Delivery pain monitoring system and method based on biofeedback
By using a multi-parameter fusion model and adaptive filtering algorithm, combined with edge computing and blockchain storage technology from biofeedback devices, the response delay and data security issues of traditional labor pain monitoring have been resolved, achieving precise pain monitoring and intelligent analgesia.
Patent Information
- Application Number
- CN202610114783.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-28
- Publication Date
- 2026-03-03
AI Technical Summary
Traditional labor pain monitoring suffers from problems such as large response delays, poor individual adaptability, insufficient data security, and signal distortion. Furthermore, biofeedback analgesia devices lack dynamic optimization mechanisms and medical records are easily tampered with.
By employing a multi-parameter fusion model, adaptive filtering algorithm, edge computing decision-making, and blockchain evidence storage technology, combined with bioelectric sensors, graphene piezoresistive materials, conductive gel electrodes, and edge computing units, precise pain monitoring and intelligent analgesia are achieved.
It improves the accuracy and individual adaptability of pain monitoring, ensures the reliability of signals and the security of data, reduces operational difficulty, and provides reliable medical records.
Smart Images
Figure CN121587679A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of medical device technology, specifically relating to a biofeedback-based labor pain monitoring system and method. Background Technology
[0002] Traditional labor pain monitoring often uses a single parameter (such as uterine contraction pressure) to assess pain intensity, which suffers from problems such as large response delays, poor individual adaptability, and insufficient data security. In existing technologies, biofeedback analgesia devices often suffer from signal distortion due to motion artifacts, electrical stimulation parameter adjustments rely on human experience and lack dynamic optimization mechanisms, and medical records are easily tampered with. This invention overcomes these technical bottlenecks by employing a multi-parameter fusion model, adaptive filtering algorithms, edge computing decision-making, and blockchain-based evidence storage technology, achieving accurate monitoring and intelligent analgesia for labor pain. Summary of the Invention
[0003] To address the aforementioned problems, this invention provides a biofeedback-based labor pain monitoring system, comprising a maternal monitoring module, a grip pressure sensor, an electrical stimulation wave output module, a maternal pain relief station, and an interactive module, with each module electrically connected to the others.
[0004] Preferably, the maternal monitoring module integrates a bioelectric sensor array to collect ECG and EMG signals in real time and uses an adaptive filtering algorithm to suppress motion artifacts; the grip pressure sensor uses graphene piezoresistive material; and the electrical stimulation wave generator is equipped with a waveform synthesis unit, supports custom waveform group editing, and has a built-in safety threshold protection circuit.
[0005] Preferably, the postpartum pain relief station is equipped with an edge computing unit, runs a real-time decision-making algorithm, and dynamically adjusts the electrical stimulation parameters according to a multi-parameter fusion model; the interaction module adopts a capacitive touch screen, and the interface layout conforms to the ergonomic standards of medical devices.
[0006] Preferably, the electrical stimulation wave output module uses conductive gel electrodes to reduce contact resistance and improve stimulation efficiency.
[0007] Preferably, it also includes an environmental sensing module and an operation log module. The environmental sensing module monitors the temperature, humidity, and light intensity of the delivery room in real time and automatically adjusts to the optimal delivery environment. The operation log module uses blockchain storage technology to ensure that medical records cannot be tampered with. The power management unit is equipped with a UPS backup power supply, which supports continuous operation for ≥3 hours after a power outage to ensure the continuity of treatment.
[0008] A biofeedback-based method for monitoring labor pain, comprising the following steps: S1. Initialization: The pregnant woman wears a maternal monitoring module on her wrist, which integrates a bioelectric sensor array with an ECG / EMG signal sampling rate ≥1kHz. She holds a pressure sensor made of graphene piezoresistive material with a resolution of 0.01kPa and a response time ≤10ms to complete the device's self-test program to confirm normal communication. The battery capacity of the external device is ≥2000mAh. S2. Data Acquisition: Real-time monitoring of vital signs such as uterine contraction pressure, grip strength, heart rate, and blood pressure; S3. Data Analysis: After receiving the signal, the pain relief station for postpartum women uses a multi-parameter fusion model (uterine contraction pressure, grip strength, heart rate variability, skin conductance) to dynamically calculate the electrical stimulation parameters. The real-time decision-making algorithm has a running cycle of ≤100ms and an adjustment delay of ≤50ms. S4. Electrical stimulation output: The electrical stimulation wave generator generates an adjustable analgesic and massage wave group of 1-150Hz (amplitude accuracy ±0.1mA, wave width 0.1-0.5ms), which is output to the pregnant woman's body surface through silver chloride coated conductive gel electrodes. The contact impedance is ≤100Ω, and the stimulation efficiency is improved by ≥20%. S5. Dynamic adjustment: Automatically adjusts parameters based on a neural network mapping model of grip strength and pain level, with a built-in safety threshold protection circuit to prevent current from exceeding 50mA; S6. Feedback Record: The operation log uses blockchain notarization technology to generate a unique hash value. The data is synchronized through 5G and Wi-Fi 6 dual-mode communication (upload speed ≥10Mbps, latency ≤50ms). Medical staff can view real-time data and treatment effects through a medical-grade capacitive screen (response time ≤200ms).
