Intelligent collaborative robot system for microneedle introduction

Abstract

The invention provides an intelligent collaborative robot system for microneedle introduction, and relates to the technical field of medical puncture. The skin cuticle thickness, the surface roughness, the hue and other characteristic parameters are obtained through optical scanning and image processing technologies, and the initial puncture depth and the initial puncture frequency are predicted through a deep learning model; in the puncture process, skin humidity, impedance, tissue density and hue changes are monitored in real time, and a depth adjustment index and a frequency adjustment index are generated to adjust the initial puncture depth and the initial puncture frequency. According to the method, real-time sensing of the skin state and accurate regulation and control of puncture parameters are achieved, and the safety of microneedle treatment is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical puncture, and specifically provides an intelligent collaborative robot system for microneedle introduction. Background Art

[0002] Microneedle introduction technology has wide applications in the fields of medical aesthetics and drug delivery. However, its effectiveness and safety highly depend on the precise control of puncture parameters. In the prior art, the puncture depth and frequency of microneedles are usually set based on experience, lacking real-time monitoring of skin conditions and dynamic adjustment. As a result, problems such as over-stimulation or insufficient penetration may occur during the puncture process due to individual differences in skin stratum corneum thickness, roughness, humidity, etc. In addition, traditional methods are difficult to accurately identify the hue changes and tissue density changes on the skin surface, and cannot timely adjust the puncture parameters to adapt to the characteristics of different skin layers. These problems limit the accuracy and safety of microneedle technology, and there is an urgent need for an intelligent system that can real-time monitor skin conditions and dynamically optimize puncture parameters.

[0003] In the prior art, the published patent with the publication number CN118787850A discloses a microneedle control method, device, electronic device and storage medium, which monitors the skin resistance during the puncture process of the microneedle and determines the motor speed of the microneedle thrust motor based on the skin resistance, wherein the microneedle thrust motor is used to push the microneedle to puncture the skin; controls the microneedle thrust motor to push the microneedle to puncture based on the motor speed; monitors the actual depth of the microneedle puncture, and stops the microneedle thrust motor when the actual depth reaches the target depth, achieving the effect of ensuring that the microneedle can reach the target depth.

[0004] The main problems of the above method are: only relying on skin resistance to adjust the thrust motor, but skin resistance can only reflect the local mechanical impedance and cannot comprehensively reflect the true state of the skin, which may lead to over-puncturing or under-puncturing and affect the safety of microneedle introduction; and only adjusting the motor speed during the puncture process, without considering the skin state changes after puncture, and unable to dynamically adjust the puncture parameters according to the real-time state.

[0005] The above information disclosed in the background art section is only used to strengthen the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent collaborative robot system for microneedle introduction to solve the problems raised in the above background art.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] The intelligent collaborative robot system for microneedle introduction specifically includes:

[0009] A skin scanning module for scanning a target skin area to be micro-needle punctured and introduced, and obtaining the thickness of the skin cutin layer;

[0010] An image processing module for collecting a surface image of the target skin area, converting the image into a grayscale image to generate a first recognition image, generating a second recognition image after converting it into the HSV color space, extracting edge pixel points based on canny edge detection in the first recognition image, generating the skin surface roughness based on the number of edge pixel points and the area of the region surrounded by the edge pixel points, obtaining the hue of the pixel points in the second recognition image, and generating the hue of the target skin area based on the hues of all pixel points, denoted as the reference hue;

[0011] A model construction module for constructing a deep learning network, using the known skin cutin layer thickness and skin surface roughness as inputs, and the puncture depth and puncture frequency of the micro-needle as labels to train a puncture parameter prediction model;

[0012] A parameter generation module for inputting the skin cutin layer thickness and skin surface roughness of the target skin area into the puncture parameter prediction model to generate initial puncture parameters, where the initial puncture parameters include an initial puncture depth and an initial puncture frequency;

[0013] A frequency adjustment module for measuring the humidity and skin layer impedance on the surface of the target skin area before puncture, denoted as the reference humidity and reference impedance, and after introducing the micro-needle into the skin according to the initial puncture parameters, monitoring the humidity change and skin layer impedance change on the surface of the target skin area in real time during the monitoring time period, and generating a frequency adjustment index to adjust the initial puncture frequency in real time;

[0014] A depth adjustment module for measuring the tissue density of the target skin area before puncture, denoted as the reference density, and monitoring the tissue density and skin surface hue of the skin layer in the target skin area in real time during the monitoring time period, and generating a depth adjustment index to adjust the initial puncture depth in real time.

