Postpartum hemorrhage real-time monitoring smart nursing device integrated with multi-modal sensors

Abstract

The present application relates to the technical field of medical care monitoring equipment, and discloses a postpartum hemorrhage real-time monitoring intelligent nursing device integrated with a multi-modal sensor and an absorbing pad body; a multi-modal sensor array is arranged in the absorbing pad body, and the multi-modal sensor array comprises a photoelectric sensing unit, a thin film pressure sensing array and a micro weight sensing unit; a master control and communication assembly is electrically connected with the multi-modal sensor array, and the master control and communication assembly is used for collecting monitoring data of the photoelectric sensing unit, the thin film pressure sensing array and the micro weight sensing unit, and wirelessly transmitting the monitoring data to a data receiving terminal, so as to realize real-time dynamic estimation of postpartum hemorrhage volume, exclude maternal body movement interference through posture compensation filtering of pressure time derivative, and output target absolute hemorrhage volume and real-time hemorrhage rate. The present application eliminates the limitation of single-dimensional measurement, and realizes objective quantization output of postpartum hemorrhage volume.

Description

Technical Field

[0001] This invention relates to the field of medical and nursing monitoring equipment technology, specifically to an intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors. Background Technology

[0002] Postpartum hemorrhage is a common obstetric complication, and objective and continuous monitoring of postpartum blood loss and real-time bleeding rate is a routine clinical nursing procedure. Current devices for monitoring postpartum hemorrhage typically employ a single physical quantity detection structure. Some devices only have a gravity sensor at the bottom of the nursing pad to measure changes in the overall physical mass of the device. This reliance on a single gravity monitoring dimension means that during actual operation, changes in the mother's posture, such as turning over, sitting up, or moving, can cause transient mechanical pressure on the gravity sensor (not caused by bleeding), leading to instantaneous changes in the sensor's output voltage signal. This, in turn, causes abnormal jumps and false alarms in the blood loss value calculated by the data processing unit. Some devices use photoelectric sensors to measure only the two-dimensional planar diffusion area of ​​blood on the surface of the absorbent material. Due to the spatial differences in vertical depth during blood penetration in porous fiber materials, the two-dimensional diffusion area data cannot accurately reflect the actual volume of blood in three-dimensional space. This results in the system calculating a lower absolute blood loss than the actual amount when the bleeding is significant and penetrates deep into the material. Furthermore, the aforementioned hardware structure that relies on single-dimensional data acquisition cannot simultaneously acquire the deformation volume and overall mass change of blood. This results in the backend data processing system lacking basic physical data for cross-validation, making it unable to execute multivariate data fusion and attitude interference filtering algorithms, and making it difficult to achieve accurate quantitative output of absolute blood loss and real-time bleeding rate.

[0003] Therefore, the purpose of this invention is to provide an intelligent nursing device for real-time monitoring of postpartum hemorrhage that integrates multimodal sensors, in order to overcome the shortcomings of the prior art. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent nursing device for real-time monitoring of postpartum hemorrhage that integrates multimodal sensors. This solves the problems of existing postpartum hemorrhage monitoring devices, which use a single type of sensor for monitoring, leading to false alarms about blood loss due to changes in the mother's body posture, and the inability to accurately calculate absolute blood loss based on single-dimensional physical measurement data.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a postpartum hemorrhage real-time monitoring intelligent nursing device integrating multimodal sensors, comprising: Absorbent pad body; A multimodal sensor array is disposed in the absorbent pad body, the multimodal sensor array including a photoelectric sensing unit, a thin film pressure sensing array and a micro weight sensing unit; The main control and communication component is electrically connected to the multimodal sensor array. The main control and communication component is used to collect monitoring data from the photoelectric sensing unit, the thin-film pressure sensing array, and the miniature weight sensing unit, and wirelessly transmit the monitoring data to the data receiving terminal for real-time dynamic estimation of postpartum hemorrhage.

[0006] Preferably, the absorbent pad body comprises, from top to bottom, a water-permeable and skin-friendly surface layer, a diffusing and diffusion layer, a polymer absorbent core layer, and a leak-proof and breathable bottom layer.

[0007] Preferably, the photoelectric sensing units are distributed in a dispersed manner inside the flow-guiding diffusion layer. The photoelectric sensing units are used to detect the light attenuation signal when blood permeates in the flow-guiding diffusion layer, and to collect blood diffusion area data.

