shape more regular, which is convenient for subsequent processing, manufacturing, and actual application. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall method flow of the present invention; Figure 2 A scatter plot of principal curvature coefficients and fit index; Figure 3 This is a schematic diagram of the overall system modules of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0018] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] Example: Please see Figure 1-2 The present invention provides a technical solution: A method for determining the size of a microneedle patch, comprising the following steps: Step 1: Obtain the three-dimensional image data of the target object's surface area to be covered, preprocess the three-dimensional image data, and divide the preprocessed three-dimensional image data into several sub-regions, each sub-region corresponding to a local surface unit of the surface area to be covered. In this embodiment, the three-dimensional image data of the body surface is acquired through a three-dimensional scanning device; The preprocessing process is as follows: First, a statistical outlier removal algorithm is used to perform noise reduction filtering on the three-dimensional image data of the body surface: For each data point, its distance to the nearest integer is calculated. The average distance between the nearest neighbors, The range of values ​​is Assuming the average distance of all points follows a Gaussian distribution, the average distance will exceed... Points within the specified range are identified as isolated noise points and removed. and The global mean and standard deviation are used to eliminate discrete noise caused by equipment jitter or environmental interference during the scanning process. Then, the moving least squares surface fitting algorithm is used to smooth the denoised point cloud data: a local neighborhood is established with each data point as the center, and the neighborhood radius is set to 2-5 times the average spacing of the point cloud. The local surface is fitted by weighted least squares method, and the points are projected onto the fitted surface along the normal direction to correct their spatial position, thereby eliminating small protrusions and irregular fluctuations on the surface, while maintaining the geometric features and edge structure of the original surface.

[0020] The preprocessed 3D body surface image data is divided into several sub-regions. Specifically, the 3D body surface image data is divided into uniform grids, and the data is then divided according to... The grid size is evenly divided into several sub-regions, each sub-region having the same size, and each sub-region corresponds to a local surface unit of the area to be covered.

[0021] Step 1 involves acquiring and preprocessing three-dimensional image data of the skin surface at the application site. This effectively eliminates isolated noise points caused by equipment vibration or environmental interference during the scanning process, while correcting minor surface protrusions and irregular fluctuations, significantly improving the accuracy and smoothness of the original data. Based on this, the data is divided into uniform grids, dividing the complex curved surface into several sub-regions of the same size. This provides standardized processing units for subsequent local curvature analysis and fit evaluation, thus laying a reliable data foundation for accurately quantifying the bending characteristics of various regions on the skin surface.

[0022] Step 2: Perform image analysis on each sub-region, extract the local point cloud data corresponding to each sub-region, and perform surface fitting based on the local point cloud data to determine the principal curvature coefficient of each sub-region. The principal curvature coefficient is used to characterize the curvature of the corresponding local surface unit; then, determine the fitting index of each sub-region based on the principal curvature coefficient. In this embodiment, image analysis is performed on each sub-region to extract the local point cloud data corresponding to each sub-region. The specific logic is as follows: A three-dimensional coordinate system is established to map each pixel in the three-dimensional image data of the body surface to a three-dimensional space. Image pixel analysis is performed on each sub-region to extract the image coordinates and depth information corresponding to all pixels within the sub-region. The X-axis and Y-axis coordinates of each pixel in the three-dimensional space are determined based on the image coordinates, and the Z-axis coordinates of each pixel in the three-dimensional space are determined based on the depth information. The X-axis, Y-axis, and Z-axis coordinates together constitute the three-dimensional space point corresponding to the pixel. All three-dimensional space points constitute the local point cloud data corresponding to the sub-region, and each three-dimensional space point uniquely corresponds to a pixel within the sub-region. Using the geometric center point of each sub-region as the query point, search the local point cloud data for the nearest point to the query point. A three-dimensional spatial point, the The value range is 20 to 50; the search results will... Three-dimensional spatial points constitute a neighborhood point set for a corresponding sub-region, and the neighborhood point set is used to characterize the local surface morphology around the center point of the sub-region; Based on the neighborhood point set, a local quadratic surface is fitted using the least squares method. The equation of the local quadratic surface is as follows: In the formula, The local planar coordinates, with the geometric center of the sub-region as the origin, are obtained by coordinate transformation from the three-dimensional spatial coordinates of each point in the neighborhood point set. This represents the height value on the surface at the corresponding point location. , , , , , The parameters of the surface to be fitted; Based on the fitted local quadratic surface parameters, calculate the first fundamental quantity of the surface in this sub-region. , , Second basic quantity , , The specific calculation formula is as follows: The principal curvature coefficients of this sub-region can then be obtained by solving the following characteristic equation; Solving this quadratic equation yields two roots. and The root with the larger absolute value is taken as the principal curvature coefficient of the subregion. .

