Microneedle control method, device and system based on skin impedance-depth fusion

Abstract

The invention is suitable for the field of medical beauty and medical instrument technology improvement, and provides a microneedle control method, device and system based on skin impedance-depth fusion, and the method comprises the steps: collecting skin impedance and puncture depth bimodal data in real time, carrying out the data fusion through a dynamic weight distribution algorithm, and obtaining a fusion value representing a puncture state; a pre-stored impedance-propulsion parameter mapping table is dynamically matched based on the real-time impedance, and the rotating speed and the thrust of the microneedle are adjusted in real time according to the fusion value; the puncture process is divided into a contact period, a propelling period and a precise needle stopping period for staged differential control; meanwhile, a layered abnormity early warning mechanism is established, and abnormity early recognition and grading processing are achieved. According to the method, the defects of data splitting, poor static threshold adaptability, extensive control, early warning lag and the like in the prior art are overcome, the puncture depth control precision and the treatment safety are remarkably improved, and the discomfort of a patient is reduced.

Description

Technical Field

[0001] This invention belongs to the field of medical device technology, specifically relating to microneedle puncture control technology, and particularly to a microneedle control method, device and system based on skin impedance-depth fusion; applicable to skin beauty, such as water light injection treatment, gold microneedling, subcutaneous drug delivery, such as vaccines, subcutaneous injection of growth factors, superficial scar repair and other scenarios, such as atrophic scar filling puncture. Background Technology

[0002] Microneedling technology is widely used in the fields of medical aesthetics and medical puncture due to its precise subcutaneous action. In medical aesthetics, it is used for treatments such as mesotherapy and gold microneedling, while in medical applications, it is used for subcutaneous drug delivery and superficial scar repair. Precise control of the puncture depth of microneedles is crucial to ensuring treatment effectiveness and patient safety.

[0003] Existing microneedle control technology mainly adopts a "single parameter control" logic, which can be divided into two main types of solutions: one is fixed depth control, which uses mechanical travel limit or infrared depth sensor to preset a fixed puncture depth, and the drive motor pushes the microneedle to the preset depth and then stops. However, it does not take into account the skin deformation caused by the difference in skin impedance (hardness), resulting in inaccurate actual puncture depth. For example, the actual depth on hard skin may be 0.3-0.5mm shallower than the preset depth. The other is single impedance protection, which only integrates a skin impedance sensor and sets a static impedance threshold. When the impedance exceeds the limit, the machine stops to avoid excessive thrust. However, it does not combine depth data, which may lead to over-puncture on low-resistance skin due to excessively fast advancement.

[0004] The main drawbacks of existing technologies include: 1) The use of depth data and impedance data separately leads to low precision in puncture depth control, with an error of 0.3-0.6 mm; 2) The static impedance threshold cannot be adapted to diverse skin types (such as oily, dry, and sensitive skin), which can easily lead to accidental shutdown of high-impedance skin or over-puncture of low-impedance skin; 3) The puncture process uses a single advancing speed and thrust, resulting in strong pain upon contact and easy overshoot due to inertia when approaching the target depth; 4) The abnormal warning is delayed, and the machine only stops after the parameters exceed the limit, making it impossible to identify potential risks in advance, such as skin damage or microneedle tilt.

[0005] Therefore, there is an urgent need for a microneedle control solution that can integrate multimodal data, dynamically adapt control parameters, perform phased and refined control, and provide early warning of anomalies. Summary of the Invention

[0006] The purpose of this invention is to provide a microneedle control method, device and system based on skin impedance depth fusion, which aims to solve the problems of low depth accuracy, poor static threshold adaptability, high puncture risk and delayed abnormal warning caused by data fragmentation.

[0007] This invention is implemented as follows: a microneedle control method based on skin impedance depth fusion, the microneedle control method comprising the following steps: S1. Real-time acquisition of skin impedance data and microneedle puncture depth data; S2. The collected skin impedance data and puncture depth data are calculated and fused using a data fusion algorithm to obtain a fusion value F that represents the puncture state. S3. Based on the real-time collected skin impedance data, dynamically match the skin impedance range of the microneedles in the pre-stored mapping table with the advancement parameters, and adjust the speed and thrust of the microneedles in real time based on the fusion value F. S4. The puncture process is divided into three stages: the contact stage, the advancement stage, and the precise needle stop stage, and different rotation speed and thrust control logic is adopted in each stage. S5. Based on the fusion value F, a multi-level anomaly early warning mechanism is set up. When an anomaly is detected, a graded response action is executed to avoid over-puncture or under-puncture.