[0015] Further, the principle for generating the skin surface roughness is as follows:

[0016] For each pixel point in the first recognition image, the pixel point and the neighboring pixel points are respectively convolved with the horizontal direction template and the vertical direction template of the Prewitt operator to generate the gray-scale difference of the pixel point in the horizontal direction and the vertical direction, and the formula is as follows:

[0017]

[0018] where P X represents the horizontal direction template of the Prewitt operator, and P YRepresents the vertical direction template of the Prewitt operator, G x Represents the horizontal direction difference of the pixel point, G y Represents the vertical direction difference of the pixel point, (x,y) represents the coordinates of the pixel point;

[0019] According to the gray difference in the horizontal and vertical directions, generate the gradient magnitude of each pixel point, and the formula is:

[0020]

[0021] Among them, G(x,y) represents the gradient magnitude of the pixel point with coordinates (x,y), G x Represents the horizontal direction difference of the pixel point, G y Represents the vertical direction difference of the pixel point;

[0022] Preset the edge threshold, and when the gradient magnitude of the pixel point is higher than the edge threshold, it is used as an edge pixel point;

[0023] Based on all edge pixel points, obtain the number of edge pixel points and the area of the region enclosed by the outermost edge pixel points, and generate the skin surface roughness. The formula is:

[0024]

[0025] Among them, R represents the skin surface roughness, S represents the area of the region enclosed by the outermost edge pixel points, and N represents the number of edge pixel points.

[0026] Furthermore, the principle for generating the hue of the target skin area is:

[0027] The principle for converting the image to the HSV color space is:

[0028] For any point in the image, its RGB color space is (R,G,B), and its HSV color space is (H,S,V). First, normalize the values of R, G, and B:

[0029]

[0030] Among them, R0 represents the normalized red channel value, G0 represents the normalized green channel value, B0 represents the normalized blue channel value, R represents the red channel value, G represents the green channel value, and B represents the blue channel value;

[0031] Based on R0, G0, and B0, generate H, S, and V. The formula is:

[0032] V = max(R0,G0,B0)

[0033]

[0034] Among them, H, S, and V represent hue, saturation, and brightness respectively.

[0035] The formula for generating the hue of the target skin area based on the hues of all pixel points is:

[0036]

[0037] Among them, H total represents the hue of the target skin area, H i represents the hue of the i-th pixel point, i represents the index of the pixel point, n represents the number of pixel points, and atan2[] represents the four-quadrant arctangent function.

[0038] Furthermore, the principle for generating the frequency adjustment index to adjust the initial puncture frequency in real time is:

[0039] The formula for generating the frequency adjustment index is:

[0040]

[0041] Among them, k f,t represents the frequency adjustment index at time t, M t represents the humidity value at time t, t represents the index of the monitoring time, and t ∈ [0, T], T represents the length of the monitoring time period, M0 represents the reference humidity, M max represents the humidity safety threshold, Z t represents the skin layer impedance at time t, Z min represents the impedance safety threshold, Z0 represents the reference impedance, w1 and w2 respectively represent the weight coefficients of humidity and impedance, w1 + w2 = 1 and w1 > w2;

[0042] The formula for adjusting the puncture frequency is:

[0043] f t = f0 × [1 + u1 × (k f,t - 0.5)]

[0044] Among them, f t represents the adjusted puncture frequency at time t, f0 represents the initial puncture frequency, u1 represents the frequency adjustment gain coefficient, and u1 ∈ (0, 1].

[0045] Furthermore, the principle for generating the depth adjustment index to adjust the initial puncture depth in real time is:

[0046] The formula for generating the depth adjustment index is:

[0047]

[0048] Among them, kd,t represents the depth adjustment index at time t, ρ t represents the tissue density at time t, ρ0 represents the reference density, H t represents the skin surface hue at time t, H max represents the maximum hue value, H max = 360°, w3 and w4 respectively represent the weight coefficients of tissue density and hue, w3 + w4 = 1 and w3 > w4;

[0049] The formula for adjusting the puncture depth is:

[0050] d t = d0×(1 - u2×k d,t )

[0051] where, d t represents the adjusted puncture depth at time t, d0 represents the initial puncture depth, u2 represents the depth adjustment gain coefficient, u2 ∈ (0, 0.5].

