Intelligent lochia nursing pad and early warning system for postpartum hemorrhage monitoring

Abstract

The invention discloses an intelligent lochia nursing pad for postpartum hemorrhage monitoring and an early warning system, and relates to the technical field of physiological monitoring. The intelligent lochia nursing pad early warning system comprises a diffusion time phase construction module, a potential reversal recognition module, a phase dislocation analysis module, an anti-reversal time grating generation module and a reverse diffusion traction regulation and control module. The intelligent lochia nursing pad for postpartum hemorrhage monitoring comprises the early warning system. According to the invention, by establishing a diffusion time phase recording chain and introducing potential inversion identification, phase dislocation analysis and time grating traction regulation and control mechanisms, space-time synchronization of flexible sensing node response is realized, signal inversion and disorder are avoided, continuous increasing of a bleeding amount curve is maintained, negative sudden change is inhibited, and the stability and reliability of an early warning signal are improved. And accurate real-time monitoring of postpartum hemorrhage is realized.

Description

Technical Field

[0001] This invention relates to the field of physiological monitoring technology, specifically to an intelligent lochia care pad and early warning system for monitoring postpartum hemorrhage. Background Technology

[0002] The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring is an intelligent sensing system nursing device that combines flexible sensing technology, biochemical detection, and wireless communication. It is used for real-time, quantitative, and automated monitoring of postpartum bleeding. The system embeds a flexible resistance and capacitance sensor array within the absorbent layer of the care pad, forming an intelligent sensing layer for the lochia infiltration process. It can detect changes in local conductivity or dielectric constant after lochia enters, automatically calculating the bleeding volume and generating a continuous change curve. Simultaneously, the system includes a wireless transmission module as the data transmission unit for the intelligent sensing system, synchronizing monitoring data to the nurse's station or mobile device. It automatically alarms based on preset thresholds when bleeding is abnormal, providing early risk warnings. To also include infection screening capabilities, the care pad integrates pH or protein component detection units, enabling the intelligent sensing system to simultaneously identify biochemical trends in lochia and make a preliminary judgment on infection status. Through this structural layout and functional integration, the intelligent sensing system breaks through the limitations of traditional lochia observation, which relies on manual labor, is difficult to quantify, and has delayed alarms. It realizes the transformation of the nursing model from passive adsorption to active monitoring, and has efficient, accurate, and visualized clinical application value.

[0003] The existing technology has the following shortcomings: Under current technological conditions, smart lochia care pads used for postpartum hemorrhage monitoring typically collect electrical parameter changes during the lochia infiltration process by deploying multiple flexible sensor nodes within the absorbent layer. When lochia spreads rapidly within a short period, the fluid infiltration rate and distribution vary significantly across different areas. This can easily lead to transient potential reversal zones between adjacent sensor nodes, causing the detection signals, which should maintain a consistent trend, to cancel each other out and become misaligned on the time axis, resulting in signal coupling disorder and phase distortion. This abnormal state is reflected in the system's output bleeding volume change curve as a negative abrupt change that does not match the actual physiological state. In existing applications, this negative abrupt change is often directly interpreted as a sudden decrease in lochia volume, causing the system to make a judgment on the bleeding status that deviates from reality. This can lead to misjudgment when the actual bleeding volume continues to increase, thus masking the dynamic evolution trend of worsening postpartum hemorrhage, affecting the timeliness of early warning, and posing potential clinical risks.

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

[0005] The purpose of this invention is to provide an intelligent lochia care pad and early warning system for monitoring postpartum hemorrhage, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: On the one hand, an intelligent lochia care pad early warning system for postpartum hemorrhage monitoring is provided, including a diffusion phase construction module, a potential reversal identification module, a phase misalignment analysis module, an anti-reversal time grating generation module, and a reverse diffusion traction control module: The diffusion phase construction module establishes a diffusion phase recording chain around the early warning of intelligent lochia care pads, collects the electrical change trajectories of multiple flexible sensor nodes in a unified time sequence, and forms a diffusion front sequence table based on the diffusion phase recording chain. The potential reversal identification module, based on the diffusion phase recording chain and the diffusion front sequence table, compares the electrical change trajectory direction of adjacent flexible sensing nodes one by one, locks out continuous segments where the change direction is reversed, and converges them to form a potential reversal suspected band. The phase misalignment analysis module extracts the corresponding time segment along the potential reversal suspected band, performs time expansion on the differential changes of adjacent flexible sensing nodes based on the diffusion phase recording chain, marks the phase misalignment contacts with asynchronous changes, and forms a list of misalignment contacts. The anti-reversal time grating generation module traces the diffusion phase record chain around the list of misaligned contacts, arranges anti-reversal time gratings before and after each phase misaligned contact, limits the superposition order of adjacent flexible sensing nodes by inserting controlled micro-time slot silence intervals, and generates time grating traction drafts. The reverse diffusion traction control module performs reverse diffusion traction control based on the time grating traction draft. Before the occurrence of negative mutation, it briefly freezes the superimposed aperture of adjacent flexible sensing nodes and replays the sampling rhythm in segments according to the time grating traction sequence, outputs a bleeding volume change curve that maintains a continuous increasing trend, and generates a stable and consistent early warning signal.

[0007] Optionally, the process of forming the diffusion front sequence table is as follows: A diffusion phase recording chain was established around the early warning of intelligent lochia care pads. Multiple flexible sensing nodes were evenly distributed in the absorbent layer structure, and each flexible sensing node was kept in signal communication with the acquisition and control unit to monitor the electrical changes during the lochia penetration process. After diffusion-triggered detection, the acquisition and control unit starts the diffusion phase recording chain initialization program, independently numbers each flexible sensing node and sets a unified time reference to ensure the time consistency of electrical change data. The acquisition and control unit synchronously acquires the electrical change trajectories of multiple flexible sensing nodes at a uniform time interval, forming a diffusion time-phase recording chain containing continuous time segments; The acquisition and control unit sorts and maps the response time and spatial distribution of each flexible sensing node according to the diffusion phase recording chain, and generates a diffusion front sequence table to characterize the diffusion direction and advancement sequence of lochia in the absorbent layer of the nursing pad.