[0008] A further technical solution of the present invention is: the data fusion algorithm in step S2 adopts a dynamic weight allocation strategy to obtain the fusion value F, and the calculation formula of the fusion value F is F=α×Z. std +β×D std Among them, Z std To standardize skin impedance values, D std The value is the standardized depth deviation, and α and β are dynamic weighting coefficients, where α + β = 1.

[0009] A further technical solution of the present invention is as follows: In step S1, skin impedance data is obtained through a contact impedance sensor with a range of 0-2000Ω and an accuracy of ±5Ω; puncture depth data is obtained through a laser displacement sensor with a range of 0-5mm and an accuracy of ±0.02mm.

[0010] A further technical solution of the present invention is as follows: In step S4, during the contact phase, a low-speed advance is adopted, and when the impedance value is detected to drop from infinity to a set threshold, it is determined that the skin is in contact; during the advance phase, the advance is carried out according to the rotation speed of the mapping table, and the rotation speed is dynamically adjusted according to the fusion value F; during the precise needle stopping phase, the rotation speed is reduced when the actual depth reaches 90% of the target depth, and the machine is stopped when the target depth is reached and the error is ≤ ±0.01mm; in the pediatric scenario, the rotation speed and thrust are reduced throughout the entire process, and the parameters of the advance phase are increased in the scar repair scenario.

[0011] A further technical solution of the present invention is: the skin impedance range-promotion parameters in step S3 include: The mapping relationship between pre-stored skin impedance range and motor speed and thrust level; The mapping relationship is queried based on real-time impedance data to determine the corresponding initial rotational speed and thrust; The rotation speed is dynamically adjusted based on the fusion value F during the propulsion process.

[0012] A further technical solution of the present invention is: the multi-level abnormality early warning mechanism in step S5 includes three-level early warning rules: Level 1 early warning: in response to a sudden drop in impedance, reduce the rotation speed and issue a prompt; Level 2 early warning: in response to depth fluctuations, suspend propulsion and recalibrate; Level 3 early warning: in response to impedance or depth exceeding limits, immediately stop the machine and issue an alarm.

[0013] Another object of the present invention is to provide a microneedle control device based on skin impedance-depth fusion, the microneedle control device comprising: A dual-modal sensing module is used to collect skin impedance data and microneedle puncture depth data in real time; The data fusion module is used to calculate and fuse the collected skin impedance data and puncture depth data through a data fusion algorithm to obtain a fusion value F that represents the puncture state. The dynamic impedance-thrust matching module is used to dynamically match the skin impedance range and thrust parameters of the microneedles in the pre-stored mapping table according to the real-time collected skin impedance data, and adjust the speed and thrust of the microneedles in real time based on the fusion value F. The phased puncture control module is used to divide the puncture process into three stages: contact period, advancement period, and precise needle stop period, and to use differentiated speed and thrust control logic in each stage. The hierarchical anomaly warning module is used to set up a multi-level anomaly warning mechanism based on the fusion value F. When an anomaly is detected, it executes a hierarchical response action to avoid over-puncture or under-puncture.

[0014] A further technical solution of the present invention is: the microneedle control device further includes a constant motor power control module, which adopts a three-loop feedback control of current-speed-thrust. The current loop stabilizes the current through a PI controller, the speed loop adjusts the speed through a PID controller, and the thrust loop measures the thrust through a strain gauge sensor. When the thrust deviation is > ±10%, speed compensation is triggered.

[0015] A further technical solution of the present invention is: the microneedle control device further includes a sensor calibration module, which performs a calibration process weekly: impedance calibration is performed by connecting to 500Ω, 1500Ω, and 2500Ω standard resistance boxes with an error ≤ ±2%; depth calibration is performed by verifying the readings at 0.5mm, 1.0mm, and 2.0mm using a standard scale plate with an error ≤ ±0.02mm; when calibration fails, the treatment function is locked and a faulty sensor is indicated.

[0016] Another objective of this invention is to provide a microneedle control system based on skin impedance-depth fusion, the microneedle control system comprising a microneedle control device, a data storage module, an emergency treatment module, a machine learning module, and a user interface; The data storage module is used to cache the real-time treatment data of the most recent 10 seconds; The emergency handling module is used to immediately cut off the power output and make the micro needle retract to the initial position at 5mm / s when a level 3 warning is triggered, the emergency stop button is pressed, or the voltage fluctuation is greater than ±20%. The built-in supercapacitor realizes power failure data protection. The machine learning module is used to optimize control parameters based on historical data; The user interface is used to input treatment parameters and display treatment status.