[0023] After completing the local quadratic surface fitting, firstly, based on the coefficients of the first term in the fitted quadratic surface parameters... , and quadratic coefficient , , Calculate the first fundamental quantity of the surface according to the principles of differential geometry. , , Second basic quantity , , The first fundamental quantity describes the arc length and angle measurement on the surface, while the second fundamental quantity characterizes the curvature of the surface in space; characteristic equations are then constructed based on these fundamental quantities. The essence of this equation is the equation satisfied by the extreme values ​​of normal curvature of a surface at a given point along different directions. The two roots obtained by solving this quadratic equation are the principal curvatures at that point, representing the maximum and minimum values ​​of the curvature of the surface at that point, respectively. Since the larger the absolute value of the principal curvature, the more severe the curvature of the surface, and since this invention needs to quantify the curvature of local surface units to evaluate the fit, the root with the larger absolute value is taken as the principal curvature coefficient of the sub-region for subsequent calculation of the fit index.

[0024] Based on the principal curvature coefficient, the fit index of each sub-region is determined, and the formula used is as follows: In the formula, For the first The fit index of each sub-region For indexes of sub-regions; Indicates the first Principal curvature coefficients of each sub-region; , These represent the maximum and minimum values ​​of the principal curvature coefficients in all subregions, respectively.

[0025] For this formula, the dependent variable Used to characterize the The fit index of each sub-region is between 0 and 1. The larger the value, the lower the curvature of the local surface unit and the better the fit between the microneedle patch and the region. The smaller the value, the higher the curvature of the region and the more likely the microneedle patch is to have problems such as poor fit, wrinkles or curling at that point. As the first The absolute value of the principal curvature coefficient of each sub-region directly reflects the degree of curvature of the local surface. From a practical physics perspective, areas with greater skin surface curvature (such as the forehead and chin) require a greater degree of deformation of the microneedle patch substrate during application to adhere to the skin. However, the substrate material has limited deformation capacity, therefore, areas with greater curvature are more difficult to adhere to and have lower adhesion. In the formula... The larger the molecule The larger it is, the more it leads to The smaller it is, the more it reflects this physical law; This formula introduces a global minimum. As a benchmark, the influence of the absolute value difference of curvature between different individuals or different parts is eliminated, making the fit index comparable. Secondly, the difference between the maximum and minimum values ​​is used as the denominator to normalize the fit index to the 0-1 range, which facilitates the unification of the evaluation criteria. Thirdly, the normalized relative curvature value is subtracted from 1, so that areas with smaller curvature (lower degree of bending) get higher fit indices. The formula takes the absolute value of the principal curvature coefficient, unifying the concave and convex directions of the surface as a measure of the degree of bending, avoiding the influence of different signs of curvature on the fit evaluation.

[0026] Table 1: Fit Index Statistics Table Please refer to the following: Figure 2 Based on the above 15 sets of data, the principal curvature coefficient and the fit index show a significant negative correlation: when the principal curvature coefficient is small, the fit index is high, indicating that the local surface curvature is lower and the fit between the microneedle patch and the body surface is better; conversely, when the principal curvature coefficient is large, the fit index is low and the fit is worse; for example, when When the minimum value of 0.08 is taken, Reaching the maximum value of 1.00 indicates that the area is nearly flat and has the best fit; while when When the maximum value of 0.45 is taken, A value dropping to the minimum of 0.00 indicates that the area has the greatest curvature and the worst fit; in terms of data distribution, It fluctuates within the range of 0.08 to 0.45. The uniform distribution between 0.00 and 1.00 reflects the significant differences in the bending characteristics of different local surface units. Cluster analysis is needed to identify continuous areas with similar adhesion to optimize the substrate size design of the microneedle patch and ensure that the effective area is maximized while maintaining adhesion.