[0008] Optionally, the steps for generating the potential reversal suspect band are as follows: The electrical change trajectories of multiple flexible sensing nodes are extracted using a diffusion time-phase recording chain, and the change direction of conductivity or dielectric constant of each flexible sensing node is analyzed in a time-based manner. Based on the spatial arrangement of the diffusion front sequence table, the electrical change trajectory directions of adjacent flexible sensing nodes are compared one by one to determine the node pairs with opposite change directions. By tracing the time intervals of node pairs with opposite directions of change along the diffusion phase recording chain, the start and end positions of the inversion segments are determined and continuous inversion segments are formed. Multiple reversal segments were aggregated according to their temporal sequence and spatial proximity to form a potential reversal suspected band that is continuous on the time axis and adjacent in spatial distribution, in order to reflect the abnormal region of electrical response during the spread of lochia.

[0009] Optionally, during the formation of the potential reversal suspected zone, the boundaries of all reversal segments are referenced to the time nodes in the diffusion phase recording chain. The temporal continuity and spatial adjacency of the potential reversal suspected zone are constrained by the node arrangement direction of the diffusion front sequence table to ensure that the reversal region continues to advance continuously in time and corresponds to the actual propagation path of lochia diffusion in space.

[0010] Optionally, the process for forming the misaligned contact list is as follows: Extract the corresponding time segment along the potential reversal suspected band, and extract the electrical change data corresponding to the time range based on the diffusion phase recording chain to obtain the electrical change curve segments of multiple flexible sensing nodes; Using the extracted time segments as time windows, the electrical change trajectories of adjacent flexible sensing nodes are unfolded over time and compared for differences, forming a continuous differential change sequence. Based on the differential change sequence, the changing trend of adjacent flexible sensing nodes is tracked along the time axis, and the positions of phase misalignment contacts with asynchronous responses are marked. The marked phase misalignment contacts are sorted and categorized according to time sequence and spatial distribution order to form a misalignment contact list with time sequence and spatial correspondence, which is used as the input data structure for the subsequent time control stage.

[0011] Optionally, when forming the misaligned contact list, each phase misaligned contact is further indexed and encoded based on the spatial location information in the diffusion front sequence table, so that the misaligned contact list has both time coordinates and spatial coordinates, and is merged according to the arrangement order of the diffusion phase record chain, so as to ensure that the subsequent time grid traction control stage can achieve accurate backtracking of contact position and stable execution of time sequence based on the list.

[0012] Optionally, the steps for generating a time-raster traction draft are as follows: Based on the time coordinates and spatial location information recorded in the list of misaligned contacts, a backtracking operation is performed along the diffusion phase recording chain to determine the time location of each phase misaligned contact and the response status of adjacent flexible sensing nodes. By utilizing the time window obtained from contact backtracking positioning, anti-reversal time grids are arranged before and after each phase misalignment contact, so that the superposition relationship of adjacent flexible sensing nodes is constrained in the time dimension. Controlled micro-timeslot silence intervals are inserted inside the anti-reversal time grid according to the response order of adjacent flexible sensing nodes to limit the response order of adjacent nodes. The anti-reversal time grid and its internal silent intervals are integrated in chronological order to generate a time grid traction draft that records the response order of nodes and the time coverage range, which is used for execution in the subsequent time traction control stage.

[0013] Optionally, the duration of the micro-slot silence interval inserted within the anti-reversal time grating is set according to the response time resolution of the flexible sensing nodes in the diffusion phase recording chain. The silence interval maintains a constant time difference between adjacent flexible sensing nodes to ensure that the node response sequence advances sequentially along the lochia diffusion direction and maintains the temporal continuity of the time grating traction manuscript.

[0014] Optionally, based on the time-grating traction draft, reverse diffusion traction control is performed. Before the occurrence of negative mutation, the superimposed aperture of adjacent flexible sensing nodes is briefly frozen. The sampling rhythm is then replayed in segments according to the time-grating traction sequence to output a continuously increasing hemorrhage change curve and generate an early warning signal. The steps are as follows: Based on the time traction command, the sampling and control information of each flexible sensing node is loaded into the time execution channel according to the time sequence, node number, silent interval and response start and end time. Based on the time-driven command, the superposition aperture of adjacent flexible sensing nodes is temporarily frozen before the occurrence of negative mutation, in order to prevent signal mutation caused by instantaneous reverse superposition. Following the time grid sequence in the time grid traction draft, the node sampling data during the short-term freeze period is replayed in segments to ensure that the bleeding volume change curve maintains a continuous increasing trend on the time axis. The bleeding volume change curve after segmented playback is continuously output over time, and a stable and consistent early warning signal is generated based on the dynamic change trend of the curve to provide real-time alerts for abnormal bleeding conditions.

[0015] On the other hand, a smart lochia care pad for postpartum hemorrhage monitoring is provided, including the smart lochia care pad early warning system for postpartum hemorrhage monitoring as described in any one of the above claims.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention establishes a diffusion phase recording chain within the nursing pad and combines it with potential reversal identification and phase misalignment analysis mechanisms to synchronously constrain the response relationships of multiple flexible sensing nodes in both time and space dimensions. By tracking the lochia diffusion process over time and determining directional consistency, the invention maintains the orderliness of node responses during rapid fluid penetration, fundamentally avoiding signal direction reversal and phase misalignment caused by uneven local diffusion. This ensures that the bleeding volume change curve maintains a continuous and stable increasing trend during the acquisition phase, thereby achieving accurate mapping and real-time monitoring of bleeding dynamics.