[0008] Preferably, the thin-film pressure sensor array is distributed in a grid pattern and is disposed between the polymer absorbent core layer and the leak-proof and breathable bottom layer. The thin-film pressure sensor array is used to detect the local pressure change signal applied to the bottom after the polymer absorbent core layer absorbs blood and expands in volume, and is used to collect blood volume deformation data.

[0009] Preferably, there are multiple miniature weight sensing units, which are respectively disposed in the bottom four corner edge areas of the leak-proof and breathable bottom layer. The miniature weight sensing units are used to detect the physical gravity signal of the absorbent pad body due to the absorption of blood, and to collect overall weight change data.

[0010] Preferably, it also includes a waterproof and anti-fouling isolation layer, which completely wraps around the outside of the main control and communication components to prevent blood from penetrating into the main control and communication components and causing short circuits or data distortion.

[0011] Preferably, the main control and communication component integrates a microprocessor and a wireless Bluetooth module. The microprocessor synchronously acquires the monitoring data and performs analog-to-digital conversion processing, and then sends the processed monitoring data to the data receiving terminal through the wireless Bluetooth module.

[0012] Preferably, the data receiving terminal includes a data processing module, and the specific steps of the data processing module for running the multi-source data fusion algorithm include: The baseline blood loss was calculated using a weighing method based on the monitoring data from the aforementioned miniature weight sensor unit. The volumetric blood loss was calculated using the volumetric method based on the monitoring data from the aforementioned thin-film pressure sensor array. The amount of blood loss due to diffusion is calculated using the area method based on the monitoring data from the aforementioned photoelectric sensing unit. The calculation results from the three sources are extracted, weighted, and fused to output the target absolute bleeding volume and real-time bleeding rate.

[0013] Preferably, the data processing module is further configured to perform posture compensation filtering: by extracting the pressure change characteristics of the thin-film pressure sensor array in a short period of time, the mother's turning or sitting posture is identified, and the pressure change characteristics are filtered as interference variables in the multi-source data fusion algorithm to eliminate false alarms of blood loss caused by non-bleeding factors.

[0014] Preferably, it also includes an alarm and interaction module and a data interface module; When the data processing module determines that the target absolute bleeding volume or the real-time bleeding rate exceeds a preset safety threshold, the alarm and interaction module is triggered to issue an abnormal bleeding alarm signal. Meanwhile, the data interface module automatically writes the target absolute bleeding volume and real-time bleeding rate into the hospital's nursing information system and electronic medical records in accordance with standard medical data protocols.

[0015] This invention provides an intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors. It has the following beneficial effects: 1. This invention uses a photoelectric sensing unit, a thin-film pressure sensing array, and a micro weight sensing unit to collect data on the planar diffusion area, gelation volume deformation, and overall physical mass of blood, respectively. In the data processing module, spatial cross-validation and correlation calculation are performed on the heterogeneous physical parameters of the above three dimensions. This overcomes the shortcomings of conventional single weight measurement or planar area measurement, which cannot accurately map the actual volume of blood permeation in three-dimensional space. By verifying the multiple physical states of blood diffusion, expansion, and weight gain, an objective absolute blood loss value is output.

[0016] 2. This invention extracts pressure abrupt change features in the output data of the thin-film pressure sensor array where the time derivative is greater than a preset threshold through a data processing module, and performs attitude compensation filtering to discard the pressure abrupt change features as interference variables. Based on the difference in pressure change gradient, it distinguishes between the slowly changing pressure caused by the slow absorption of blood and the transient high-frequency pressure caused by the change in the mother's posture, eliminates mechanical compression interference caused by the transfer of physical center of gravity, and solves the problem of false blood loss reports caused by abnormal jumps in sensor values.

[0017] 3. This invention configures the absorbent pad body as a series of stacked layers: a water-permeable and skin-friendly surface layer, a diffusing layer, a polymer absorbent core layer, and a leak-proof and breathable bottom layer. The photoelectric sensing unit is distributed inside the diffusing layer, and the thin-film pressure sensing array is placed at the bottom of the polymer absorbent core layer. The physical path of blood's horizontal capillary permeation and vertical gel phase expansion at different depths is matched with the physical sensing working surface of each sensor at the physical structure level, ensuring the accuracy of the acquisition of the bottom layer's physical signals. Attached Figure Description

[0018] Figure 1 This is a perspective view of the present invention; Figure 2 This is a schematic diagram of the internal structure of the present invention; Figure 3 This is a top view of the internal structure of the present invention; Figure 4 This is a schematic diagram of the underlying structure of the present invention.