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] By integrating skin scanning, image processing, and deep learning technologies, the present invention realizes the accurate measurement of the thickness of the skin stratum corneum and the skin surface roughness, predicts the initial puncture depth and puncture frequency based on the skin stratum corneum and skin roughness, ensures that the initial puncture parameters match the physiological characteristics of the target skin area, and improves the scientificity of setting the initial puncture parameters; by real-time monitoring of the humidity and impedance changes in the target skin area, it can timely adjust the puncture frequency according to the dynamic changes of the skin condition. This adaptability significantly improves the effectiveness of the treatment, avoids potential problems caused by inappropriate initial parameters, effectively reduces the error during the puncture process, avoids skin damage or unnecessary side effects caused by improper frequency, enhances the safety of the treatment, and since the humidity and impedance changes are closely related to the skin absorption ability, timely adjustment of the puncture frequency can effectively improve the drug penetration rate and enhance the treatment effect.

[0054] The present invention also measures the tissue density before puncture and real-time monitors the tissue density and skin surface hue during the treatment process, and can dynamically adjust the puncture depth according to the actual physiological characteristics of the skin. This precise control effectively avoids improper puncture caused by individual differences, improves the safety and effectiveness of the treatment. On the one hand, it ensures that the microneedle puncture reaches the ideal depth, so that the drug can penetrate more effectively into the target skin layer, which helps to improve the bioavailability of the drug. On the other hand, it can real-time monitor the tissue density changes in the skin layer, timely identify potential risk factors and make corresponding adjustments, reduce the incidence of complications, and ensure the safety of patients. Brief Description of the Drawings

[0055] Figure 1 Schematic diagram of the system module of the embodiment of the present invention. Detailed implementation manners

[0056] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0057] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should be the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0058] Embodiment:

[0059] Please refer to Figure 1 , the present invention provides a technical solution:

[0060] An intelligent collaborative robot system for microneedle introduction, specifically including:

[0061] A skin scanning module, configured to scan a target skin area that needs to be introduced by microneedle puncture, and obtain the thickness of the skin cutin layer;

[0062] In this embodiment, a target skin area that needs to be introduced by microneedle puncture is located through a high-resolution optical sensor and a confocal microscope. Near-infrared light is emitted through optical coherence tomography, and a reflected signal is received to generate a cross-sectional image. Since the cutin layer has a high reflectivity and is different from the underlying living cell layer, the target skin area is evenly divided into five equal parts in both the horizontal and vertical directions to form 25 intersection points. The thickness of the cutin layer is sampled at each intersection point, and the average value of the cutin layer thickness at the 25 intersection points is calculated as the skin cutin layer thickness.

[0063] The image processing module is used to collect the surface image of the target skin area, convert the image into a grayscale image to generate the first recognition image, generate the second recognition image after converting it to the HSV color space, extract edge pixel points based on canny edge detection in the first recognition image, generate the skin surface roughness based on the number of edge pixel points and the area of the region enclosed by the edge pixel points, obtain the hue of the pixel points in the second recognition image, and generate the hue of the target skin area based on the hues of all pixel points, denoted as the reference hue;

[0064] The principle for generating the skin surface roughness is as follows:

[0065] For each pixel point in the first recognition image, the pixel point and the neighborhood pixel points are respectively convolved with the horizontal direction template and the vertical direction template of the Prewitt operator to generate the gray - level differences in the horizontal and vertical directions of the pixel point. The formula is as follows:

[0066]

[0067] where, P X represents the horizontal direction template of the Prewitt operator, P Y represents the vertical direction template of the Prewitt operator, G x represents the horizontal direction difference of the pixel point, G y represents the vertical direction difference of the pixel point, and (x,y) represents the coordinates of the pixel point;

[0068] According to the gray - level differences in the horizontal and vertical directions, generate the gradient magnitude of each pixel point. The formula is as follows:

[0069]

[0070] where, G(x,y) represents the gradient magnitude of the pixel point with coordinates (x,y), G x represents the horizontal direction difference of the pixel point, G y represents the vertical direction difference of the pixel point;

[0071] Preset an edge threshold. When the gradient magnitude of the pixel point is higher than the edge threshold, it is regarded as an edge pixel point;

[0072] The gradient magnitude synthesizes the gray - level changes in the horizontal and vertical directions and is used to measure the edge strength at this pixel point. The larger the gradient magnitude of the pixel point, the more likely it is an edge pixel point; the edge threshold is used to distinguish between edge and non - edge pixel points.