[0027] Step 3: Based on the fit index of each sub-region, perform cluster analysis on all sub-regions, and divide adjacent sub-regions whose fit indices are within the same preset index range into the same cluster to obtain several clusters; extract the local point cloud data of all sub-regions contained in each cluster, and fit the extracted local point cloud data to generate the region boundary contour line of each cluster, and incorporate it into the initial micro-needle patch contour line set. In this embodiment, cluster analysis is performed on all sub-regions based on the fit index of each sub-region, specifically as follows: S31, Create an empty cluster set and initialize all sub-regions to an unvisited state; S32, select any unvisited sub-region as the initial seed point for the current cluster, mark it as visited, and use the initial seed point as the current seed point; S33, search for unvisited sub-regions adjacent to the current seed point; for each adjacent sub-region found, determine whether its fit index falls within the corresponding preset index range, and the upper and lower limits of the preset index range are respectively the fit index of the initial seed point in the current cluster plus or minus 0.1. S34, If the judgment result of an adjacent sub-region is yes, then the adjacent sub-region is marked as visited and included in the current cluster. At the same time, the adjacent sub-region is used as the new current seed point, and the search and judgment process in S33 is recursively repeated until no new adjacent sub-region that meets the conditions and has not been visited can be found in the current cluster. S35, save the current cluster to the cluster set and return to S32, until all sub-regions are marked as visited, thus obtaining several clusters composed of adjacent sub-regions whose fit index is located in the same preset index range, and calculate the mean fit index of the sub-regions in each cluster as the comprehensive fit index of the cluster. For each cluster, local point cloud data of all sub-regions contained in the cluster are extracted, and all extracted local point cloud data are merged into the overall point cloud data of the cluster. Based on the overall point cloud data, the edge points of the cluster are extracted using the convex hull algorithm, and curve fitting is performed on the edge points to generate the region boundary contour line of the cluster as an initial micro-needle patch contour line. All initial micro-needle patch contour lines are summarized to construct an initial micro-needle patch contour line set.

[0028] This step employs a region-growing-based clustering algorithm. Based on the fit index of each sub-region, spatially adjacent sub-regions with similar fit are merged into the same cluster. Specifically, all sub-regions are first initialized to an unvisited state, and an empty cluster set is created. Then, one unvisited sub-region is randomly selected as a seed point and marked as visited. Using this seed point as the center, adjacent unvisited sub-regions are searched outwards. For each found adjacent sub-region, its fit index is compared with the current seed point's fit index within the same preset index range. If the condition is met, the adjacent sub-region is marked as visited and added to the current cluster. The seed point is used as a new seed point to continue the recursive search until the current cluster can no longer find a new unvisited adjacent sub-region that meets the conditions. At this point, the resulting cluster is saved to the cluster set, and a new seed point is selected from the unvisited sub-regions to repeat the above process until all sub-regions are marked as visited. Finally, several clusters are obtained, which are composed of adjacent sub-regions whose fitting index is within the same preset interval. For each cluster, the local point cloud data of all the sub-regions contained therein are extracted and merged into the overall point cloud data. The convex hull algorithm is used to extract edge points and perform curve fitting to generate the region boundary contour line of the cluster, which is then included in the initial micro-needle patch contour line set.

[0029] First, by employing the dual constraints of spatial adjacency and similarity of fit index, continuous regions with similar curvature characteristics can be accurately identified, avoiding the forced inclusion of regions with excessively large curvature differences into the same microneedle patch range, thereby effectively reducing the risk of wrinkles during application. Furthermore, the recursive search method based on region growth ensures the connectivity and integrity of clusters, enabling the generated contour lines to accurately reflect the curvature distribution pattern of the skin surface. In addition, by generating independent boundary contour lines for each cluster, a rich pool of candidates is provided for subsequent selection of the contour line with the largest effective area or for contour fusion optimization, which helps to maximize the effective working area of ​​the microneedle patch while ensuring fit.