[0019] Among them, 100 is the absorbent pad body; 110 is the water-permeable and skin-friendly surface layer; 120 is the diffusing layer; 130 is the superabsorbent polymer core layer; 140 is the leak-proof and breathable bottom layer; 200 is the multimodal sensor array; 210 is the photoelectric sensing unit; 220 is the thin film pressure sensing array; 230 is the miniature weight sensing unit; 300 is the main control and communication components; and 400 is the waterproof and stain-resistant isolation layer. Detailed Implementation

[0020] The technical solutions in the embodiments 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.

[0021] Please see the appendix Figure 1 - Appendix Figure 4 This invention provides an intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors, comprising: Absorbent pad body 100; multimodal sensor array 200, disposed in absorbent pad body 100, multimodal sensor array 200 includes photoelectric sensing unit 210, thin film pressure sensing array 220 and miniature weight sensing unit 230; main control and communication component 300, electrically connected to multimodal sensor array 200, main control and communication component 300 is used to collect monitoring data from photoelectric sensing unit 210, thin film pressure sensing array 220 and miniature weight sensing unit 230, and wirelessly transmit the monitoring data to data receiving terminal for real-time dynamic estimation of postpartum hemorrhage.

[0022] Specifically, this technical solution adopts a multi-source physical signal acquisition structure. In terms of physical connection structure, the photoelectric sensing unit 210, thin-film pressure sensing array 220, and miniature weight sensing unit 230 inside the multi-modal sensor array 200 are respectively connected to the input interface of the main control and communication component 300 through flexible printed circuit cables. The photoelectric sensing unit 210 provides physical data in the two-dimensional planar diffusion dimension, the thin-film pressure sensing array 220 provides physical data in the three-dimensional deformation dimension, and the miniature weight sensing unit 230 provides physical data in the overall mass dimension. After the main control and communication component 300 is powered on, it synchronously reads the analog or digital electrical signals of the above three dimensions according to the set sampling frequency, and sends the three monitoring data to the data receiving terminal in the form of radio frequency electromagnetic waves through the radio frequency antenna. Through the above components and connection structure, the basic data of bleeding volume can be synchronously obtained from multiple physical quantities, solving the problem that a single parameter is easily affected by external environmental interference.

[0023] The absorbent pad body 100 includes, from top to bottom, a water-permeable and skin-friendly surface layer 110, a diffusing and diffusion layer 120, a super absorbent polymer core layer 130, and a leak-proof and breathable bottom layer 140.

[0024] Specifically, the permeable and skin-friendly surface layer 110 is made of hydrophilic non-woven fabric material, with a porosity parameter set to allow blood to pass through without surface accumulation; the flow-guiding and diffusion layer 120 is made of fluff pulp or wood pulp fiber material, closely attached to the bottom of the permeable and skin-friendly surface layer 110, providing capillary action to guide the blood passing through the permeable and skin-friendly surface layer 110 to spread horizontally in all directions; the superabsorbent polymer core layer 130 contains sodium polyacrylate superabsorbent polymer particles, closely attached to the bottom of the flow-guiding and diffusion layer 120, absorbing the blood that permeates through the flow-guiding and diffusion layer 120 and undergoing volume expansion, converting the free blood into a gel solid; the leak-proof and breathable bottom layer 140 is made of polyethylene cast film and is set at the bottom of the structure to block liquid water from penetrating downwards. Through the above-mentioned top-down material arrangement structure, the flow path and phase change of blood are controlled, while providing mounting reference surfaces of different depths for the multimodal sensor array 200.

[0025] The photoelectric sensing units 210 are distributed in a scattered manner inside the flow diffusion layer 120. The photoelectric sensing units 210 are used to detect the light attenuation signal when blood permeates in the flow diffusion layer 120, and are used to collect blood diffusion area data.