[0017] The beneficial effects of this invention are: by fusing dual-modal data and dynamically assigning weights, single-parameter errors are eliminated, and the depth control accuracy is improved from ±0.05mm in the prior art. The accuracy has been improved to ±0.02mm, increasing the depth attainment rate from 85% to 99%, solving the core problem of "mechanical travel meeting the target but actual depth deviation." Dynamic impedance-propulsion parameters cover various skin types, including oily, dry, and sensitive skin, improving treatment compatibility by 40% and avoiding the defects of "high impedance shutdown and low impedance over-penetration." Phased control of rotation speed and thrust reduces patient discomfort scores by 35% and overshoot rate from 12% to 2%. Layered abnormality warning identifies abnormalities 0.3-0.5 seconds in advance, reducing the abnormal damage rate from 8% to 1%. Operation time is reduced from 3-5 minutes in existing technologies to within 1 minute, reducing reliance on professional experience and adapting to standardized clinical treatment and home use scenarios. Three-loop feedback control of the motor ensures constant power, regular sensor calibration ensures data accuracy, emergency handling and power failure protection mechanisms avoid unexpected risks, and a machine learning module enables dynamic parameter optimization. The system has high integration, reducing operation time by approximately 60%, and continuous optimization through machine learning further reduces reliance on operator experience. Attached Figure Description

[0018] Figure 1 This is a flowchart of a microneedle control method based on skin impedance depth fusion provided in an embodiment of the present invention. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0020] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] like Figure 1 The flowchart of the microneedle control method based on skin impedance depth fusion provided by the present invention is shown below, and its details are as follows: Step S1: Real-time acquisition of skin impedance data and microneedle puncture depth data; skin impedance data is acquired through a contact impedance sensor with a range of 0-2000Ω and an accuracy of ±5Ω, which can accurately reflect skin hardness. The higher the impedance value, the tougher the skin. Puncture depth data is acquired through a laser displacement sensor with a range of 0-5mm and an accuracy of ±0.02mm, which avoids the defect of infrared sensors being affected by skin color. The sampling frequency of both types of data is set to 100Hz to ensure the real-time nature of data acquisition.

[0022] Step S2: The collected skin impedance data and puncture depth data are fused using a data fusion algorithm to obtain a fusion value F representing the puncture state. The data fusion algorithm employs a dynamic weight allocation strategy, and the formula for calculating the fusion value F is: F = α × Z std +β×D std Zstd is the standardized skin impedance value, calculated as "actual impedance value / maximum range (2000Ω)", with a value ranging from 0 to 1. A higher value indicates more resilient skin. D std To standardize the depth deviation value, the calculation method is "|target depth - actual depth| / target depth", with a value range of 0-1. The larger the value, the more severe the depth deviation. α and β are dynamic weighting coefficients that satisfy α+β=1 and are adjusted in real time based on the stability of sensor data: if sweat on the skin surface causes impedance data fluctuations with a fluctuation amplitude >10%, then α is reduced (adjusted from 0.6 to 0.3) and β is increased (adjusted from 0.4 to 0.7); if slight tilting of the microneedle causes depth data fluctuations (fluctuation amplitude >5%), then β is reduced (adjusted from 0.4 to 0.2) and α is increased (adjusted from 0.6 to 0.8).

[0023] Step S3: Based on the real-time collected skin impedance data, dynamically match the skin impedance range and advancement parameters of the microneedles in the pre-stored mapping table, and adjust the speed and thrust of the microneedles in real time based on the fusion value F. The pre-stored mapping table shows the correspondence between "skin impedance range - motor speed - thrust level", with speed and thrust being positively correlated (under constant power, a decrease in speed leads to an increase in thrust). First, query the mapping table based on the real-time impedance data to determine the corresponding initial speed and thrust. During advancement, dynamically adjust the speed according to the fusion value F to ensure that the puncture state meets expectations. For example, impedance 0-800Ω corresponds to high speed and low thrust, suitable for sensitive skin; impedance 1501-2000Ω corresponds to low speed and high thrust mode, suitable for thick stratum corneum or scarred skin. The mapping table is shown below.

[0024] Step S4 divides the puncture process into three stages: contact phase, advancement phase, and precise needle stopping phase. Differentiated speed and thrust control logic is used in each stage. Contact phase: Low-speed advancement (e.g., default speed 1000 r / min) is used. When the impedance value drops from infinity to a set threshold (e.g., 1300 Ω within the 801-1500 Ω range), the microneedle is considered to be in contact with the skin. Advancement phase: Advancement is performed at the speed matched to the mapping table (e.g., 1500 r / min), and the speed is dynamically adjusted in real-time according to the fusion value F (e.g., speed reduction of 10%-20% when F>0.8). (Speed ​​increase by 10%-20% when F<0.3); Precise needle stop period: When the actual depth reaches 90% of the target depth, reduce the rotation speed (e.g., reduce to 750r / min); When the actual depth reaches the target depth and the error is ≤±0.01mm, the motor stops; Scenario adaptation: In pediatric scenarios, reduce the rotation speed and thrust throughout the entire process, tighten the needle stop threshold to reduce trauma; In scar repair scenarios, increase the parameters during the advancement phase to break through tough tissue, and relax the needle stop threshold to ensure the depth is achieved; The system supports automatic loading of parameters according to the treatment scenario and allows manual fine-tuning (fine-tuning range ±20%).