[0017] This invention utilizes the synergistic effect of an anti-reversal time grating and back-diffusion traction control to construct a controlled temporal rhythm in the signal superposition stage, ensuring a clear sequential order for the sampling outputs of each flexible sensing node. Through short-term freeze and segmented playback mechanisms, it effectively suppresses negative abrupt changes caused by asynchronous node responses, ensuring that the bleeding volume curve maintains temporal continuity and numerical consistency even in complex diffusion environments. This approach improves the stability and reliability of the early warning signal, providing nursing staff with a dependable basis for judging dynamic bleeding trends. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a schematic diagram of a module of an intelligent lochia care pad early warning system for monitoring postpartum hemorrhage, provided in an embodiment of the present invention. Detailed Implementation

[0020] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0021] This invention provides, for example Figure 1 The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring includes a diffusion phase construction module, a potential reversal identification module, a phase misalignment analysis module, an anti-reversal time grating generation module, and a reverse diffusion traction control module. The diffusion phase construction module establishes a diffusion phase recording chain around the early warning of intelligent lochia care pads, collects the electrical change trajectories of multiple flexible sensor nodes in a unified time sequence, and forms a diffusion front sequence table based on the diffusion phase recording chain. To achieve time-series monitoring of lochia diffusion and accurate capture of bleeding volume trends, a diffusion phase recording chain can be established around the intelligent lochia care pad. This involves continuously collecting and sequentially organizing the electrical change trajectories of multiple flexible sensing nodes to form a diffusion front sequence table reflecting the dynamics of fluid diffusion. The specific steps are as follows:

[0022] In the diffusion-triggered detection phase, multiple flexible sensing nodes are uniformly distributed within the absorbent layer structure of the smart lochia care pad, and each flexible sensing node is kept in signal communication with the acquisition and control unit to monitor local electrical changes during the lochia infiltration process. When lochia comes into contact with the surface of any flexible sensing node, the conductivity or dielectric constant of that node will undergo a slight change, thereby triggering the acquisition and control unit to initiate the initialization program of the diffusion phase recording chain. In this phase, the acquisition and control unit assigns an independent number to each flexible sensing node and sets a unified time base, ensuring that all subsequently acquired data are calibrated from the same time starting point. This unified time base method ensures that the electrical change information acquired by each node has strict temporal comparability, thus laying a stable temporal foundation for the subsequent formation of the diffusion front sequence table.

[0023] During the node response acquisition phase, the acquisition control unit synchronously acquires the electrical parameters of all flexible sensing nodes at uniform time intervals. The impedance changes, capacitance responses, or conductivity changes exhibited by each flexible sensing node during lochia diffusion are recorded in real time. To ensure the continuity and integrity of the acquisition, the acquisition control unit cyclically visits each flexible sensing node at a fixed sampling frequency, sequentially reading the node's electrical change values ​​and attaching a time stamp. At this point, the diffusion time-phase recording chain gradually forms time-series data segments, each segment corresponding to the electrical state of a flexible sensing node at a specific time point. Because the diffusion of lochia within the absorbent layer exhibits directionality and hierarchy, the response times of different flexible sensing nodes to liquid penetration vary from microseconds to milliseconds. Therefore, maintaining consistency in the time sequence is crucial during acquisition to ensure that the response order of the nodes in the diffusion time-phase recording chain accurately reflects the physical frontier of lochia penetration. Through continuous acquisition in this phase, the complete electrical change trajectory from initial liquid penetration to local saturation can be obtained, providing a data source for subsequent time-series processing.

[0024] During the time series processing phase, the acquisition and control unit rearranges and merges the electrical change trajectories of different flexible sensor nodes on a unified time axis based on the existing time segment data in the diffusion phase recording chain. Specifically, the response time of each flexible sensor node is paired with its electrical change amplitude, and all data are sorted into a continuous chain using time sequence as an index. This time sorting not only records the initial response time of each flexible sensor node but also retains the duration of the electrical change, so as to determine the diffusion rate and penetration range of lochia in subsequent analysis. During the processing, the acquisition and control unit simultaneously maps the time series with the spatial distribution based on the geometric distribution relationship of each flexible sensor node in the absorbent layer of the nursing pad, so that the diffusion phase recording chain reflects not only the temporal sequence but also the spatial direction of liquid diffusion. This method can form a dynamic recording structure with dual spatial and temporal dimensions, providing a complete information chain for identifying the diffusion path and flow direction of lochia.

[0025] During the diffusion front sequence generation stage, the acquisition and control unit compares and calculates the first response times of all flexible sensor nodes based on the already time-sorted diffusion phase record chain, and arranges them chronologically to form a diffusion front sequence table. The diffusion front sequence table uses time as the horizontal axis and the flexible sensor node number as the vertical axis, recording the temporal progression of lochia from the initial infiltration point to the farthest diffusion boundary. This table clearly reflects the propagation trajectory and velocity change trend of the liquid diffusion front within the nursing pad, and can be used to determine the main direction and boundary range of lochia diffusion. The generation process of the diffusion front sequence table also incorporates the spatial mapping information from the previous stage, creating a complete correspondence between the diffusion phase record chain in both time and space dimensions, thus enabling subsequent data analysis to be based on the actual diffusion process for judgment and calculation. At this point, the establishment of the diffusion phase recording chain and the formation of the diffusion front sequence table have completed the closed loop. All electrical change trajectories collected in the previous stage have been integrated into a traceable data structure in a time-continuous manner, providing a reliable basis for subsequent identification of potential reversal phenomena, extraction of phase misalignment contacts, and execution of time grating traction control.

