A multi-parameter monitoring system and method for mothers and infants integrating skin contact monitoring

Abstract

This application provides a maternal and infant multi-parameter monitoring system and method integrating skin contact monitoring. It includes: a contact sensing module for generating raw contact data based on continuous changes in pressure sensing data and the matching status of identity sensing data; a vital sign acquisition module for collecting vital sign data from the mother and newborn; a contact state modeling module for generating maternal and infant contact state segments and generating contact duration statistics and contact quality parameters; a time calibration module for generating contact context monitoring data; a correlation discrimination module for performing correlation discrimination on changes in contact quality and fluctuations in vital signs to generate contact-related monitoring events; and a monitoring result output module for generating maternal and infant multi-parameter monitoring results based on contact-related monitoring events. This application can improve data fusion accuracy, enhance the accuracy of temporal correlation, and improve the timeliness of complex risk identification.

Description

Technical Field

[0001] This application relates to the field of medical monitoring technology, and in particular to a maternal and infant multi-parameter monitoring system and method that integrates skin contact monitoring. Background Technology

[0002] Mother-infant skin-to-skin contact is a crucial aspect of early postpartum care and neonatal monitoring, providing healthcare professionals with essential information to assess the contact process, changes in vital signs, and care risks. With the application of bedside monitoring equipment, wearable sensors, and nursing information systems, mother-infant monitoring is gradually shifting from manual observation to data-driven monitoring. Current technologies typically monitor vital signs of both mother and newborn using parameters such as heart rate, blood oxygen saturation, respiration, and body temperature. Other methods utilize pressure sensors or inductive recognition to determine skin-to-skin contact and track the duration of contact.

[0003] However, in existing maternal and infant monitoring systems, skin contact monitoring and vital sign monitoring are mostly independent. Contact status data is usually only used to determine whether contact has occurred or the cumulative contact time, failing to serve as contextual information for vital sign analysis in multi-parameter monitoring decisions. When contact is interrupted, the contact area changes, or the contact quality deteriorates during maternal-infant contact, the system struggles to perform time-series correlation analysis between these contact changes and fluctuations in the newborn's vital signs such as blood oxygenation, heart rate, and respiration. Furthermore, different monitoring devices vary in data sampling frequency, transmission delay, and time stamping methods. Simply displaying data using a single timestamp makes it difficult to accurately determine the correspondence between changes in contact status and changes in physiological parameters.

[0004] Therefore, existing technologies still suffer from problems such as low integration of maternal and infant skin contact status with multi-parameter vital sign data, insufficient support for judging abnormal vital signs during the contact process, and untimely identification of complex risk scenarios, making it difficult to meet the needs of refined monitoring during maternal and infant skin contact care. Summary of the Invention

[0005] In view of this, embodiments of this application provide a maternal and infant multi-parameter monitoring system and method that integrates skin contact monitoring to solve the problems of insufficient integration of contact monitoring and vital signs, inaccurate temporal correlation, and delayed identification of complex risks in the prior art.

[0006] The first aspect of this application provides a maternal and infant multi-parameter monitoring system integrating skin contact monitoring, comprising: a contact sensing module for collecting pressure sensing data and identity sensing data during maternal and infant skin contact, and generating raw contact data based on the continuous changes in pressure sensing data and the matching status of identity sensing data; a vital sign acquisition module for collecting vital sign data of the mother and newborn, and configuring acquisition time information for the vital sign data; a contact state modeling module for analyzing the continuity, intensity change, and contact range of the raw contact data, generating maternal and infant contact state segments, and generating contact duration statistics and contact quality parameters based on the maternal and infant contact state segments; a time calibration module for time-aligning the raw contact data, contact duration statistics, contact quality parameters, and vital sign data with the state switching nodes in the maternal and infant contact state segments as calibration benchmarks, and generating contact context monitoring data; an association discrimination module for association discrimination of contact quality changes and vital sign fluctuations based on the contact context monitoring data and the physiological response time range after contact state changes, and generating contact-related monitoring events; and a monitoring result output module for generating maternal and infant multi-parameter monitoring results based on contact-related monitoring events.