[0026] Specifically, the dispersed arrangement refers to the arrangement of multiple photoelectric sensing units 210 in a matrix array at preset distances, such as 2 cm in both the horizontal and vertical directions, within the horizontal plane of the flow diffusion layer 120. These units are fixed in the fiber structure of the flow diffusion layer 120. In the absence of bleeding, the infrared light emitted by the light-emitting diodes is reflected in the fiber structure and received by the photosensitive receiver, generating a first voltage value. When blood permeates into the flow diffusion layer 120 and covers the location of a specific photoelectric sensing unit 210, the blood absorbs and blocks the infrared light, causing a decrease in the light intensity received by the photosensitive receiver, generating a second voltage value lower than the first voltage value. The main control and communication component 300 generates and outputs blood diffusion area data by recording the number and coordinate positions of the photoelectric sensing units 210 that output the second voltage value, thus solving the problem that initial bleeding cannot be detected in time by changes in weight.

[0027] The thin-film pressure sensor array 220 is distributed in a grid pattern and is disposed between the polymer absorbent core layer 130 and the leak-proof and breathable bottom layer 140. The thin-film pressure sensor array 220 is used to detect the local pressure change signal applied to the bottom after the polymer absorbent core layer 130 absorbs blood and expands in volume, and is used to collect blood volume deformation data.

[0028] Specifically, the thin-film pressure sensor array 220 consists of two flexible polyester films containing conductive carbon paste patterns. The wires on the upper and lower films intersect each other perpendicularly to form grid intersections. Each grid intersection constitutes an independent piezoresistive sensor. After the polymer absorbent core layer 130 absorbs blood, it undergoes irreversible volume expansion, applying mechanical pressure to the thin-film pressure sensor array 220 below in the vertical direction. As the amount of blood absorbed increases, the local pressure on the corresponding grid intersection increases, the contact area between the upper and lower films increases, and the resistance value of the grid intersection decreases. The main control and communication component 300 scans the resistance change values ​​of each grid intersection to establish a three-dimensional matrix reflecting the increase in thickness of different areas of the polymer absorbent core layer 130, and then calculates and outputs blood volume deformation data, solving the problem that planar area measurement cannot reflect three-dimensional absorption.

[0029] Multiple miniature weight sensing units 230 are respectively set in the bottom four corner edge areas of the leak-proof and breathable bottom layer 140. The miniature weight sensing units 230 are used to detect the physical gravity signal increased by the absorption pad body 100 due to blood absorption, and to collect overall weight change data.

[0030] Specifically, the miniature weight sensing unit 230 uses a weighing sensor based on metal strain gauges, with four units respectively bonded and fixed to the upper left, lower left, upper right, and lower right corners of the bottom surface of the rectangular leak-proof and breathable bottom layer 140. When the absorbent pad body 100 is laid on the bed surface, these four corners constitute the main force support points for the entire device in contact with the bed surface. As blood continues to flow into the absorbent pad body 100, the overall physical mass of the device increases. The metal strain gauges inside the miniature weight sensing units 230 distributed at the four corners undergo mechanical elastic deformation due to the increase in gravity, causing a millivolt-level change in the output voltage of the Wheatstone bridge circuit. The main control and communication component 300 collects and accumulates the voltage changes of the four miniature weight sensing units 230, and calculates the overall weight change data of the absorbent pad body 100 based on the pre-calibrated voltage-mass correspondence curve, providing a measurement benchmark for the absolute mass dimension.

[0031] It also includes a waterproof and anti-fouling isolation layer 400, which completely wraps around the outside of the main control and communication components 300 to prevent blood from penetrating into the main control and communication components 300 and causing short circuits or data distortion.

[0032] Specifically, the waterproof and anti-fouling isolation layer 400 uses medical-grade polyurethane film material or a silicone sealing sleeve structure formed by low-pressure injection molding. The circuit board of the main control and communication component 300 and its surface-mounted electronic components are placed inside the waterproof and anti-fouling isolation layer 400. The edges of the waterproof and anti-fouling isolation layer 400 are completely sealed by hot pressing or ultrasonic welding process, with only airtight perforations reserved for the sensor cable to pass through. Through the physical barrier of the waterproof and anti-fouling isolation layer 400, direct contact between liquid blood, amniotic fluid or body fluid and the exposed metal pins of the main control and communication component 300 is blocked, preventing short circuits and burnout between circuit board components caused by conductive liquids. At the same time, it avoids distortion of analog signal acquisition caused by pin oxidation and extends the working life of the electronic module in humid environments.

[0033] The main control and communication component 300 integrates a microprocessor and a wireless Bluetooth module. The microprocessor synchronously acquires monitoring data and performs analog-to-digital conversion processing, and then sends the processed monitoring data to the data receiving terminal through the wireless Bluetooth module.