[0025] For different treatment scenarios, the three-stage control parameters are refined (taking a target depth of 1.0 mm as an example): Adaptation logic: Pediatric scenario: Reduce rotation speed and thrust throughout the entire process, and tighten the needle stop threshold to reduce trauma; Scar repair: Improve the parameters during the advancement phase to break through tough tissue, and relax the needle stopping threshold to ensure the target depth is achieved; Dynamic switching: The system automatically loads parameters according to the treatment scenario and supports manual fine-tuning (within ±20%).

[0026] Step S5: Based on the fusion value F, a multi-level anomaly warning mechanism is set. When an anomaly is detected, a graded response action is executed to avoid over-puncture or under-puncture. The multi-level anomaly warning mechanism includes three warning rules: Level 1 warning: In response to a sudden drop in impedance (such as due to local skin damage), the rotation speed is immediately reduced and a prompt is issued through the user interface; Level 2 warning: In response to depth data fluctuations exceeding the set range, the advancement action is paused and the sensor recalibration process is initiated; Level 3 warning: In response to impedance or depth exceeding the limit, the machine is immediately stopped and an alarm signal is issued.

[0027] Based on dual-modal data, a three-level early warning rule is set to identify anomalies in advance: Threshold definition Response mechanism Implementation details of sensor installation and calibration Sensor Installation Specifications Calibration process and cycle Weekly calibration: Impedance calibration: Connect to a standard resistance box (500Ω, 1500Ω, 2500Ω), and the system will automatically correct the measured value (error must be ≤ ±2%). Depth calibration: Align the micro needle with the standard scale (accuracy 0.01mm) and verify the readings at 0.5mm, 1.0mm, and 2.0mm (error must be ≤ ±0.02mm).

[0028] Calibration failure handling: The system will indicate the specific faulty sensor (e.g., "Laser depth sensor calibration out of tolerance"). The treatment function is automatically locked until calibration is successful.

[0029] Feedback regulation mechanism for constant motor power control The system employs a three-loop feedback control mechanism—current, speed, and thrust—to ensure constant motor output power (fluctuation ≤ ±5%). Feedback Link Current loop: Real-time monitoring of motor operating current (sampling frequency 1kHz), and stabilization of current output via PI controller (proportional coefficient Kp=0.8, integral time Ti=0.05s); Speed ​​loop: Calculates real-time speed based on encoder data (1024 lines / revolution resolution), and adjusts the deviation from the target speed using PID (Kp=0.5, Ki=0.1, Kd=0.02); Thrust ring: Thrust is directly measured via strain gauge sensors (range 0-1N, accuracy ±0.005N). When the thrust deviation > ±10%, speed compensation is triggered (Δspeed = K × Δthrust, K = 200r). min -1 / N).

[0030] Load change response strategies When increased skin resistance leads to increased load: The current loop prioritizes increasing the current (not exceeding 120% of the rated value); If the power cannot be maintained even when the current reaches the upper limit, the speed loop reduces the speed (reduction ≤30%) to ensure thrust; When the load suddenly drops (e.g., microneedles detach from the skin): Cut off 30% of the power output within 0.1 seconds; Trigger speed limit (maximum 1000 r / min) to prevent motor from idling and overspeeding.

[0031] Machine learning logic for optimizing treatment data feedback An optimization model is built based on historical treatment data (cumulative ≥1000 cases) to achieve dynamic parameter adaptation: Data feature dimensions Model Training and Application 1. Training logic: The Gradient Boosting Regression Tree (GBDT) model is used as input, with skin attributes and treatment parameters as input, and the "optimal thrust adjustment coefficient" is output. Models are trained separately for each treatment scenario (cosmetics / pediatrics / scars), and the model is updated every 200 cumulative data increments.

[0032] 2. Application process: Before treatment: The system calls the corresponding model based on the input skin type (e.g., "sensitive skin + face") and pre-generates thrust adjustment suggestions; During treatment: Real-time data is input into the model every 50ms to dynamically correct the thrust (correction range ≤ ±15%, to avoid drastic fluctuations); Post-treatment: Update model weights based on feedback on treatment outcomes to enhance the impact of positive cases.

[0033] Model safety mechanism Set upper and lower limits for parameter adjustment (e.g., thrust must not exceed 1.5 times the base value) to prevent abnormal model output; The right of manual veto is retained, and doctors are encouraged to manually override model recommendations.