[0030] Step 4: Filter and optimize the initial set of microneedle patch outlines to obtain candidate outlines, smooth the candidate outlines to generate the target microneedle patch outline, and determine the area enclosed by the target microneedle patch outline as the base size of the microneedle patch. In this embodiment, the initial microneedle patch outline set is screened and optimized, and the specific steps include: S41, traverse each initial microneedle patch outline in the initial microneedle patch outline set. For any initial microneedle patch outline, calculate the area of ​​the region it encloses and use it as the area value of the corresponding initial microneedle patch outline. The area calculation method is as follows: project the initial microneedle patch outline along the normal of the area to be applied onto a two-dimensional plane to obtain a two-dimensional closed curve. Calculate the area of ​​the region enclosed by this two-dimensional closed curve using the polygon area calculation formula. Sort each initial microneedle patch outline according to the comprehensive adhesion index from largest to smallest, select the initial microneedle patch outline at the top of the sort as the current preferred object, and perform the following steps: S42: Determine whether the area value of the current preferred object is greater than or equal to the preset minimum effective area threshold. If yes, then use the current preferred object as a candidate contour line and end the optimization process; otherwise, proceed to S43. S43: Using the current preferred object as the reference fusion object, traverse other initial micro-needle patch outlines adjacent to it in descending order of area value. For each traversed initial micro-needle patch outline, determine whether it meets the fusion conditions with the reference fusion object. The fusion conditions include: the clusters corresponding to the traversed initial micro-needle patch outline and the reference fusion object are adjacent in spatial position, and the absolute difference of their comprehensive fit index is less than a preset difference threshold. If an initial micro-needle patch outline that is traversed satisfies the fusion condition, its corresponding cluster is merged into the cluster corresponding to the benchmark fusion object to obtain a fusion cluster. The comprehensive fit index of the fusion cluster is calculated, and the region boundary outline of the fusion cluster is generated as the updated current preferred object. S44, return to execute S42 until the optimization process ends or fusion cannot continue. When fusion cannot continue, select the next initial microneedle patch outline in the sorted order as the current preferred object, return to execute S42 until the optimization process ends or all initial microneedle patch outlines have been selected as the current preferred objects. If all initial microneedle patch outlines have been selected as the current preferred targets, but candidate outlines have not yet been determined, then other locations of the target objects will be selected as the application sites.

[0031] For step 4, firstly, all initial contour lines are traversed and the area enclosed by each contour is calculated. Then, according to the comprehensive fit index corresponding to each cluster in descending order, each contour line is judged and optimized as the current preferred object. If the area of ​​the current preferred object meets the preset minimum effective area threshold, it is directly used as a candidate contour line and the process ends. If the area does not meet the standard, the contour line is used as a benchmark, and other adjacent contour lines are traversed in descending order of area. Contour lines that are spatially adjacent and whose comprehensive fit index difference is less than the preset threshold are gradually merged. The merged contour lines are updated and the area is re-judged. If the merged area meets the standard, it is used as a candidate contour line and the process ends. If it still does not meet the standard, the next contour line with the second largest comprehensive fit index is selected as the new preferred object, and the above area judgment and fusion optimization process is repeated. If all contour lines and their fusion results cannot meet the area requirements, other positions of the target object are selected as the application sites.

[0032] The advantages of this approach are as follows: First, prioritizing the overall fit index as the selection order ensures that the area with the best fit characteristics is chosen as the starting point for the base design, fundamentally guaranteeing good adhesion between the microneedle patch and the skin. Second, using area thresholds ensures that the final base size has sufficient practical value, avoiding situations where the size is too small to effectively support the microneedle array or result in poor application. Third, employing an iterative fusion optimization mechanism maximizes the effective area while prioritizing fit, achieving a balance between fit and area optimization. Fourth, introducing dual fusion conditions of spatial adjacency and similar fit ensures that the fused area maintains good fit characteristics, avoiding a significant decrease in fit due to blindly expanding the area.

[0033] A cubic B-spline curve is fitted to the discrete point sequence on the candidate contour line, and the resulting cubic B-spline curve is used as the target microneedle patch contour line. Extract all image pixel areas enclosed by the outline of the target microneedle patch in the three-dimensional image data of the body surface, and determine the surface area corresponding to the image pixel area in three-dimensional space as the base size of the microneedle patch to be applied.