[0034] Specifically, the main control and communication component 300 includes a printed circuit board carrier. The microprocessor and the wireless Bluetooth module are fixed to the surface of the printed circuit board using surface mount technology. The microprocessor is equipped with multiple general-purpose input / output pins and analog-to-digital conversion pins, which are connected to the output wires of the photoelectric sensing unit 210, the thin-film pressure sensing array 220, and the miniature weight sensing unit 230, respectively. The working process is as follows: the microprocessor reads the continuous analog voltage signal output by the sensor through the analog-to-digital conversion pins, executes the analog-to-digital conversion instruction, and converts the analog voltage signal into a discrete digital signal represented in binary. Subsequently, the microprocessor packages the discrete digital signal according to a preset data packet format and sends the data packet to the wireless Bluetooth module through the serial port of a universal asynchronous transceiver. The wireless Bluetooth module modulates the radio frequency signal according to the Bluetooth Low Energy communication protocol and radiates the data packet into space until it is captured by the Bluetooth receiving antenna of the external data receiving terminal.

[0035] The data receiving terminal is equipped with a data processing module. The specific steps of the data processing module to run the multi-source data fusion algorithm include: calculating the baseline blood loss based on the monitoring data of the micro weight sensor unit 230 using the weighing method; calculating the volumetric blood loss based on the monitoring data of the thin-film pressure sensor array 220 using the volume method; calculating the diffusion blood loss based on the monitoring data of the photoelectric sensor unit 210 using the area method; and comprehensively extracting the calculation results of the three and performing weighted fusion to output the target absolute bleeding volume and real-time bleeding rate.

[0036] Specifically, the data processing module consists of a central processing unit and a memory storing computer-executable instructions. The specific execution steps of the multi-source data fusion algorithm are as follows: The pre-stored blood density constant is called, and the overall increase in mass acquired by the micro weight sensing unit 230 is divided by the blood density constant to obtain the baseline blood loss value. Substitute the pressure increase at each grid intersection in the thin-film pressure sensor array 220 into the preset deformation volume mapping equation, integrate to calculate the overall volume increase, and obtain the volumetric blood loss value. The total number of nodes of the photoelectric sensing unit 210 that outputs low voltage signals is multiplied by the coverage area parameter calibrated for a single node and the saturated liquid absorption capacity per unit area of ​​the absorbent pad body 100 to obtain the value of diffuse blood loss. Apply the weighted summation formula: Target absolute bleeding = ×Base blood loss + ×Volume blood loss+ × Blood loss due to diffusion. (Among them) , , These are the preset weighting coefficients corresponding to the three parameters, and , , The sum equals 1. The data processing module records the time difference between two consecutive calculations of the target absolute bleeding volume, divides the difference between the two target absolute bleeding volumes by the time difference, and calculates and outputs the real-time bleeding rate value.

[0037] The data processing module is also used to perform posture compensation filtering: by extracting the pressure change features of the thin-film pressure sensor array 220 in a short period of time, the mother's turning or sitting posture is identified, and the pressure change features are used as interference variables in the multi-source data fusion algorithm to filter out false alarms of blood loss caused by non-bleeding factors.

[0038] Specifically, the execution process of posture compensation filtering is as follows: Before running the multi-source data fusion algorithm, the data processing module sets a sliding time window with a duration of 1 to 3 seconds. The data processing module calculates the time derivative, i.e., the rate of change, of the pressure data output by the thin-film pressure sensor array 220 within the sliding time window. Since the pressure increase caused by liquid absorption is a slow and gradual process, the shift in the body's center of gravity caused by the mother turning over or sitting up will cause the thin-film pressure sensor array 220 to generate instantaneous high-frequency pressure changes in a local area. When the data processing module determines that the time derivative of the pressure data is greater than the preset body movement judgment threshold, it confirms that a posture change caused by non-bleeding factors has occurred. At this time, the data processing module triggers the filtering mechanism, marking all monitoring data output by the multi-modal sensor array 200 within the current sliding time window as invalid interference variables and not substituting them into the multi-source data fusion algorithm calculation. At the same time, it retrieves the valid historical data from the previous sliding time window as the input value for the current moment for calculation. By discarding transient change signals, the problem of abnormal jumps in the amount of bleeding caused by the mother's physical compression is solved.