[0034] Fault tolerance mechanisms for abnormal recovery and emergency handling Anomaly recovery process Emergency response mechanism 1. Emergency shutdown: Triggering conditions: Level 3 warning, emergency stop button pressed, power supply abnormality (voltage fluctuation > ±20%); Action: Immediately cut off the motor and RF power output, and the microneedle automatically retracts to the initial position (speed 5mm / s).

[0035] 2. Data fault tolerance: Real-time data caching (data from the last 10 seconds) automatically saves data after abnormal shutdown and supports fault tracing; Power loss protection: Built-in supercapacitor ensures that critical data (such as current depth and exception code) is written to Flash before power failure.

[0036] 3. Manual intervention: Emergency mode: Press and hold the operation button for 5 seconds to enter, supporting manual control of microneedle advance and retreat (speed limited to 30% of normal); Fault code lookup: Read detailed fault causes and troubleshooting guides through the device display screen or the accompanying APP.

[0037] It forms a closed loop in terms of parameter refinement, scenario adaptability, and safety tolerance, which can directly support equipment research and development and clinical application.

[0038] Another object of the present invention is to provide a microneedle control device based on skin impedance-depth fusion, the microneedle control device comprising: A dual-modal sensing module is used to acquire skin impedance data and microneedle puncture depth data in real time; it includes a contact impedance sensor and a laser displacement sensor, and the sensor parameters are consistent with the data acquisition step (S1).

[0039] The data fusion module is used to calculate and fuse the collected skin impedance data and puncture depth data through a data fusion algorithm to obtain a fusion value F that represents the puncture state, and realizes dynamic weight allocation and fusion value calculation. The dynamic impedance-thrust matching module is used to dynamically match the skin impedance range and thrust parameters of the microneedles in the pre-stored mapping table according to the real-time collected skin impedance data, and adjust the speed and thrust of the microneedles in real time based on the fusion value F. The phased puncture control module divides the puncture process into three phases: contact phase, advancement phase, and precise needle stop phase. It employs differentiated speed and thrust control logic in each phase, outputting speed and thrust control signals for each phase, and supports scenario-based parameter adaptation and manual fine-tuning.

[0040] The hierarchical anomaly warning module is used to set up a multi-level anomaly warning mechanism based on the fusion value F. When an anomaly is detected, it executes a hierarchical response action to avoid over-puncture or under-puncture.

[0041] The microneedle control device also includes a constant motor power control module, employing a three-loop feedback control system of current, speed, and thrust to ensure constant motor output power (fluctuation ≤ ±5%). The current loop monitors the motor operating current in real time (sampling frequency 1kHz) and stabilizes the current output through a PI controller (proportional coefficient Kp=0.8, integral time Ti=0.05s). The speed loop calculates the real-time speed based on encoder data (resolution 1024 lines / revolution), and the deviation from the target speed is adjusted by a PID controller (Kp=0.5, Ki=0.1, Kd=0.02) to control the speed. The thrust loop measures the thrust through a strain gauge sensor (range 0-1N, accuracy ±0.005N). When the thrust deviation is > ±10%, speed compensation is triggered (Δspeed = K × Δthrust, K=200r). min - ¹ / N).

[0042] The microneedle control device also includes a sensor calibration module, used to perform periodic sensor calibration procedures to ensure data acquisition accuracy; the calibration procedure is performed weekly: impedance calibration is performed by connecting to 500Ω, 1500Ω, and 2500Ω standard resistance boxes, with an error ≤ ±2%; depth calibration is performed by verifying readings at 0.5mm, 1.0mm, and 2.0mm using a standard scale plate (accuracy 0.01mm), with an error ≤ ±0.02mm; if calibration fails, the treatment function is locked and a faulty sensor is indicated, specifying the faulty sensor (e.g., "laser depth sensor calibration error"), and the treatment function is automatically locked until calibration is successful.

[0043] Another objective of this invention is to provide a microneedle control system based on skin impedance-depth fusion, the microneedle control system comprising a microneedle control device, a data storage module, an emergency treatment module, a machine learning module, and a user interface; The data storage module is used to cache real-time treatment data from the last 10 seconds, including skin impedance data, puncture depth data, fusion value F, advancement parameters, etc., and supports fault tracing after abnormal shutdown.