[0009] The beneficial effects of this invention are as follows: The system employs an LMS adaptive filtering algorithm to suppress motion artifacts, ensuring the accuracy of ECG / EMG signal acquisition and avoiding signal distortion caused by motion interference in traditional devices, thus improving the reliability of pain monitoring. A neural network mapping model between grip strength and pain level enables a non-linear correlation between pain intensity and electrical stimulation parameters, enhancing the individual adaptability of analgesic effects and avoiding the blindness of manual experience-based adjustments. Blockchain technology ensures full-process data traceability from initialization, data acquisition, analysis, electrical stimulation output to feedback recording, providing reliable evidence for medical disputes and supporting long-term tracking and evaluation of treatment effects. The system is suitable for various scenarios, including tertiary hospitals, community hospitals, and maternal and child health centers. It is easy to deploy, and the user interface conforms to medical device ergonomic standards, reducing the operational difficulty for medical staff and increasing the system's adoption rate. Attached Figure Description
[0010] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0011] Figure 1 This is a schematic diagram of module connections according to an embodiment of the present invention.
[0012] In the picture: Maternal monitoring module 1, bioelectric sensor 101, grip pressure sensor 2, electrical stimulation wave output module 3, postpartum pain relief station 4, interaction module 5, environmental perception module 6, operation log module 7, power management unit 8. Detailed Implementation
[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Example
[0014] This system employs a closed-loop control architecture encompassing physiological signal acquisition, multimodal fusion, intelligent decision-making, and dynamic intervention. The maternal monitoring module features a bioelectric sensor array with dedicated ECG and EMG electrodes. An LMS adaptive filtering algorithm suppresses motion artifacts in real time, ensuring clear signals even during strong contractions. The grip pressure sensor utilizes graphene piezoresistive material with a GF value as high as 603 kPa⁻¹, accurately capturing minute changes in the mother's grip strength at the 0.1N level. The electrical stimulation wave output module is equipped with a waveform synthesis unit, supporting custom waveform editing. The built-in safety threshold protection circuit employs a dual-redundancy design, ensuring the stimulation current does not exceed the safe threshold of 20mA.
[0015] During the active phase of labor, the mother holds a graphene piezoresistive sensor, and the system collects her grip pressure and bioelectrical signals in real time. When contractions occur, the grip pressure increases sharply from a baseline of 15N to 45N within 3 seconds, the ECG shows a heart rate increase to 110 bpm, and the EMG electromyography signal amplitude increases by 300%. The decision algorithm on the edge computing unit immediately activates the multi-parameter fusion model and dynamically adjusts the electrical stimulation parameters: switching the original low-frequency massage wave frequency of 2Hz and pulse width of 200μs to the analgesic wave group frequency of 100Hz and pulse width of 50μs, and automatically increasing the stimulation intensity by 10% according to the pressure change. At the same time, the electrical stimulation is precisely transmitted to the T10-L1 spinal cord segments through the conductive gel electrode contact impedance of <5kΩ.
[0016] The environmental sensing module monitors the delivery room temperature and humidity in real time, maintaining them at 24-26℃ and light intensity at 300-500 lux. When humidity exceeds 65%, the fresh air system is automatically activated for adjustment. When the light sensor detects insufficient natural light, the LED supplemental lighting system is triggered to simulate a sunrise environment. The UPS backup power supply takes over immediately after a power outage, ensuring treatment continuity. Within a 3-hour power supply period, at least 6 complete adjustments to electrical stimulation parameters can still be made.
[0017] The operation log module uses blockchain notarization technology. Each operation record generates a hash value containing a timestamp, operator ID, and parameter change details, and is stored immutably through a distributed ledger. The maternity interaction module uses a 12-inch capacitive touchscreen. The interface layout strictly follows medical device ergonomic standards, with key operation buttons ≥20mm in size and an interface color contrast ≥3:1, ensuring accurate touch control even when medical staff are wearing gloves.