[0073] Based on all edge pixel points, obtain the number of edge pixel points and the area of the region enclosed by the outermost - circle edge pixel points, and generate the skin surface roughness. The formula is as follows:

[0074]

[0075] Among them, R represents the skin surface roughness, S represents the area of the region enclosed by the outermost edge pixel points, and N represents the number of edge pixel points.

[0076] The skin surface roughness represents the number of pixel points within the area enclosed by the pixel points. The outermost edge pixel points reflect the actual edge of the target skin area. Within the area enclosed by the actual edge, if there are still pixel points with gradient magnitudes greater than the edge threshold, they are also recognized as edge pixel points. The reason for the generation of such edge pixel points is that the target skin area is relatively rough. During image detection, due to the unevenness within the area, there will also be pixel points with relatively high gradient magnitudes, which in turn affects the overall number of edge pixel points. The more edge pixel points there are, the more uneven the target skin area is and the higher the skin surface roughness. On the contrary, if the number of edge pixel points is smaller, it means that except for the outermost edge pixel points, there are fewer pixel points with high gradient magnitudes inside, and the corresponding skin surface roughness is lower; the skin surface roughness is inversely proportional to the area of the region enclosed by the outermost edge pixel points and directly proportional to the number of edge pixel points.

[0077] In this embodiment, the principle for generating the hue of the target skin area is as follows:

[0078] The principle for converting the image to the HSV color space is as follows:

[0079] For any point in the image, its RGB color space is (R, G, B), and its HSV color space is (H, S, V). First, normalize the values of R, G, and B:

[0080]

[0081] Among them, R0 represents the normalized red channel value, G0 represents the normalized green channel value, B0 represents the normalized blue channel value, R represents the red channel value, G represents the green channel value, and B represents the blue channel value;

[0082] Based on R0, G0, and B0, the formulas for generating H, S, and V are as follows:

[0083] V = max(R0, G0, B0)

[0084]

[0085]

[0086] Among them, H, S, and V represent hue, saturation, and brightness respectively.

[0087] The formula for generating the hue of the target skin area based on the hues of all pixel points is as follows:

[0088]

[0089] Among them, H total represents the hue of the target skin area, H i represents the hue of the i-th pixel point, i represents the index of the pixel point, n represents the number of pixel points, and atan2[] represents the four-quadrant arctangent function.

[0090] The hue is an angular value, and its value range is [0°, 360°], and 0° and 360° are equivalent. Therefore, the arithmetic mean cannot be directly used when calculating the average hue. The hue is transformed into a vector in the rectangular coordinate system through the four-quadrant arctangent function. For the hue H i of the i-th pixel point, its corresponding horizontal component and vertical component are cos(H i ), sin(H i ), respectively. Each hue is transformed into a vector and summed up, and the result is and Then, the angle between the point and the positive direction of the horizontal axis is calculated through the four-quadrant arctangent function, and the degree of the included angle is the average hue, that is, the hue of the target skin area, and it is also the reference hue before puncture.

[0091] The model construction module is used to construct a deep learning network, using the known skin cutin thickness and skin surface roughness as inputs, and the puncture depth and puncture frequency of the microneedle as labels to train the puncture parameter prediction model;

[0092] In this embodiment, the structure of the deep learning network is as follows:

[0093] Output layer: It contains 2 neurons and is used to input the skin cutin thickness and skin surface roughness;

[0094] The first hidden layer: It contains 64 neurons and uses the ReLU function for activation;

[0095] The second hidden layer: It contains 32 neurons and uses the ReLU function for activation;

[0096] Output layer: It contains 2 neurons and is used to output the puncture depth and puncture frequency.

[0097] The parameter generation module is used to input the skin cutin thickness and skin surface roughness of the target skin area into the puncture parameter prediction model to generate initial puncture parameters, and the initial puncture parameters include the initial puncture depth and the initial puncture frequency;

[0098] In this embodiment, initial puncture parameters are generated based on the thickness of the skin cutin layer and the surface roughness of the skin. When performing microneedle insertion, the skin cutin layer is penetrated first. Both the thickness of the cutin layer and the surface roughness of the skin mainly reflect the characteristics of the cutin layer. Therefore, the initial puncture parameters required to penetrate the cutin layer are determined based on the characteristics of the cutin layer, and then the initial puncture parameters are adjusted according to the changes in skin parameters during the actual insertion process.