[0044] The emergency response module is used to deal with emergencies and ensure treatment safety. When a level 3 warning is triggered, the emergency stop button is pressed, or the power supply voltage fluctuates by more than ±20%, the power output is immediately cut off and the microneedle retracts to the initial position at 5mm / s. The built-in supercapacitor ensures that critical data (such as current depth and abnormal code) is written to the Flash storage before power failure. The machine learning module is used to optimize control parameters based on historical data and improve adaptability. Model construction: Gradient boosting regression tree (GBDT) model is used, and the model is trained separately according to the treatment scenario (medical aesthetics / pediatrics / scars); Data characteristics: Input skin attributes (such as skin type, treatment area) + treatment parameters (such as target depth), output "optimal thrust adjustment coefficient"; Training and Update: Initial training begins after accumulating ≥1000 historical data points, and the model is updated every 200 data points. Application process: Before treatment, the corresponding model is called based on the input skin type to pre-generate thrust adjustment suggestions; during treatment, real-time data is input into the model every 50ms to dynamically correct the thrust (correction range ≤ ±15%); after treatment, the model weights are updated based on the effect feedback. Safety mechanisms: Set upper and lower limits for parameter adjustment (e.g., thrust must not exceed 1.5 times the base value), retain the right of manual veto, and support doctors to manually override model suggestions.

[0045] The user interface is used to input treatment parameters (such as target depth and treatment scenario), display treatment status (such as current stage, real-time depth, and impedance value), present warning information and fault codes, and support manual operation by the user (such as starting / stopping treatment and fine-tuning parameters).

[0046] Example 1: Microneedle Control Method This embodiment provides a microneedle control method based on skin impedance depth fusion. Taking medical aesthetic mesotherapy (target depth 1.8mm) as an example, its workflow is as follows: 1. System Initialization and Parameter Settings The user inputs treatment parameters through the interactive interface: target depth 1.8mm, treatment scenario "medical aesthetic water light injection"; after the system starts, it automatically calls the "medical aesthetic scenario - skin impedance range - advancement parameter" mapping table, starts the dual-modal sensing module, and sets the sampling frequency to 100Hz; the sensor calibration module automatically performs self-check (if more than 7 days have passed since the last calibration, it prompts the user to perform weekly calibration).

[0047] 2. Data Acquisition and Fusion (S1-S2) The dual-modal sensing module acquires data in real time: a contact impedance sensor obtains skin impedance data (e.g., 1320Ω), and a laser displacement sensor obtains puncture depth data; the data fusion module calculates the standardized value: Z. std =1320Ω / 2000Ω=0.66, if the current depth is 0.9mm, then D std =|1.8mm-0.9mm| / 1.8mm=0.5; Since the sensor data has no significant fluctuations, the dynamic weights α=0.6, β=0.4, and the fusion value F=0.6×0.66+0.4×0.5=0.596.

[0048] 3. Staged puncture control and dynamic adjustment (S3-S4) Contact phase: The motor propels at the default low speed of 1000r / min. When the impedance sensor detects that the impedance drops from infinity to 1300Ω (801-1500Ω range), it determines that the system has made contact with the skin and switches to the propulsion phase. Advancement phase: Based on the impedance data of 1320Ω, the matching rotation speed of 1500r / min is obtained by referring to the mapping table. Based on the fusion value F=0.596 (which is within the normal range of 0.3-0.8), this rotation speed is maintained, and the data is monitored and the fusion value is updated in real time. Precise needle stop period: When the laser displacement sensor detects that the actual depth reaches 1.62mm (1.8mm×90%), the system will reduce the rotation speed to 750r / min; if it continues to advance to the actual depth of 1.8mm and the error is +0.008mm (≤±0.01mm), the motor will stop immediately.

[0049] 4. Anomaly Monitoring and Early Warning (S5) During treatment, if local sweating causes impedance data fluctuations of up to 12%, the data fusion module automatically adjusts the dynamic weights: α=0.3, β=0.7, and recalculates the fusion value. If a sudden drop in impedance is detected (e.g., from 1320Ω to 600Ω), a first-level warning is triggered, the rotation speed is reduced to 500r / min, and the interface displays "Local skin abnormality, it is recommended to check the skin condition." If the depth data fluctuation reaches 8% (microneedle slight tilt), a second-level warning is triggered, the advancement is paused, the laser displacement sensor is automatically recalibrated, and advancement resumes after calibration.

[0050] 5. Treatment completion and data storage After the motor stops, the data storage module caches the treatment data of the last 10 seconds (including target depth, actual depth, average impedance, rotation speed at each stage, etc.); the interactive interface displays "treatment completed", and the machine learning module incorporates the treatment data (skin type "neutral skin - face", control parameters, effect feedback) into the model training dataset for subsequent parameter optimization.

[0051] 6. Emergency Response Scenarios If the user presses the emergency stop button during treatment, the emergency handling module immediately cuts off the motor and radio frequency power output, and the microneedle retracts to its initial position at a speed of 5mm / s; the built-in supercapacitor ensures that treatment data is written to Flash, and the interface displays the fault code "Emergency Stop Triggered" and recovery instructions.