[0034] Cubic B-spline curve fitting of discrete point sequences on candidate contour lines transforms the original contours composed of discrete points into smooth, continuous parametric curves. This effectively eliminates potential local abrupt changes and jagged edges on the original contour lines, resulting in a more regular and smooth microneedle patch substrate shape. Simultaneously, cubic B-spline curves possess excellent local support and geometric invariance, allowing for fine-tuning of local areas while maintaining the overall contour shape, avoiding excessive smoothing that could lead to contour distortion. Based on this, all image pixel regions enclosed by the target contour line in the three-dimensional image data of the body surface are extracted, and the corresponding three-dimensional surface region is determined as the substrate size of the microneedle patch. This ensures that the final determined substrate shape precisely matches the surface morphology of the application site, thereby obtaining a regular and processable contour while maintaining adhesion, facilitating subsequent mold manufacturing and actual microneedle patch production.

[0035] Please see Figure 3 The present invention also provides a microneedle patch size determination system, comprising: The image acquisition and segmentation module is used to acquire three-dimensional image data of the body surface of the target object to be applied, preprocess the three-dimensional image data of the body surface, and segment the preprocessed three-dimensional image data of the body surface into several sub-regions, each sub-region corresponding to a local surface unit of the body surface to be applied. The curvature and fit calculation module is used to perform image analysis on each sub-region, extract the local point cloud data corresponding to each sub-region, and perform surface fitting based on the local point cloud data to determine the principal curvature coefficient of each sub-region. The principal curvature coefficient is used to characterize the curvature of the corresponding local surface unit; and then, the fit index of each sub-region is determined based on the principal curvature coefficient. The clustering module is used to perform cluster analysis on all sub-regions based on the fit index of each sub-region. It divides adjacent sub-regions whose fit indices are within the same preset index range into the same cluster, thereby obtaining several clusters. It extracts the local point cloud data of all sub-regions contained in each cluster, and uses the extracted local point cloud data to fit and generate the region boundary contour line of each cluster, which is then incorporated into the initial micro-needle patch contour line set. The size determination module is used to filter and optimize the initial set of micro-needle patch outlines to obtain candidate outlines, smooth the candidate outlines to generate the target micro-needle patch outline, and determine the area enclosed by the target micro-needle patch outline as the base size of the micro-needle patch.

[0036] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for determining the size of a micropatch.

[0037] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0038] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0039] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0040] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for determining the size of a microneedle patch, characterized in that, The specific steps include: Step 1: Obtain the three-dimensional image data of the target object's surface area to be covered, preprocess the three-dimensional image data, and divide the preprocessed three-dimensional image data into several sub-regions, each sub-region corresponding to a local surface unit of the surface area to be covered. Step 2: Perform image analysis on each sub-region, extract the local point cloud data corresponding to each sub-region, and perform surface fitting based on the local point cloud data to determine the principal curvature coefficient of each sub-region. The principal curvature coefficient is used to characterize the curvature of the corresponding local surface unit; then, determine the fitting index of each sub-region based on the principal curvature coefficient. Step 3: Based on the fit index of each sub-region, perform cluster analysis on all sub-regions, and divide adjacent sub-regions whose fit indices are within the same preset index range into the same cluster to obtain several clusters; extract the local point cloud data of all sub-regions contained in each cluster, and fit the extracted local point cloud data to generate the region boundary contour line of each cluster, and incorporate it into the initial micro-needle patch contour line set. Step 4: Filter and optimize the initial set of microneedle patch outlines to obtain candidate outlines. Smooth the candidate outlines to generate the target microneedle patch outline, and determine the area enclosed by the target microneedle patch outline as the base size of the microneedle patch.

2. The method for determining the size of a microneedle patch according to claim 1, characterized in that: The three-dimensional image data of the body surface is acquired by a three-dimensional scanning device; the preprocessing includes denoising filtering of the three-dimensional image data of the body surface to eliminate isolated noise points, and smoothing to eliminate irregular protrusions on the surface. The preprocessed three-dimensional image data of the body surface is divided into several sub-regions. Specifically, the three-dimensional image data of the body surface is divided into a uniform grid. The three-dimensional image data of the body surface is divided into several sub-regions according to a preset grid size. Each sub-region has the same size and each sub-region corresponds to a local surface unit of the part to be covered.