[0018] Furthermore, the real-time decision-making algorithm running on the edge computing unit comprises three core parts: 1. Preprocessing the original signal by applying a 50Hz power frequency notch filter and a 0.5-15Hz bandpass filter; 2. Calculating 12 key parameters, such as the rate of change of gripping pressure and the root mean square value of the EMG signal, through feature extraction; 3. The decision model adopts the random forest algorithm, dynamically adjusting the electrical stimulation frequency, pulse width, and amplitude based on the multi-parameter fusion results, ensuring that the stimulation intensity does not exceed 80% of the safe threshold during the peak of uterine contractions, and returns to the basic analgesia level during the interval between contractions. Example
[0019] After the pregnant woman was admitted to the hospital, medical staff assisted her in wearing a wrist-mounted maternal monitoring module with an integrated bioelectric sensor array. This module, made of flexible silicone, conforms to the skin of the pregnant woman's wrist and incorporates an ECG / EMG dual-mode sensor array with a sampling rate set to 1.2kHz. Motion artifacts are suppressed using a differential amplifier. A handheld graphene piezoresistive grip force sensor is magnetically attached to the pregnant woman's palm, with a resolution stabilized at 0.01kPa after calibration. The response time is optimized to 8ms using an FPGA real-time processing circuit. After the device's self-test program is initiated, it sequentially checks the sensor communication status (SPI bus), battery power (2500mAh lithium battery), and the 5G / Wi-Fi 6 module handshake signal. After all parameters are confirmed and displayed on an OLED screen, the system enters standby mode.
[0020] Uterine contraction pressure is monitored in real time by a flexible contraction sensor. This sensor uses a PVDF piezoelectric film with a response frequency covering 0.1-10Hz, capable of capturing minute pressure changes during contractions. Grip pressure values are continuously acquired by a graphene sensor, and the data stream is transmitted at a sampling rate of 100Hz after low-pass filtering. Heart rate is calculated by extracting R-wave intervals from ECG electrodes. Blood pressure is automatically measured every 5 minutes using an oscillometric cuff blood pressure monitor. Skin conductance is acquired by electrodes driven by a constant current source. All data is synchronized to the central processing unit via the BLE 5.2 protocol, ensuring that the time alignment error of multiple parameters is ≤10ms.
[0021] After receiving multi-parameter data, the postpartum pain relief station uses a multi-parameter fusion model for real-time calculation. Model inputs include: contraction pressure normalized to 0-100%, grip strength standardized, heart rate variability (HRV) time-domain / frequency-domain indices, and skin conductance (SC) change rate. A pain level score of 0-10 is dynamically generated using a hybrid model of Support Vector Machine (SVM) and Long Short-Term Memory (LSTM). The decision algorithm, based on a reinforcement learning framework, runs every 80ms, outputting electrical stimulation parameter adjustment instructions; the latency is optimized to 40ms via hardware acceleration.
[0022] The electrical stimulation wave generator produces adjustable analgesic and massage wave groups based on algorithmic instructions. The frequency range of 1-150Hz is achieved through Direct Digital Synthesis (DDS) technology, with amplitude accuracy controlled by a 16-bit DAC to ensure an error of ±0.1mA. The wave width is adjusted within the range of 0.1-0.5ms via an FPGA-gated circuit. The stimulation wave is output through silver chloride-coated conductive gel electrodes. The electrode contact impedance is controlled at 80Ω through an impedance matching circuit, and the stimulation efficiency is improved by 25% through pulse width modulation (PWM) optimization. The electrodes feature a split design, allowing for precise stimulation of acupoints on the pregnant woman's lower back and abdomen.
[0023] The neural network mapping model between grip strength and pain level is trained based on actual clinical data. The input is a grip strength time series, and the output is a predicted pain level. The model extracts features using a convolutional neural network (CNN) and combines this with a fully connected layer to output the predicted value. When the predicted pain level exceeds a threshold, the system automatically increases the electrical stimulation intensity; if the grip strength suddenly decreases (e.g., during childbirth when the mother relaxes), the stimulation intensity is reduced. A safety threshold protection circuit is implemented using a hardware overcurrent protection chip. When the detected current exceeds 45mA (with a 5mA safety margin), the output is immediately cut off and an alarm is triggered.
[0024] Operation logs utilize blockchain technology to generate unique hash values and are encrypted using the SHA-256 algorithm to ensure data immutability. Data is synchronized via dual 5G NSA / SA and Wi-Fi 6 links, with upload speeds consistently maintained at 12Mbps and latency controlled within 40ms. Medical staff can view real-time data on a medical-grade capacitive touchscreen with a 180ms response time. The screen supports multi-touch and gesture controls, allowing access to historical data curves, stimulation parameter trajectories, and treatment effect analysis reports to support clinical decision-making.