[0026] The potential reversal identification module, based on the diffusion phase recording chain and the diffusion front sequence table, compares the electrical change trajectory direction of adjacent flexible sensing nodes one by one, locks out continuous segments where the change direction is reversed, and converges them to form a potential reversal suspected band. To identify regions of potential reversal caused by differences in liquid permeability between adjacent flexible sensing nodes during the dynamic process of lochia diffusion, the electrical change trajectories of multiple flexible sensing nodes can be compared one by one based on the established diffusion phase record chain and diffusion front sequence table. Based on this, continuous segments of potential reversal can be identified, ultimately converging to form a suspected potential reversal band. The specific steps are as follows:

[0027] In the electrical trajectory direction extraction stage, the electrical change trajectories of multiple flexible sensing nodes included in the diffusion phase recording chain are used to directionally analyze the response changes of each node during the lochia penetration process. After contact with the liquid, the conductivity or dielectric constant of each flexible sensing node exhibits a specific trend over time, which is stored in the diffusion phase recording chain as continuous time segments. By sequentially reading the response time corresponding to each node in the diffusion front sequence table along the time axis, the upward or downward direction of the electrical change of each node can be obtained. To ensure accuracy in subsequent comparisons, the change direction of each node is normalized and sorted with a time reference, ensuring that the change directions of all nodes are compared from a unified time starting point. In this stage, the diffusion phase recording chain not only serves as a data carrier but also bears the responsibility of time series constraints, ensuring that the electrical change directions of different nodes can be analyzed within the same time frame, thus providing an aligned data foundation for subsequent comparisons of the directions of adjacent nodes.

[0028] During the adjacent node direction comparison phase, based on the spatial arrangement of the flexible sensing nodes in the diffusion front sequence table, adjacent node pairs are selected one by one, and their electrical change trajectories in the diffusion phase recording chain are compared. Since the diffusion of lochia is not uniform, different nodes may experience slight response delays in time, and the electrical change directions of some nodes may show opposite trends; that is, while the conductivity of one node continues to rise, the conductivity of neighboring nodes may briefly decrease. To accurately identify such direction reversals, it is necessary to compare the change curves of adjacent nodes on a unified time axis. By comparing the change directions of each adjacent node pair, two states can be identified: one is a normal segment with consistent change directions, and the other is an abnormal segment with opposite change directions. Since the diffusion front sequence table has already marked the temporal order of the nodes, the time period and corresponding spatial range of the direction reversal can be determined during the comparison process. The key to this stage lies in utilizing the temporal continuity of the diffusion phase recording chain and the spatial sequence of the diffusion front sequence table to combine temporal changes with spatial distribution. This makes the determination of direction reversal not only depend on the electrical changes of a single node, but also on the time difference response relationship between adjacent nodes, thereby improving the completeness and consistency of the identification.

[0029] In the reversal fragment localization stage, for the node pairs with opposite change directions identified in the previous stage, their continuous response intervals in the diffusion phase recording chain are further traced along the time axis to determine the start and end positions of the reversal fragments. The determination of each reversal fragment must satisfy two conditions: temporal continuity and spatial continuity. That is, within a certain time range, at least two adjacent nodes must exhibit a continuously opposite electrical change direction, and this relationship must occur during the advancement of the diffusion front. This method eliminates sporadic direction reversals caused by single instantaneous fluctuations, making the locked reversal fragments more continuous and representative. In this process, the diffusion phase recording chain provides temporally continuous data on electrical changes, while the diffusion front sequence table provides the spatial diffusion path; their combined effect provides a two-dimensional constraint for reversal fragment localization. By comparing the boundary time points of reversal fragments from multiple adjacent nodes, the temporal and spatial coverage of the continuous reversal region can be obtained, thus integrating discrete node direction reversal phenomena into a complete fragment structure. This fragmented localization method provides a clear input basis for subsequent aggregation processing of the reversal region, ensuring that the aggregation results accurately reflect the abnormal states that occur during the diffusion process.

[0030] During the convergence phase of the potential reversal suspected band, multiple reversal segments located in the previous phase are aggregated according to their temporal sequence and spatial proximity to form a continuous potential reversal suspected band. This phase analyzes the spatial arrangement of adjacent nodes in the diffusion front sequence table, merging temporally continuous and spatially adjacent reversal segments and integrating their corresponding time intervals and spatial ranges into a complete reversal region band. The formation of the potential reversal suspected band not only describes the area where the reversal occurs but also records its start and end times in the diffusion phase recording chain, enabling subsequent phase misalignment analysis and time grating manipulation of this region. During the convergence process, to maintain consistency between temporal and spatial information, the boundaries of all reversal segments are referenced to the time nodes of the diffusion phase recording chain, ensuring that the final potential reversal suspected band is continuous on the time axis and spatially adjacent. By synchronously combining temporal and spatial characteristics, the potential reversal suspected band not only possesses temporal tracking capabilities but also regional representation capabilities, intuitively reflecting the continuous region where the electrical response is disordered during the rapid diffusion phase of lochia. Thus, the entire potential reversal identification process forms a closed loop. The diffusion phase recording chain provides timing information, and the diffusion front sequence table provides spatial indexing. The combination of the two makes the comparison of the direction of electrical change trajectory, the locking of reversal segments, and the convergence of reversal regions a continuously executed process, providing a stable data foundation for the subsequent extraction of phase misalignment contacts and time grating traction control.

[0031] The phase misalignment analysis module extracts the corresponding time segment along the potential reversal suspected band, performs time expansion on the differential changes of adjacent flexible sensing nodes based on the diffusion phase recording chain, marks the phase misalignment contacts with asynchronous changes, and forms a list of misalignment contacts. To identify subtle changes in the asynchronous response of adjacent flexible sensing nodes during lochia diffusion within the suspected potential reversal zone, it is necessary to extract the corresponding time segments along the suspected potential reversal zone. Then, based on the diffusion phase recording chain, the differential changes of adjacent flexible sensing nodes are temporally unfolded. By continuously tracking and comparing the marked asynchronous phase misalignment contacts, a list of misalignment contacts for subsequent time-controlled adjustments is ultimately formed. The specific steps are as follows:

[0032] In the time segment extraction stage, based on the time start and end boundaries contained in the converged potential reversal suspected band, electrical change data corresponding completely to this time range are extracted along the diffusion phase recording chain. The potential reversal suspected band typically covers a continuous, spatially adjacent reversal region over a period of time. Multiple flexible sensing nodes within this region exhibit opposite directions of conductivity change during the rapid diffusion of lochia. By retrieving node data corresponding to this region's time in the diffusion phase recording chain, a complete electrical change curve segment can be extracted. Each time segment extraction should include the node's initial response time, duration of change, and magnitude of change, ensuring that the time segment comprehensively reflects the electrical dynamics within the potential reversal region. During extraction, the temporal sequence structure of the diffusion phase recording chain ensures that the response data of each node is continuously arranged on the time axis, without misalignment or omission. Through this stage of processing, the potential reversal suspected band can be converted from spatial region information into an operable set of time segments, providing accurate data input for subsequent differential change decomposition.