[0099] A frequency adjustment module is configured to measure the humidity and skin layer impedance on the surface of the target skin area before puncture, denoted as the reference humidity and reference impedance. After inserting the microneedles into the skin according to the initial puncture parameters, the humidity change and skin layer impedance change on the surface of the target skin area are monitored in real time during the monitoring time period, and a frequency adjustment index is generated based on the humidity change and impedance change to adjust the initial puncture frequency in real time.

[0100] In this embodiment, the principle for generating a frequency adjustment index to adjust the initial puncture frequency in real time is as follows:

[0101] The formula for generating the frequency adjustment index is:

[0102]

[0103] where k f,t represents the frequency adjustment index at time t, M t represents the humidity value at time t, t represents the index of the monitoring time, and t ∈ [0, T], where T represents the length of the monitoring time period, M0 represents the reference humidity, and M max represents the humidity safety threshold, Z t represents the skin layer impedance at time t, and Z min represents the impedance safety threshold, Z0 represents the reference impedance, and w1 and w2 respectively represent the weight coefficients of humidity and impedance, w1 + w2 = 1 and w1 > w2.

[0104] The frequency adjustment index reflects the adjustment ratio of the puncture frequency under the influence of the surface humidity and skin layer impedance. The surface humidity reflects the amount of tissue fluid oozed out during the microneedle puncture and insertion process. The higher the humidity, the more tissue fluid oozes out. When the surface humidity approaches the safety threshold, the puncture frequency needs to be reduced to avoid excessive stimulation. Humidity is inversely proportional to the puncture frequency and thus inversely proportional to the frequency adjustment index. When the humidity is lower than the reference humidity, the skin surface is too dry, and the friction during microneedle puncture increases, which may cause difficulty in penetrating the microneedles. At this time, the puncture frequency should be increased to ensure effective penetration of the microneedles through the cutin layer.

[0105] is to normalize the skin layer impedance within the safe range. When the impedance is lower than the safety threshold Z minWhen this occurs, it may cause abnormal microneedle insertion. This could be because the microneedles have penetrated into the dermis layer, and the skin has been overly stimulated. In this case, the frequency needs to be reduced to ensure safety. When the impedance is high, it indicates that it is more difficult for the microneedles to continue penetrating, and the puncture frequency needs to be increased to ensure smooth penetration. Humidity can directly reflect whether tissue fluid has oozed out, so the weight coefficient of humidity is higher. Let w1 = 0.7 and w2 = 0.3.

[0106] The formula for adjusting the puncture frequency is:

[0107] f t = f0 × [1 + u1 × (k f,t - 0.5)]

[0108] Among them, f t represents the puncture frequency adjusted at time t, f0 represents the initial puncture frequency, u1 represents the frequency adjustment gain coefficient, and u1 ∈ (0, 1].

[0109] f t reflects the puncture frequency adjusted at time t based on the initial puncture frequency, and the puncture frequency adjusted at time t actually acts at time t + 1. Taking 0.5 as the balance point indicates that the penetration state at this time does not damage the skin and is not difficult to penetrate. When k f,t > 0.5, it is necessary to increase the puncture frequency to enhance the penetration effect. When k f,t < 0.5, it is necessary to reduce the puncture frequency to reduce the damage to the skin. The adjustment amplitude is controlled by the gain coefficient. The larger the gain coefficient u1, the more sensitive the adjustment, but oscillation may occur. The smaller the gain coefficient u1, the smoother the adjustment, but response delay may occur. In actual situations, a larger gain coefficient, such as 0.5 to 1, is taken for skin with a thicker stratum corneum, and a smaller gain coefficient, such as 0.2 to 0.5, is taken for skin with a thinner stratum corneum.

[0110] The depth adjustment module is used to measure the tissue density of the target skin area before puncture, denoted as the reference density, and to monitor the tissue density of the skin layer and the skin surface hue of the target skin area in real time during the monitoring period, and generate a depth adjustment index to adjust the initial puncture depth in real time.