3. The method for determining the size of a microneedle patch according to claim 2, characterized in that: Image analysis is performed on each sub-region to extract the corresponding local point cloud data. The specific logic is as follows: A three-dimensional coordinate system is established to map each pixel in the three-dimensional image data of the body surface to a three-dimensional space. Image pixel analysis is performed on each sub-region to extract the image coordinates and depth information corresponding to all pixels within the sub-region. The X-axis and Y-axis coordinates of each pixel in the three-dimensional space are determined based on the image coordinates, and the Z-axis coordinates of each pixel in the three-dimensional space are determined based on the depth information. The X-axis, Y-axis, and Z-axis coordinates together constitute the three-dimensional space point corresponding to the pixel. All three-dimensional space points constitute the local point cloud data corresponding to the sub-region, and each three-dimensional space point uniquely corresponds to a pixel within the sub-region. Using the geometric center point of each sub-region as the query point, search the local point cloud data for the nearest point to the query point. A three-dimensional spatial point, which will be searched Each three-dimensional spatial point constitutes a neighborhood point set for a corresponding sub-region, and the neighborhood point set is used to characterize the local surface morphology around the center point of the sub-region; Based on the neighborhood point set, a local quadratic surface is fitted using the least squares method. The local quadratic surface is a quadratic polynomial function of local planar coordinates with the center point of the sub-region as the origin. The function consists of quadratic coefficients, linear coefficients, and a constant term. The local planar coordinates are obtained by coordinate transformation of the three-dimensional spatial coordinates of each three-dimensional point in the neighborhood point set. The function value of the quadratic polynomial function is the height value on the surface at the corresponding point position. Based on the fitted local quadratic surface parameters, calculate the first and second fundamental quantities of the surface in this sub-region. The first fundamental quantity of the surface includes... , , Used to describe the arc length and angle measurement on a curved surface, the second fundamental quantity of the curved surface includes , , , used to characterize the degree of curvature of a surface in space; Then, the principal curvature coefficients of the sub-region are obtained by solving the characteristic equation composed of the first and second fundamental quantities. The characteristic equation is a quadratic equation in one variable with the principal curvature coefficients as unknowns. The three coefficients of the equation are obtained by the combination operation of the first and second fundamental quantities, respectively. Solving the quadratic equation yields two roots, and the root with the largest absolute value is taken as the principal curvature coefficient of the subregion.

4. The method for determining the size of a microneedle patch according to claim 3, characterized in that: Based on the principal curvature coefficient, the fitting index of each sub-region is determined. Specifically, for each sub-region, the principal curvature coefficient of that sub-region is obtained, and the maximum and minimum values ​​of the principal curvature coefficients of all sub-regions are obtained. The difference between the maximum and minimum values ​​is used as the denominator, and the difference between the absolute value and the minimum value of the principal curvature coefficient of that sub-region is used as the numerator. The quotient of the two is calculated, and then 1 is subtracted from the quotient. The result is used as the fitting index of that sub-region. The fit index ranges from 0 to 1. A higher fit index indicates a higher degree of fit between the microneedle patch and the body surface in that sub-region, while a lower fit index indicates a lower degree of fit.

5. The method for determining the size of a microneedle patch according to claim 1, characterized in that: Cluster analysis was performed on all sub-regions based on their fit index, specifically as follows: S31, Create an empty cluster set and initialize all sub-regions to an unvisited state; S32, select any unvisited sub-region as the initial seed point for the current cluster, mark it as visited, and use the initial seed point as the current seed point; S33, Search for unvisited sub-regions adjacent to the current seed point; For each adjacent sub-region found, determine whether its fit index falls within the corresponding preset index range, and the upper and lower limits of the preset index range are respectively the fit index of the initial seed point in the current cluster plus a preset amplitude value and minus a preset amplitude value. S34, If the judgment result of an adjacent sub-region is yes, then the adjacent sub-region is marked as visited and included in the current cluster. At the same time, the adjacent sub-region is used as the new current seed point, and the search and judgment process in S33 is recursively repeated until no new adjacent sub-region that meets the conditions and has not been visited can be found in the current cluster. S35, save the current cluster to the cluster set and return to S32, until all sub-regions are marked as visited, thus obtaining several clusters composed of adjacent sub-regions whose fit index is located in the same preset index range, and calculate the mean fit index of the sub-regions in each cluster as the comprehensive fit index of the cluster. For each cluster, local point cloud data of all sub-regions contained in the cluster are extracted, and all extracted local point cloud data are merged into the overall point cloud data of the cluster. Based on the overall point cloud data, the edge points of the cluster are extracted using the convex hull algorithm, and curve fitting is performed on the edge points to generate the region boundary contour line of the cluster as an initial micro-needle patch contour line. All initial micro-needle patch contour lines are summarized to construct an initial micro-needle patch contour line set.