[0033] In the differential change unfolding stage, the extracted time segments are used as time windows to unfold and compare the electrical change trajectories of adjacent flexible sensing nodes. Each pair of adjacent nodes often exhibits subtle differences in response speed and change amplitude during lochia diffusion. When the liquid diffusion front crosses a certain region, nodes closer to the front respond earlier, while nodes further away respond slightly later. By unfolding the electrical change sequences of adjacent nodes along the time axis, this relationship of response delay and inconsistency can be clearly observed. To ensure the continuity of the unfolded time dimension, the time markers of adjacent nodes in the diffusion phase recording chain are used as the benchmark points for unfolding, ensuring that the differential changes in each time segment are extended forward with the same time step. During the unfolding process, the focus is on the differences in the direction, amplitude, and response start point of the change curves of adjacent nodes within the same time interval, thus forming a continuous differential change sequence. This differential change sequence not only reveals the asynchronous response between nodes within the potential reversal region but also provides data for subsequent labeling of phase-misaligned contacts. This time-expansion-based differential analysis method can transform complex node response relationships into a traceable temporal structure, ensuring the accuracy and continuity of subsequent misalignment annotations.

[0034] In the phase misalignment annotation stage, for the asynchronous response regions identified during the differential change unfolding stage, the change trends of adjacent nodes are gradually traced along the time axis to mark the positions of phase misalignment contacts. A phase misalignment contact refers to a specific point where, within the same time period, the electrical change trajectories of two adjacent flexible sensing nodes show temporal discrepancies, reversed amplitudes, or inconsistent directions of change. These contacts typically appear at moments of drastic velocity changes or uneven local penetration at the lochia diffusion front. By overlaying the electrical change trajectories of adjacent nodes on a unified time axis, regions with consistent phase and those with phase misalignment can be clearly distinguished. When the electrical response of adjacent nodes exhibits a time delay exceeding a preset time resolution limit, that location can be marked as a phase misalignment contact. To ensure the continuity of annotation, all contacts are annotated according to the temporal order of the diffusion phase record chain, ensuring that each contact can be traced back to the original node response sequence on the time axis. Simultaneously, the annotation process should incorporate spatial location information from the diffusion front sequence table, ensuring that each contact not only has a time coordinate but also a spatial index, thereby enabling spatiotemporal correlation analysis in subsequent stages. Through this stage of operation, the asynchronous response phenomenon that was originally scattered in time segments is accurately located as a discrete set of contacts, laying the foundation for generating a list of misaligned contacts.

[0035] During the contact point list generation phase, all phase misalignment contacts marked in the previous phase are organized and categorized according to chronological and spatial distribution order, forming a misalignment contact point list with complete time series and spatial correspondence. This list, with time as the main axis, uniformly records the time coordinates, corresponding node numbers, adjacent node pair numbers, and spatial location indexes of each phase misalignment contact point, and summarizes them according to the arrangement order of the diffusion phase record chain. To maintain temporal continuity, the arrangement of all contacts must be consistent with the diffusion phase record chain, allowing the list to serve as the input data structure for the subsequent time-grid traction control phase. The generation of the misalignment contact point list not only realizes the transformation from time segments to discrete contacts but also eliminates duplicate or overlapping contacts through the organization process, ensuring that each record entry independently corresponds to a specific response anomaly location within a time period. By uniformly encoding temporal and spatial information, the misalignment contact list forms a multi-dimensional data index structure that can be directly called in the subsequent control phase to determine the time freeze point or time traction sequence. Thus, the entire process of extracting time segments along the suspected potential reversal band, unfolding differential changes, and marking phase misalignment contacts is completed in a closed loop. The original electrical anomalies in the suspected potential reversal band are concretized into a set of locatable and traceable contacts, thereby enabling subsequent time grid construction and traction control to be implemented at more precise time points.

[0036] The anti-reversal time grating generation module traces the diffusion phase record chain around the list of misaligned contacts, arranges anti-reversal time gratings before and after each phase misaligned contact, limits the superposition order of adjacent flexible sensing nodes by inserting controlled micro-time slot silence intervals, and generates time grating traction drafts. To effectively suppress potential reversal and curve abrupt changes caused by asynchronous responses from adjacent flexible sensing nodes during lochia diffusion monitoring, the diffusion phase recording chain can be traced back based on the list of misaligned contacts. Anti-reversal time gratings can be arranged before and after each phase misaligned contact, and the superposition order of adjacent flexible sensing nodes can be limited by inserting controlled micro-timeslot silence intervals. Finally, a time grating traction draft is generated to achieve precise traction of the time series and time-series management of the response order. The specific steps are as follows:

[0037] During the contact backtracking and localization phase, based on the time coordinates and spatial location information recorded in the generated list of misaligned contacts, a backtracking operation is performed along the diffusion phase recording chain to determine the accurate position of each phase misaligned contact in the time series and the response states of its preceding and following adjacent nodes. The diffusion phase recording chain contains the electrical change trajectory, response time, and diffusion front advancement sequence of each flexible sensing node. By backtracking this recording chain, the response points of the preceding and following nodes corresponding to each misaligned contact can be obtained. During this process, it is necessary to maintain the integrity of the time sequence, ensuring that the positioning of each misaligned contact conforms to its actual position on the time axis while preserving its relative relationship in the spatial sequence of the diffusion front. Contact backtracking is not only used to determine the absolute time point of the misaligned contact, but more importantly, it provides preceding and following reference intervals for the subsequent time grid arrangement, ensuring that there are clear buffer time periods available for adjustment before and after each misaligned contact. In this way, a contextual time frame can be established for each misaligned contact, providing data boundaries and time windows for the setting of the anti-reversal time grid.