[0111] In this embodiment, the principle for generating the depth adjustment index to adjust the initial puncture depth in real time is as follows:

[0112] The formula for generating the depth adjustment index is:

[0113]

[0114] Among them, k d,t represents the depth adjustment index at time t, ρ t represents the tissue density at time t, ρ0 represents the reference density, H tRepresents the skin surface hue at time t, H max Represents the maximum hue value, H max = 360°, w3 and w4 respectively represent the weight coefficients of tissue density and hue, w3 + w4 = 1 and w3 > w4;

[0115] The depth adjustment index reflects the adjustment ratio of the puncture depth under the influence of tissue density and skin surface hue, Represents the relative change rate of tissue density. When ρ t > ρ0, it indicates that the tissue density increases, meaning the microneedle enters a denser skin layer, such as the dermis layer. At this time, the puncture depth needs to be reduced to avoid excessive damage. When ρ t < ρ0, it indicates that the tissue density decreases, meaning the microneedle is in a looser stratum corneum or epidermis layer, and the penetration depth needs to be increased to ensure effective penetration. Reflects whether the hue after change is close to red. The closer the value is to 1, the closer the hue after change is to red. If it is closer to red than the reference hue, it may be caused by skin bleeding, and the puncture depth needs to be reduced. If the hue is close to the reference hue, then the value is close to 0, indicating that the skin state is normal; the depth adjustment index is inversely proportional to the tissue density and inversely proportional to the deviation between the hue after change and the red hue. The tissue density directly reflects the physical state of the skin during microneedle puncture, and the hue indirectly reflects the skin state based on whether there is bleeding. Therefore, the tissue density has a higher weight, taking w3 = 0.6 and w4 = 0.4.

[0116] The formula for adjusting the puncture depth is:

[0117] d t = d0×(1 - u2×k d,t )

[0118] Among them, d t represents the adjusted puncture depth at time t, d0 represents the initial puncture depth, u2 represents the depth adjustment gain coefficient, and u1 ∈ (0, 0.5].

[0119] d t reflects the puncture depth adjusted based on the reference puncture depth. When k d,t > 0, it indicates that the skin tissue density increases or the hue is close to red, and the puncture depth needs to be reduced. When k d,t < 0, it indicates that the skin tissue density is relatively stable and the hue is also close to the reference hue, and the puncture depth can be increased. The gain coefficient u2 reflects the sensitivity of depth adjustment. For thick stratum corneum, a larger gain coefficient is taken to quickly penetrate the dense tissue, and for thin stratum corneum, a smaller gain coefficient is taken to avoid excessive penetration into the dermis layer.

[0120] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data and performing software simulation to get a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0121] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0122] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all of them should be covered within the protection scope of this application.

Claims

1. An intelligent collaborative robot system for microneedle introduction, characterized in that, Specifically include: A skin scanning module for scanning a target skin area to be micro-needle punctured and introduced, and obtaining the thickness of the skin cutin layer; An image processing module for collecting a surface image of the target skin area, converting the image into a grayscale image to generate a first recognition image, generating a second recognition image after converting it into the HSV color space, extracting edge pixel points based on canny edge detection in the first recognition image, generating the skin surface roughness based on the number of edge pixel points and the area of the region surrounded by the edge pixel points, obtaining the hue of the pixel points in the second recognition image, and generating the hue of the target skin area based on the hue of all pixel points, denoted as the reference hue; A model construction module for constructing a deep learning network, using the known skin cutin layer thickness and skin surface roughness as inputs, and the puncture depth and puncture frequency of the micro-needle as labels to train a puncture parameter prediction model; A parameter generation module for inputting the skin cutin layer thickness and skin surface roughness of the target skin area into the puncture parameter prediction model to generate initial puncture parameters, where the initial puncture parameters include an initial puncture depth and an initial puncture frequency; A frequency adjustment module for measuring the humidity and skin layer impedance on the surface of the target skin area before puncture, denoted as the reference humidity and reference impedance, and after introducing the micro-needle into the skin according to the initial puncture parameters, monitoring the humidity change and skin layer impedance change on the surface of the target skin area in real time during the monitoring period, and generating a frequency adjustment index to adjust the initial puncture frequency in real time based on the humidity change and impedance change; A depth adjustment module for measuring the tissue density of the target skin area before puncture, denoted as the reference density, and monitoring the tissue density and skin surface hue of the skin layer of the target skin area in real time during the monitoring period, and generating a depth adjustment index to adjust the initial puncture depth in real time.