6. The method for determining the size of a microneedle patch according to claim 5, characterized in that: The initial set of microneedle patch outlines is screened and optimized, and the specific steps include: S41, Traverse each initial microneedle patch outline in the initial microneedle patch outline set. For any initial microneedle patch outline, calculate the area of ​​the region it encloses and use it as the area value of the corresponding initial microneedle patch outline. The area calculation method is as follows: Project the initial microneedle patch outline along the normal of the area to be applied onto a two-dimensional plane to obtain a two-dimensional closed curve. Calculate the area of ​​the region enclosed by the two-dimensional closed curve using the polygon area calculation formula. Sort each initial microneedle patch outline in descending order of comprehensive adhesion index, select the initial microneedle patch outline at the top of the sorted list as the current preferred object, and perform the following steps: S42: Determine whether the area value of the current preferred object is greater than or equal to the preset minimum effective area threshold. If yes, then use the current preferred object as a candidate contour line and end the optimization process; otherwise, proceed to S43. S43: Using the current preferred object as the reference fusion object, traverse other initial micro-needle patch outlines adjacent to it in descending order of area value. For each traversed initial micro-needle patch outline, determine whether it meets the fusion conditions with the reference fusion object. The fusion conditions include: the clusters corresponding to the traversed initial micro-needle patch outline and the reference fusion object are adjacent in spatial position, and the absolute difference of their comprehensive fit index is less than a preset difference threshold. If an initial micro-needle patch outline that is traversed satisfies the fusion condition, its corresponding cluster is merged into the cluster corresponding to the benchmark fusion object to obtain a fusion cluster. The comprehensive fit index of the fusion cluster is calculated, and the region boundary outline of the fusion cluster is generated as the updated current preferred object. S44, return to execute S42 until the optimization process ends or fusion cannot continue. When fusion cannot continue, select the next initial microneedle patch outline in the sorted order as the current preferred object, return to execute S42 until the optimization process ends or all initial microneedle patch outlines have been selected as the current preferred objects. If all initial microneedle patch outlines have been selected as the current preferred targets, but candidate outlines have not yet been determined, then other locations of the target objects will be selected as the application sites.

7. The method for determining the size of a microneedle patch according to claim 6, characterized in that: A cubic B-spline curve is fitted to the discrete point sequence on the candidate contour line, and the resulting cubic B-spline curve is used as the target microneedle patch contour line. Extract all image pixel areas enclosed by the outline of the target microneedle patch in the three-dimensional image data of the body surface, and determine the surface area corresponding to the image pixel area in three-dimensional space as the base size of the microneedle patch to be applied.

8. A microneedle patch size determination system, characterized in that: The microneedle patch size determination system is used to execute the microneedle patch size determination method according to any one of claims 1-7, comprising: The image acquisition and segmentation module is used to acquire three-dimensional image data of the body surface of the target object to be applied, preprocess the three-dimensional image data of the body surface, and segment the preprocessed three-dimensional image data of the body surface into several sub-regions, each sub-region corresponding to a local surface unit of the body surface to be applied. The curvature and fit calculation module is used to perform image analysis on each sub-region, extract the local point cloud data corresponding to each sub-region, and perform surface fitting based on the local point cloud data to determine the principal curvature coefficient of each sub-region. The principal curvature coefficient is used to characterize the curvature of the corresponding local surface unit; and then, the fit index of each sub-region is determined based on the principal curvature coefficient. The clustering module is used to perform cluster analysis on all sub-regions based on the fit index of each sub-region. It divides adjacent sub-regions whose fit indices are within the same preset index range into the same cluster, thereby obtaining several clusters. It extracts the local point cloud data of all sub-regions contained in each cluster, and uses the extracted local point cloud data to fit and generate the region boundary contour line of each cluster, which is then incorporated into the initial micro-needle patch contour line set. The size determination module is used to filter and optimize the initial set of micro-needle patch outlines to obtain candidate outlines, smooth the candidate outlines to generate the target micro-needle patch outline, and determine the area enclosed by the target micro-needle patch outline as the base size of the micro-needle patch.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements a method for determining the size of a micropatch as described in any one of claims 1-7.

Images