[0039] It also includes an alarm and interaction module and a data interface module; when the data processing module determines that the target absolute bleeding volume or real-time bleeding rate exceeds the preset safety threshold, it triggers the alarm and interaction module to issue an abnormal bleeding alarm signal; at the same time, the data interface module automatically writes the target absolute bleeding volume and real-time bleeding rate into the hospital's nursing information system and electronic medical records in accordance with standard medical data protocols.

[0040] Specifically, the alarm and interaction module includes a speaker and a display screen, while the data interface module includes a network communication chip. The data processing module's storage area is pre-written with two comparison threshold parameters: an absolute bleeding volume alarm threshold (e.g., 500 ml) and a bleeding rate alarm threshold. In each calculation cycle, the data processing module compares the currently output target absolute bleeding volume with the absolute bleeding volume alarm threshold and the real-time bleeding rate with the bleeding rate alarm threshold. When either comparison result is determined to be greater than the target absolute bleeding volume, the data processing module outputs a high-level trigger signal to the alarm and interaction module. Upon receiving this signal, the alarm and interaction module drives the speaker to emit a specific frequency beep and displays the current bleeding volume value in red text on the display screen to alert medical staff. Simultaneously, the data interface module receives the target absolute bleeding volume value, real-time bleeding rate value, and current timestamp transmitted from the data processing module. It encapsulates this data into a data packet conforming to the HL7 standard protocol format and sends it to the hospital's local server via Ethernet or wireless LAN. The server stores the corresponding field values ​​and associates them with the mother's personal health record in the nursing information system and electronic medical record database, achieving automated data archiving.

[0041] Working principle: When postpartum bleeding occurs, the blood sequentially passes through the absorbent pad body 100 from top to bottom, consisting of a water-permeable and skin-friendly surface layer 110, a diffusing layer 120, and a super absorbent polymer core layer 130, reaching the leak-proof and breathable bottom layer 140. During the blood permeation process, the photoelectric sensing unit 210 located inside the diffusing layer 120 detects the light attenuation signal when the blood permeates the diffusing layer 120, and collects and outputs the blood diffusion area data. The thin film pressure sensing array 220, which is distributed in a grid pattern between the super absorbent polymer core layer 130 and the leak-proof and breathable bottom layer 140, detects the local pressure change signal applied to the bottom after the super absorbent polymer core layer 130 absorbs blood and expands in volume, and collects and outputs the blood volume deformation data.

[0042] Miniature weight sensing units 230, located at the four corners of the bottom edge of the leak-proof and breathable bottom layer 140, detect the physical gravity signal of the absorption pad body 100 due to blood absorption, collect and output overall weight change data. When the main control and communication component 300, which is electrically connected to the multimodal sensor array 200 and externally wrapped with a waterproof and stain-resistant isolation layer 400, is working, the microprocessor integrated inside synchronously collects the blood diffusion area data output by the photoelectric sensing unit 210, the blood volume deformation data output by the thin film pressure sensing array 220, and the overall weight change data output by the miniature weight sensing unit 230. After the microprocessor performs analog-to-digital conversion processing on the monitoring data output by the photoelectric sensing unit 210, the thin film pressure sensing array 220, and the miniature weight sensing unit 230, it sends the processed monitoring data to the data receiving terminal through the wireless Bluetooth module.

[0043] After receiving the monitoring data, the data processing module in the data receiving terminal runs a multi-source data fusion algorithm. The data processing module extracts the pressure change characteristics of the monitoring data of the thin film pressure sensor array 220 in a short period of time to identify the mother's turning or sitting posture. In the multi-source data fusion algorithm, the pressure change characteristics are used as interference variables to filter and perform posture compensation filtering to eliminate pressure data changes caused by non-bleeding factors.

[0044] After filtering, the data processing module calculates the baseline blood loss using the weighing method based on the monitoring data from the micro weight sensor unit 230, the volumetric blood loss using the volumetric method based on the monitoring data from the thin-film pressure sensor array 220, and the diffusion blood loss using the area method based on the monitoring data from the photoelectric sensor unit 210. Subsequently, the data processing module extracts the calculation results of the baseline blood loss, volumetric blood loss, and diffusion blood loss, performs weighted fusion, and outputs the target absolute blood loss and real-time bleeding rate.