[0052] Other scenario adaptation instructions In pediatric settings (subcutaneous injection of vaccine, target depth 0.5mm): The system reduces rotation speed (800r / min during contact, 1000r / min during propulsion, and 400r / min during needle stop) and thrust throughout the entire process, tightening the needle stop threshold (error ≤ ±0.005mm). Scar repair scenario (target depth 2.0mm): Increase the rotation speed to 1800r / min during the advancement phase and relax the needle stopping threshold (error ≤ ±0.015mm) to ensure breakthrough of tough scar tissue.

[0053] Example 2: Microneedle Control Device and System This embodiment provides a microneedle control device and system for implementing the above method.

[0054] The microneedle control device includes: Dual-modal sensing module: includes a contact impedance sensor (range 0-2000Ω, accuracy ±5Ω, mounted at the root of the microneedle array) and a laser displacement sensor (range 0-5mm, accuracy ±0.02mm, mounted at a 45° angle to the side of the probe).

[0055] Data fusion module: Implemented by a microprocessor, it executes a dynamic weight allocation algorithm and calculates the fusion value F.

[0056] Dynamic impedance-thrust matching module: It stores a mapping table between impedance range and motor speed / thrust, and can output adjustment commands based on real-time impedance and fusion value F.

[0057] The phased puncture control module calls different control logics based on the current puncture depth's phase (contact phase, advancement phase, and precise needle cessation phase).

[0058] Layered anomaly warning module: Set three levels of warning rules and their response strategies.

[0059] The motor power constant control module adopts a three-loop feedback structure consisting of a current loop (PI control), a speed loop (PID control), and a thrust loop (strain gauge feedback) to ensure stable power.

[0060] Sensor calibration module: Automatic or manual prompts to perform calibration weekly. Impedance calibration uses standard resistance boxes (500Ω, 1500Ω, 2500Ω), and depth calibration uses standard scale plates (0.5mm, 1.0mm, 2.0mm). Treatment is prohibited if calibration fails.

[0061] The microneedle control system also includes: Data storage module: caches real-time data from the last 10 seconds for fault tracing.

[0062] Emergency handling module: In the event of a level 3 warning, emergency stop button press, or power failure, the power is immediately cut off, the microneedle is controlled to retract at a speed of 5mm / s, and critical data is saved at the moment of power failure using a supercapacitor.

[0063] Machine learning module: Based on accumulated treatment data (skin attributes, treatment parameters, effect feedback), it uses models such as GBDT for training, dynamically optimizes the thrust adjustment coefficient, and provides initial parameter suggestions for different scenarios.

[0064] User interface: Touch screen, used to set treatment parameters (target depth, scene mode), display real-time data, warning information and treatment status.

[0065] Sensor installation and calibration details: The impedance sensor is sealed and fixed to the base of the microneedle with epoxy resin, insulated from the microneedle (impedance ≥10MΩ), and located approximately 5mm from the needle tip. The laser depth sensor is secured with metal clips and shock-absorbing pads, with its light spot focused approximately 0.5mm directly above the puncture point. Calibration is recommended weekly to ensure measurement accuracy.

[0066] Anomaly recovery mechanism: Level 1 warning recovery: After the abnormal conditions (such as a sudden drop in impedance) disappear, the original parameters are automatically restored. If there are no abnormalities after 3 seconds of continuous monitoring, the warning is lifted.

[0067] Level 2 warning recovery: Pause propulsion for 0.5 seconds, recalibrate the depth zero point, then restart propulsion at 80% of the original rotation speed, and verify subsequent depth accuracy.

[0068] Level 3 warning recovery: Manual intervention is required. After confirming that the abnormality has been eliminated, system reset and full parameter calibration are performed. Treatment can only continue after passing the test.

[0069] Through the detailed implementation scheme combining the above-described hardware and software, this invention ensures high precision, high safety, and a good user experience in the microneedle puncture process, enabling those skilled in the art to implement this invention based on the teachings of this specification.

[0070] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A microneedle control method based on skin impedance-depth fusion, characterized in that, The microneedle control method includes the following steps: S1. Real-time acquisition of skin impedance data and microneedle puncture depth data; S2. The collected skin impedance data and puncture depth data are calculated and fused using a data fusion algorithm to obtain a fusion value F that represents the puncture state. S3. Based on the real-time collected skin impedance data, dynamically match the skin impedance range of the microneedles in the pre-stored mapping table with the advancement parameters, and adjust the speed and thrust of the microneedles in real time based on the fusion value F. S4. The puncture process is divided into three stages: the contact stage, the advancement stage, and the precise needle stop stage, and different rotation speed and thrust control logic is adopted in each stage. S5. Based on the fusion value F, a multi-level anomaly early warning mechanism is set up. When an anomaly is detected, a graded response action is executed to avoid over-puncture or under-puncture.