[0025] During labor, as the mother enters the active phase, uterine contractions gradually increase in pressure, leading to a rise in grip strength. The system monitored in real-time contraction pressure peaks at 80%, grip strength at 0.5 kPa, HRV decreasing to 30 ms, skin conductance increasing by 15%, and a predicted pain level of 8. The decision algorithm immediately adjusted the electrical stimulation parameters: frequency increased to 120 Hz, amplitude increased to 3.0 mA, and pulse width to 0.3 ms. The stimulation wave was applied to acupoints in the pregnant woman's lumbosacral region via electrodes, effectively relieving pain. Medical staff observed through a capacitive touchscreen that the pain level decreased to 4 after stimulation, and uterine contraction pressure remained stable, confirming a good analgesic effect. Data from the entire process was synchronized to the blockchain to ensure the integrity and traceability of medical records, while dual-mode communication ensured the stability and real-time nature of data transmission, meeting clinical needs.
[0026] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A biofeedback-based labor pain monitoring system, characterized in that: It includes a maternal monitoring module (1), a grip pressure sensor (2), an electrical stimulation wave output module (3), a postpartum pain relief station (4), and an interactive module (5), with each module electrically connected to the others.
2. The biofeedback-based labor pain monitoring system according to claim 1, characterized in that: The maternal monitoring module (1) integrates an array of bioelectric sensors (101) to collect ECG and EMG signals in real time and uses an adaptive filtering algorithm to suppress motion artifacts; the grip pressure sensor (2) uses graphene piezoresistive material; the electrical stimulation wave output module (3) is equipped with a waveform synthesis unit, supports custom waveform group editing, and has a built-in safety threshold protection circuit.
3. The biofeedback-based labor pain monitoring system according to claim 2, characterized in that: The postpartum pain relief station (4) is equipped with an edge computing unit and runs a real-time decision-making algorithm to dynamically adjust the electrical stimulation parameters according to the multi-parameter fusion model; the interaction module (5) adopts a capacitive touch screen and the interface layout conforms to the ergonomic standards of medical equipment.
4. The biofeedback-based labor pain monitoring system according to claim 3, characterized in that: The electrical stimulation wave output module (3) uses conductive gel electrodes to reduce contact resistance and improve stimulation efficiency.
5. The biofeedback-based labor pain monitoring system according to claim 4, characterized in that: It also includes an environmental sensing module (6) and an operation log module (7). The environmental sensing module (6) monitors the temperature, humidity and light intensity of the delivery room in real time and automatically adjusts it to the best delivery environment. The operation log module (7) adopts blockchain storage technology to ensure that medical records cannot be tampered with. The power management unit (8) is equipped with a UPS backup power supply to support continuous operation for ≥3 hours after power failure, ensuring the continuity of treatment.
6. A method for monitoring labor pain based on biofeedback, characterized in that, The method steps include: S1. Initialization: The pregnant woman wears a maternal monitoring module (1) on her wrist, which integrates a bioelectric sensor (101) array, with an ECG / EMG signal sampling rate ≥1kHz, and holds a pressure sensor (2) made of graphene piezoresistive material with a resolution of 0.01kPa and a response time ≤10ms. The device completes a self-test procedure to confirm normal communication, and the battery capacity in the external device is ≥2000mAh. S2. Data Acquisition: Real-time monitoring of vital signs such as uterine contraction pressure, grip strength, heart rate, and blood pressure; S3. Data analysis: After receiving the signal, the pain relief station (4) uses a multi-parameter fusion model (uterine contraction pressure, grip strength value, heart rate variability, skin conductance) to dynamically calculate the electrical stimulation parameters. The real-time decision algorithm has a running cycle of ≤100ms and an adjustment delay of ≤50ms. S4. Electrical stimulation output: The electrical stimulation wave generator generates an adjustable analgesic and massage wave group of 1-150Hz (amplitude accuracy ±0.1mA, wave width 0.1-0.5ms), which is output to the pregnant woman's body surface through silver chloride coated conductive gel electrodes. The contact impedance is ≤100Ω, and the stimulation efficiency is improved by ≥20%. S5. Dynamic adjustment: Automatically adjusts parameters based on a neural network mapping model of grip strength and pain level, with a built-in safety threshold protection circuit to prevent current from exceeding 50mA; S6. Feedback Record: The operation log uses blockchain notarization technology to generate a unique hash value. The data is synchronized through 5G and Wi-Fi 6 dual-mode communication (upload speed ≥10Mbps, latency ≤50ms). Medical staff can view real-time data and treatment effects through a medical-grade capacitive screen (response time ≤200ms).