[0045] When the data processing module determines that the target absolute bleeding volume or real-time bleeding rate exceeds the preset safety threshold, the data processing module outputs a trigger command to the alarm and interaction module. The alarm and interaction module receives the trigger command and issues an abnormal bleeding alarm signal. Simultaneously, the data processing module transmits the target absolute bleeding volume and real-time bleeding rate to the data interface module. The data interface module writes the target absolute bleeding volume and real-time bleeding rate into the hospital's nursing information system and electronic medical records according to the standard medical data protocol.

Claims

1. A smart nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors, characterized in that, include: Absorbent pad body (100); A multimodal sensor array (200) is disposed in the absorbent pad body (100). The multimodal sensor array (200) includes a photoelectric sensing unit (210), a thin film pressure sensing array (220), and a miniature weight sensing unit (230). The main control and communication component (300) is electrically connected to the multimodal sensor array (200). The main control and communication component (300) is used to collect monitoring data from the photoelectric sensing unit (210), the thin film pressure sensing array (220) and the miniature weight sensing unit (230), and wirelessly transmit the monitoring data to the data receiving terminal for real-time dynamic estimation of postpartum hemorrhage.

2. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 1, characterized in that, The absorbent pad body (100) comprises, from top to bottom, a water-permeable and skin-friendly surface layer (110), a diffusing and diffusion layer (120), a polymer water-absorbing core layer (130), and a leak-proof and breathable bottom layer (140).

3. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 2, characterized in that, The photoelectric sensing units (210) are distributed in a scattered manner inside the flow diffusion layer (120). The photoelectric sensing units (210) are used to detect the light attenuation signal when blood permeates in the flow diffusion layer (120) and to collect blood diffusion area data.

4. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 2, characterized in that, The thin-film pressure sensor array (220) is distributed in a grid pattern and is disposed between the polymer absorbent core layer (130) and the leak-proof and breathable bottom layer (140). The thin-film pressure sensor array (220) is used to detect the local pressure change signal applied to the bottom after the polymer absorbent core layer (130) absorbs blood and expands in volume, and is used to collect blood volume deformation data.

5. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 2, characterized in that, The number of miniature weight sensing units (230) is multiple, and they are respectively set in the bottom four corner edge areas of the leak-proof and breathable bottom layer (140). The miniature weight sensing units (230) are used to detect the physical gravity signal of the absorbent pad body (100) due to the absorption of blood, and to collect overall weight change data.

6. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 1, characterized in that, It also includes a waterproof and anti-fouling isolation layer (400), which completely wraps around the outside of the main control and communication component (300) to prevent blood from penetrating into the main control and communication component (300) and causing short circuits or data distortion.

7. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 1, characterized in that, The main control and communication component (300) integrates a microprocessor and a wireless Bluetooth module. The microprocessor synchronously acquires the monitoring data and performs analog-to-digital conversion processing. Then, the processed monitoring data is sent to the data receiving terminal through the wireless Bluetooth module.

8. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 1, characterized in that, The data receiving terminal is equipped with a data processing module, and the specific steps by which the data processing module runs the multi-source data fusion algorithm include: The baseline blood loss was calculated using a weighing method based on the monitoring data from the miniature weight sensing unit (230). The volumetric blood loss was calculated using the volumetric method based on the monitoring data from the thin-film pressure sensor array (220). The amount of blood loss due to diffusion is calculated using the area method based on the monitoring data from the photoelectric sensing unit (210). The calculation results from the three sources are extracted, weighted, and fused to output the target absolute bleeding volume and real-time bleeding rate.

9. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 8, characterized in that, The data processing module is also used to perform posture compensation filtering: by extracting the pressure change characteristics of the thin film pressure sensing array (220) in a short period of time, the mother's turning over or sitting up posture is identified, and the pressure change characteristics are used as interference variables in the multi-source data fusion algorithm to filter out false reports of blood loss caused by non-bleeding factors.

10. The intelligent nursing device for real-time monitoring of postpartum hemorrhage integrating multimodal sensors according to claim 8, characterized in that, It also includes an alarm and interaction module and a data interface module; When the data processing module determines that the target absolute bleeding volume or the real-time bleeding rate exceeds a preset safety threshold, the alarm and interaction module is triggered to issue an abnormal bleeding alarm signal. Meanwhile, the data interface module automatically writes the target absolute bleeding volume and real-time bleeding rate into the hospital's nursing information system and electronic medical records in accordance with standard medical data protocols.

Images