[0038] During the anti-reversal time grid arrangement stage, anti-reversal time grids are set before and after each phase misalignment contact using the time window obtained from contact backtracking positioning. An anti-reversal time grid is a time barrier used to constrain the superposition relationship of adjacent flexible sensing nodes in the time dimension. Its function is to prevent adjacent nodes from producing opposite electrical response trends at the same time by dividing the time series into stages. During arrangement, the starting point of each anti-reversal time grid is usually set within the response time segment before the misalignment contact, and the ending point is set within the stable response interval after the misalignment contact. The length of the time grid can be dynamically adjusted according to the average response interval of each node in the diffusion phase recording chain to ensure that the time grid can completely cover the time fluctuation interval where the misalignment contact is located. During arrangement, it is necessary to ensure that the boundaries between different time grids do not overlap to prevent the control area from interfering with the time order of other normal response nodes. In this way, each misalignment contact is demarcated by a time grid, forming a set of non-interfering but closely connected time protection zones, allowing for precise control within a clear time frame when subsequently inserting micro-timeslot silence intervals.

[0039] During the micro-timeslot silence interval insertion phase, controlled micro-timeslot silence intervals are inserted within each time grid according to the response sequence of adjacent flexible sensing nodes, surrounding the pre-arranged anti-reversal time grid. The purpose of these silence intervals is to artificially establish a brief signal gap in the time dimension, ensuring a sequential order in the superimposed responses of adjacent nodes, thereby avoiding potential reversal and signal cancellation caused by simultaneous responses. The duration of each micro-timeslot silence interval is set based on the node response time resolution in the diffusion phase recording chain, ensuring that the interval time guarantees the response sequence without affecting the overall acquisition continuity. During insertion, the time difference between each silence interval and the response points of the preceding and following nodes must be kept constant to form a stable time connection pattern. The silence intervals are set according to the correspondence between the diffusion direction and the node number; that is, silence intervals are inserted sequentially along the direction of diffusion front advancement, so that the nodes traversed by the liquid diffusion front respond one by one according to the time gradient, forming a linear response chain from front to back. By inserting micro-time slots at the time level, the coupling effect of instantaneous signal superposition between adjacent nodes can be effectively weakened, so that the electrical change trend presents a progressive continuity in time, thus laying a structured foundation for the generation of subsequent time grid traction drafts.

[0040] During the time-grid traction draft generation stage, all anti-reversal time grids and their internally inserted micro-slot silence intervals are integrated into a continuous time-control structure according to time sequence, generating a complete time-grid traction draft. The time-grid traction draft, with time as its main axis, records the time period of each misaligned contact, the response order of preceding and following nodes, the coverage area of ​​the anti-reversal time grid, and the duration of the silence interval. The traction draft has a layered structure, containing both time annotations at the misaligned contact level and time dependencies at the node response level, thus forming a complete time-traction path. Through the time-grid traction draft, the sequential order of node responses throughout the entire lochia diffusion process can be time-sequentially tractioned, enabling each node to perform data acquisition and overlay operations according to a predetermined rhythm during future sampling. The time-grid traction draft not only records time constraint parameters but also preserves the physical correlation of response changes through the original information of the diffusion phase recording chain, allowing it to be directly used for time freezing, reverse traction, or segmented playback operations in subsequent execution stages. Thus, through the continuous execution of contact point backtracking positioning, time grid arrangement, silent interval insertion and traction draft generation, the transformation from misaligned contact point data to an executable time traction structure was completed, realizing the temporal constraint on the superposition order of adjacent flexible sensing nodes.

[0041] The reverse diffusion traction control module performs reverse diffusion traction control based on the time grating traction draft. Before the negative mutation occurs, it briefly freezes the superimposed aperture of adjacent flexible sensing nodes and replays the sampling rhythm in segments according to the time grating traction sequence. It outputs a bleeding volume change curve that maintains a continuous increasing trend and generates a stable and consistent early warning signal. To ensure the temporal continuity and response stability of the bleeding volume change curve in postpartum hemorrhage monitoring, reverse diffusion traction control can be performed based on the time-grating traction draft. This involves briefly freezing the superimposed aperture of adjacent flexible sensor nodes before a negative abrupt change occurs, and then replaying the sampling rhythm in segments according to the time-grating traction sequence. This maintains a continuous increasing trend in the bleeding volume change curve and generates a stable and consistent early warning signal during the sampling output stage. The specific steps are as follows:

[0042] During the time-traction command loading phase, the time-grating traction draft generated in the previous phase is used as the basis for time series control. According to the recorded time sequence, node number, silence interval, and response start and end times, the sampling control information of each flexible sensing node is loaded into the time execution channel. The time-grating traction draft defines the response order and time distribution interval of each node during the diffusion process. This information is converted into a time command set that can directly drive the sampling rhythm during loading. To ensure that the response rhythms between different flexible sensing nodes strictly follow the time-grating sequence, all time-traction commands are executed with a unified time reference, thus making the entire traction process continuous and seamless in time. During the loading phase, the time-traction commands also need to be matched with the electrical response characteristics of each node recorded in the diffusion phase recording chain, so that each node can accurately correspond to its physical position and response time window in the diffusion phase recording chain during execution. By loading the time-traction commands, the response sequence relationship between flexible sensing nodes can be established in the time dimension, so that the execution of subsequent short-term freeze and playback rhythms has a clear starting order and execution boundary.