2. The intelligent collaborative robot system for microneedle introduction according to claim 1, wherein: The principle for generating the skin surface roughness in the image processing module is as follows: For each pixel point in the first recognition image, the pixel point and the neighborhood pixel points are respectively convolved with the horizontal direction template and the vertical direction template of the Prewitt operator to generate the gray-scale difference of the pixel point in the horizontal direction and the vertical direction. The formula is as follows: Among them, P X represents the horizontal direction template of the Prewitt operator, P Y represents the vertical direction template of the Prewitt operator, G x represents the horizontal direction difference of the pixel point, G y represents the vertical direction difference of the pixel point, and (x, y) represents the coordinates of the pixel point; According to the gray-scale difference in the horizontal direction and the vertical direction, the gradient amplitude of each pixel point is generated. The formula is as follows: Among them, G(x, y) represents the gradient magnitude of the pixel point with coordinates (x, y), and G x represents the horizontal direction difference of the pixel point, and G y represents the vertical direction difference of the pixel point; A preset edge threshold is set, and when the gradient amplitude of the pixel point is higher than the edge threshold, it is used as an edge pixel point; Based on all edge pixel points, the number of edge pixel points and the area of the region surrounded by the outermost edge pixel points are obtained to generate the skin surface roughness. The formula is as follows: Where, R represents the skin surface roughness, S represents the area of the region surrounded by the outermost edge pixel points, and N represents the number of edge pixel points.

3. The intelligent collaborative robot system for microneedle introduction according to claim 1, characterized in that: The principle for generating the hue of the target skin area in the image processing module is as follows: The principle for converting the image into the HSV color space is as follows: For any point in the image, its RGB color space is (R, G, B), and its HSV color space is (H, S, V). First, the values of R, G, and B are normalized: Among them, R0 represents the normalized red channel value, G0 represents the normalized green channel value, B0 represents the normalized blue channel value, R represents the red channel value, G represents the green channel value, and B represents the blue channel value; Based on R0, G0, and B0, H, S, and V are generated, and the formulas are as follows: V = max(R0, G0, B0) Among them, H, S, and V represent hue, saturation, and brightness respectively. The formula for generating the hue of the target skin area based on the hues of all pixels is: Among them, H total represents the hue of the target skin area, and H i represents the hue of the i-th pixel point, where i represents the index of the pixel point, n represents the number of pixel points, and atan2[ ] represents the four-quadrant arctangent function.

4. The intelligent collaborative robot system for microneedle introduction according to claim 1, wherein: The principle for generating the frequency adjustment index in the frequency adjustment module to adjust the initial puncture frequency in real time is: The formula for generating the frequency adjustment index is: Among them, k f,t represents the frequency adjustment index at time t, M t represents the humidity value at time t, t represents the index of the monitoring time, and t ∈ [0, T], where T represents the length of the monitoring time period, M0 represents the reference humidity, and M max represents the humidity safety threshold, Z t represents the skin layer impedance at time t, and Z min represents the impedance safety threshold, Z0 represents the reference impedance, w1 and w2 respectively represent the weight coefficients of humidity and impedance, w1 + w2 = 1 and w1 > w2; The formula for adjusting the puncture frequency is: f t = f0 × [1 + u1 × (k f,t - 0.5)] Among them, f t represents the puncture frequency adjusted at time t, f0 represents the initial puncture frequency, and u1 represents the frequency adjustment gain coefficient, where u1 ∈ (0, 1].

5. The intelligent collaborative robot system for microneedle introduction according to claim 3, characterized in that: The principle for generating the depth adjustment index in the depth adjustment module to adjust the initial puncture depth in real time is: The formula for generating the depth adjustment index is: where k d,t represents the depth adjustment index at time t, ρ t represents the tissue density at time t, ρ0 represents the reference density, H t represents the skin surface hue at time t, H max represents the maximum hue value, H max = 360°, w3 and w4 respectively represent the weight coefficients of tissue density and hue, w3 + w4 = 1 and w3 > w4; The formula for adjusting the puncture depth is: d t = d0 × (1 - u2 × k d,t ) Among them, d t represents the puncture depth adjusted at time t, d0 represents the initial puncture depth, and u2 represents the depth adjustment gain coefficient, where u2 ∈ (0, 0.5].

Images