2. The microneedle control method according to claim 1, characterized in that, In step S2, the data fusion algorithm uses a dynamic weight allocation strategy to obtain the fusion value F. The formula for calculating the fusion value F is F = α × Z. std +β×D std Among them, Z std To standardize skin impedance values, D std The value is the standardized depth deviation, and α and β are dynamic weighting coefficients, where α + β = 1.

3. The microneedle control method according to claim 2, characterized in that, In step S1, skin impedance data is obtained through a contact impedance sensor with a range of 0-2000Ω and an accuracy of ±5Ω; puncture depth data is obtained through a laser displacement sensor with a range of 0-5mm and an accuracy of ±0.02mm.

4. The microneedle control method according to any one of claims 1-3, characterized in that, In step S4, during the contact phase, a low-speed advance is used, and contact with the skin is determined when the impedance value drops from infinity to a set threshold. During the advance phase, the advance is carried out according to the rotation speed of the mapping table, and the rotation speed is dynamically adjusted according to the fusion value F. During the precise needle stopping phase, the rotation speed is reduced when the actual depth reaches 90% of the target depth, and the machine is stopped when the target depth is reached and the error is ≤ ±0.01mm. In the pediatric scenario, the rotation speed and thrust are reduced throughout the entire process, while the parameters during the advance phase are increased in the scar repair scenario.

5. The microneedle control method according to claim 4, characterized in that, The skin impedance range-propulsion parameters in step S3 include: The mapping relationship between pre-stored skin impedance range and motor speed and thrust level; The mapping relationship is queried based on real-time impedance data to determine the corresponding initial rotational speed and thrust; The rotation speed is dynamically adjusted based on the fusion value F during the propulsion process.

6. The microneedle control method according to claim 5, characterized in that, The multi-level anomaly warning mechanism in step S5 includes three levels of warning rules: Level 1 warning: in response to a sudden drop in impedance, the rotation speed is reduced and a warning is issued; Level 2 warning: in response to depth fluctuations, the propulsion is paused and recalibrated. Level 3 warning: In response to impedance or depth exceeding limits, immediately shut down and issue an alarm.

7. A microneedle control device based on skin impedance-depth fusion, characterized in that, The microneedle control device includes: A dual-modal sensing module is used to collect skin impedance data and microneedle puncture depth data in real time; The data fusion module is used to calculate and fuse the collected skin impedance data and puncture depth data through a data fusion algorithm to obtain a fusion value F that represents the puncture state. The dynamic impedance-thrust matching module is used to dynamically match the skin impedance range and thrust parameters of the microneedles in the pre-stored mapping table according to the real-time collected skin impedance data, and adjust the speed and thrust of the microneedles in real time based on the fusion value F. The phased puncture control module is used to divide the puncture process into three stages: contact period, advancement period, and precise needle stop period, and to use differentiated speed and thrust control logic in each stage. The hierarchical anomaly warning module is used to set up a multi-level anomaly warning mechanism based on the fusion value F. When an anomaly is detected, it executes a hierarchical response action to avoid over-puncture or under-puncture.

8. The microneedle control device according to claim 7, characterized in that, The microneedle control device also includes a constant motor power control module, which adopts a three-loop feedback control of current, speed and thrust. The current loop stabilizes the current through a PI controller, the speed loop adjusts the speed through a PID controller, and the thrust loop measures the thrust through a strain gauge sensor. When the thrust deviation is greater than ±10%, speed compensation is triggered.

9. The microneedle control device according to claim 8, characterized in that, The microneedle control device also includes a sensor calibration module, which performs a calibration process weekly: impedance calibration is performed by connecting to 500Ω, 1500Ω, and 2500Ω standard resistance boxes with an error of ≤±2%; depth calibration is performed by verifying readings at 0.5mm, 1.0mm, and 2.0mm using a standard scale plate with an error of ≤±0.02mm; if calibration fails, the treatment function is locked and a faulty sensor is indicated.

10. A microneedle control system based on skin impedance-depth fusion, characterized in that, The microneedle control system includes the microneedle control device, data storage module, emergency processing module, machine learning module, and user interface as described in any one of claims 7-9. The data storage module is used to cache the real-time treatment data of the most recent 10 seconds; The emergency handling module is used to immediately cut off the power output and make the micro needle retract to the initial position at 5mm / s when a level 3 warning is triggered, the emergency stop button is pressed, or the voltage fluctuation is greater than ±20%. The built-in supercapacitor realizes power failure data protection. The machine learning module is used to optimize control parameters based on historical data; The user interface is used to input treatment parameters and display treatment status.

Images