[0043] During the short-time freeze trigger execution phase, based on the time nodes of the time traction command set, a short-time freeze is applied to the superposition aperture of adjacent flexible sensing nodes within the time segment preceding the impending negative abrupt change. Short-time freeze refers to temporarily halting the superposition of the electrical responses of multiple nodes within an extremely short time interval, thereby preventing signal abrupt changes caused by instantaneous reverse superposition within that interval. The freeze trigger point is determined based on the contact boundaries in the time grating traction draft and the response delay data of the diffusion phase recording chain. When the system detects that the response interval between adjacent nodes enters the predetermined inversion sensitive area, a freeze command is immediately triggered. During the freeze, individual sampling of each node continues, but its output is temporarily excluded from the global superposition calculation to avoid brief drops or fluctuations in the curve due to local response asynchrony. The duration of the freeze is dynamically determined based on the average response interval between nodes, typically set to cover the period where signal inversion may occur. Through the execution of this phase, the transient interference effects of adjacent nodes can be effectively isolated without changing the sampling density, providing a clean time window for subsequent segmented playback of the sampling rhythm.

[0044] During the segmented playback sampling rhythm implementation phase, the node sampling data generated during the previous short-term freeze period is played back in segments according to the time grid order defined in the time grid traction draft. Segmented playback refers to releasing and re-overlaying the electrical response data of each node temporarily stored during the freeze period segment by segment according to the time traction order, so that the hemorrhage change curve maintains a continuous increasing trend on the time axis. During playback, the data release order of each flexible sensing node strictly follows the time period division of the time grid traction draft, that is, the next node can only enter the overlay interval after the playback of the previous node is completed. This ensures that the direction of the response data of each node on the time axis is always consistent with the physical direction of lochia diffusion, avoiding signal distortion caused by data overlap or reverse output. During playback, the output of each segment not only restores the data information of the frozen period, but also, through smooth connection with the previous time period, makes the overall hemorrhage change curve present a natural increasing trend both visually and logically. This segmented playback method enables the system to maintain the integrity of the sampling rhythm when facing transient interference or asynchronous response, thereby ensuring the timing stability of the entire data stream.

[0045] In the continuous curve and early warning signal output stage, the blood loss change curve generated by segmented playback and superposition is continuously output over time, and a stable and consistent early warning signal is generated based on the dynamic trend of the curve. The key to this stage is the time sequence control formed by the time grating traction structure, which ensures that the final output curve maintains a monotonically increasing trend in the time dimension, thus truly reflecting the actual changes in postpartum hemorrhage. During the curve output process, all data from the flexible sensor nodes are superimposed according to the traction sequence, and their time intervals and response rhythms completely correspond to the structure of the time grating traction draft. This ordered output method effectively eliminates negative abrupt changes caused by instantaneous reversals, signal misalignments, or local noise, ensuring a smooth transition of the entire curve during dynamic changes. The generation of the early warning signal is based on the continuously increasing characteristics of the blood loss change curve. Before the curve exceeds the preset rate of change or critical threshold, an early warning signal is automatically issued for nursing staff to intervene in advance. By combining the triggering conditions of the early warning signal with the time traction structure, the early warning output no longer relies on single-point fluctuations but on the trend judgment of continuous time periods, thereby improving the accuracy and timeliness of monitoring. Thus, through the complete link of time-driven command loading, short-term freeze trigger execution, segmented playback sampling rhythm implementation, and continuous curve and early warning signal output, the full-time closed-loop control from time-grid traction draft to final early warning output has been realized, enabling the lochia diffusion monitoring process to form a stable response order in the time dimension.

[0046] This invention establishes a diffusion phase recording chain within the nursing pad and combines it with potential reversal identification and phase misalignment analysis mechanisms to synchronously constrain the response relationships of multiple flexible sensing nodes in both time and space dimensions. By tracking the lochia diffusion process over time and determining directional consistency, the invention maintains the orderliness of node responses during rapid fluid penetration, fundamentally avoiding signal direction reversal and phase misalignment caused by uneven local diffusion. This ensures that the bleeding volume change curve maintains a continuous and stable increasing trend during the acquisition phase, thereby achieving accurate mapping and real-time monitoring of bleeding dynamics.

[0047] This invention utilizes the synergistic effect of an anti-reversal time grating and back-diffusion traction control to construct a controlled temporal rhythm in the signal superposition stage, ensuring a clear sequential order for the sampling outputs of each flexible sensing node. Through short-term freeze and segmented playback mechanisms, it effectively suppresses negative abrupt changes caused by asynchronous node responses, ensuring that the bleeding volume curve maintains temporal continuity and numerical consistency even in complex diffusion environments. This approach improves the stability and reliability of the early warning signal, providing nursing staff with a dependable basis for judging dynamic bleeding trends.

[0048] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A smart lochia care pad early warning system for postpartum hemorrhage monitoring, characterized in that, It includes a diffusion phase construction module, a potential reversal identification module, a phase misalignment analysis module, an anti-reversal time grating generation module, and a reverse diffusion traction control module: The diffusion phase construction module establishes a diffusion phase recording chain around the early warning of intelligent lochia care pads, collects the electrical change trajectories of multiple flexible sensor nodes in a unified time sequence, and forms a diffusion front sequence table based on the diffusion phase recording chain. The potential reversal identification module, based on the diffusion phase recording chain and the diffusion front sequence table, compares the electrical change trajectory direction of adjacent flexible sensing nodes one by one, locks out continuous segments where the change direction is reversed, and converges them to form a potential reversal suspected band. The phase misalignment analysis module extracts the corresponding time segment along the potential reversal suspected band, performs time expansion on the differential changes of adjacent flexible sensing nodes based on the diffusion phase recording chain, marks the phase misalignment contacts with asynchronous changes, and forms a list of misalignment contacts. The anti-reversal time grating generation module traces the diffusion phase record chain around the list of misaligned contacts, arranges anti-reversal time gratings before and after each phase misaligned contact, limits the superposition order of adjacent flexible sensing nodes by inserting controlled micro-time slot silence intervals, and generates time grating traction drafts. The reverse diffusion traction control module performs reverse diffusion traction control based on the time grating traction draft. Before the occurrence of negative mutation, it briefly freezes the superimposed aperture of adjacent flexible sensing nodes and replays the sampling rhythm in segments according to the time grating traction sequence. It outputs a bleeding volume change curve that maintains a continuous increasing trend and generates a stable and consistent early warning signal.

2. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 1, characterized in that, The process of forming the diffusion front ordinal table is as follows: A diffusion phase recording chain is established around the early warning of intelligent lochia care pads. Multiple flexible sensing nodes are evenly distributed in the liquid absorption layer structure, and each flexible sensing node is kept in signal communication with the acquisition and control unit. After diffusion-triggered detection, the acquisition and control unit starts the diffusion phase recording chain initialization program to independently number each flexible sensing node and set a unified time reference. The acquisition and control unit synchronously acquires the electrical change trajectories of multiple flexible sensing nodes at a uniform time interval, forming a diffusion time-phase recording chain; The acquisition and control unit sorts and maps the response time and spatial distribution of each flexible sensing node according to the diffusion phase recording chain, and generates a diffusion front sequence table.

3. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 2, characterized in that, The steps for generating the potential reversal suspect band are as follows: The electrical change trajectories of multiple flexible sensing nodes are extracted using a diffusion time-phase recording chain, and the change direction of conductivity or dielectric constant of each flexible sensing node is analyzed in a time-based manner. Based on the spatial arrangement of the diffusion front sequence table, the electrical change trajectory directions of adjacent flexible sensing nodes are compared one by one to determine the node pairs with opposite change directions. By tracing the time intervals of node pairs with opposite directions of change along the diffusion phase recording chain, the start and end positions of the inversion segments are determined and continuous inversion segments are formed. Multiple inversion segments are aggregated according to their temporal sequence and spatial proximity to form a potential inversion band that is continuous on the time axis and adjacent in spatial distribution.

4. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 3, characterized in that, During the formation of the potential reversal zone, the boundaries of all reversal segments are referenced to the time nodes in the diffusion phase recording chain. The temporal continuity and spatial adjacency of the potential reversal zone are constrained by the node arrangement direction of the diffusion front sequence table.

5. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 3, characterized in that, The process of generating the misaligned contact list is as follows: Extract the corresponding time segment along the potential reversal suspected band, and extract the electrical change data corresponding to the time range based on the diffusion phase recording chain to obtain the electrical change curve segments of multiple flexible sensing nodes; Using the extracted time segments as time windows, the electrical change trajectories of adjacent flexible sensing nodes are unfolded over time and compared for differences, forming a continuous differential change sequence. Based on the differential change sequence, the changing trend of adjacent flexible sensing nodes is tracked along the time axis, and the positions of phase misalignment contacts with asynchronous responses are marked. The marked phase misalignment contacts are sorted and categorized according to time sequence and spatial distribution order to form a list of misalignment contacts with time sequence and spatial correspondence.

6. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 5, characterized in that, When forming the list of misaligned contacts, each phase misaligned contact is further indexed and encoded based on the spatial location information in the diffusion front sequence table, so that the list of misaligned contacts has both time coordinates and spatial coordinates, and is merged according to the arrangement order of the diffusion phase record chain.

7. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 5, characterized in that, The steps for generating a time-raster traction draft are as follows: Based on the time coordinates and spatial location information recorded in the list of misaligned contacts, a backtracking operation is performed along the diffusion phase recording chain to determine the time location of each phase misaligned contact and the response status of adjacent flexible sensing nodes. By utilizing the time window obtained from contact backtracking positioning, anti-reversal time grids are arranged before and after each phase misalignment contact, so that the superposition relationship of adjacent flexible sensing nodes is constrained in the time dimension. Controlled micro-timeslot silence intervals are inserted inside the anti-reversal time grating according to the response order of adjacent flexible sensing nodes; The anti-reversal time grid and its internal silent intervals are integrated in chronological order to generate a time grid traction draft that records the response order of nodes and the time coverage range.

8. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 7, characterized in that, The duration of the micro-slot silence interval inserted within the anti-reversal time grating is set according to the response time resolution of the flexible sensing nodes in the diffusion phase recording chain, and the silence interval maintains a constant time difference between adjacent flexible sensing nodes.

9. The intelligent lochia care pad early warning system for postpartum hemorrhage monitoring according to claim 7, characterized in that, Based on the time-grating traction draft, reverse diffusion traction control is performed. Before the occurrence of negative mutation, the superimposed aperture of adjacent flexible sensing nodes is briefly frozen. The sampling rhythm is then replayed in segments according to the time-grating traction sequence to output a continuously increasing hemorrhage change curve and generate an early warning signal. The steps are as follows: Based on the time traction command, the sampling and control information of each flexible sensing node is loaded into the time execution channel according to the time sequence, node number, silent interval and response start and end time. Based on the time-driven command, the superposition aperture of adjacent flexible sensing nodes is temporarily frozen before the occurrence of negative mutation, in order to prevent signal mutation caused by instantaneous reverse superposition. Following the time grid sequence in the time grid traction draft, the node sampling data during the short-term freeze period is replayed in segments to ensure that the bleeding volume change curve maintains a continuous increasing trend on the time axis. The bleeding volume change curve after segmented playback is continuously output over time, and a stable and consistent early warning signal is generated based on the dynamic change trend of the curve.

10. A smart lochia care pad for monitoring postpartum hemorrhage, characterized in that, The intelligent lochia care pad includes the intelligent lochia care pad early warning system for postpartum hemorrhage monitoring as described in any one of claims 1-9.

Images