[0007] The second aspect of this application provides a method for multi-parameter maternal and infant monitoring based on the system of the first aspect, which integrates skin contact monitoring. The method includes: collecting pressure sensing data and identity sensing data during maternal and infant skin contact, and generating raw contact data based on continuous changes and matching states; collecting vital sign data of the mother and newborn, and configuring collection time information; modeling the contact state of the raw contact data to generate maternal and infant contact state segments, contact duration statistics, and contact quality parameters; using the state switching nodes in the maternal and infant contact state segments as calibration benchmarks, mapping the vital sign data to the corresponding maternal and infant contact state segments to generate contact context monitoring data; and performing correlation discrimination on the contact context monitoring data according to the physiological response time range after contact state changes to generate contact-related monitoring events, and generating multi-parameter maternal and infant monitoring results accordingly.

[0008] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: The system employs a contact sensing module to collect pressure and identity sensing data during mother-infant skin contact, generating raw contact data based on continuous changes in pressure sensing data and the matching status of identity sensing data. A vital signs acquisition module collects vital signs data from both mother and newborn, configuring acquisition time information for the data. A contact state modeling module analyzes the continuity, intensity changes, and contact range of the raw contact data, generating mother-infant contact state segments and producing contact duration statistics and contact quality parameters based on these segments. A time calibration module uses state switching nodes within the mother-infant contact state segments as calibration benchmarks to align the raw contact data, contact duration statistics, contact quality parameters, and vital signs data in time, generating contact context monitoring data. A correlation discrimination module, based on the contact context monitoring data and the physiological response time range after contact state changes, correlates and discriminates changes in contact quality and fluctuations in vital signs, generating contact-related monitoring events. A monitoring result output module generates multi-parameter monitoring results for mother and infant based on contact-related monitoring events. This application improves data fusion accuracy, enhances the accuracy of temporal correlation, and improves the timeliness of complex risk identification. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the structural composition of a maternal and infant multi-parameter monitoring system that integrates skin contact monitoring, provided in an embodiment of this application. Figure 2 This is a flowchart illustrating a multi-parameter monitoring method for mothers and infants that integrates skin contact monitoring, provided in an embodiment of this application. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] In existing technologies, maternal and infant monitoring typically focuses on vital sign monitoring. This involves collecting parameters such as heart rate, blood oxygenation, respiration, and body temperature from the mother and newborn using bedside monitoring devices, wearable devices, or non-invasive sensing devices, and triggering alarms based on the threshold ranges of each parameter. Additionally, some existing solutions can determine whether the mother and baby are in skin-to-skin contact through pressure sensing and sensor recognition, and can statistically analyze the total duration of contact, continuous contact time, or intermittent contact time, thus replacing manual recording of the skin-to-skin contact process.

[0013] However, in existing maternal and infant monitoring systems, skin contact monitoring and vital sign monitoring are usually independent. Contact status data is mostly used for individual display or duration statistics, failing to serve as contextual information for multi-parameter vital sign analysis in monitoring decisions. When contact is interrupted, the contact area changes, or the contact quality deteriorates during mother-infant contact, the system struggles to accurately correlate these changes with fluctuations in the newborn's heart rate, blood oxygenation, respiration, and other vital signs. Furthermore, differences in data sampling frequency, transmission delay, and time stamping methods among different monitoring devices lead to inaccurate temporal correspondences between changes in contact status and changes in vital signs, thus affecting the timeliness and accuracy of identifying complex risk events.

[0014] Therefore, the actual technical problem this application aims to solve is that existing technologies suffer from insufficient fusion of contact monitoring and vital signs, inaccurate temporal correlation, and delayed identification of complex risks. This technical problem is not simply a data display issue, but rather involves how to fuse contact status data, contact quality data, and vital signs data under a unified time reference, and further form correlation results that can be used for multi-parameter monitoring and judgment of mothers and infants.

[0015] To address the aforementioned technical issues, this application proposes a multi-parameter maternal and infant monitoring system integrating skin contact monitoring. This system collects pressure sensing data and identity sensing data during maternal-infant skin contact through a contact sensing module, and generates raw contact data based on the continuous changes in pressure sensing data and the matching status of identity sensing data. It also collects vital sign data from the mother and newborn through a vital sign acquisition module, configuring the acquisition time information. Furthermore, a contact state modeling module analyzes the continuity, intensity changes, and contact range of the raw contact data to generate maternal-infant contact state segments, and further generates contact duration statistics and contact quality parameters.

[0016] Building upon this foundation, this application further uses the state transition nodes within the mother-infant contact state segment as a calibration benchmark to perform time alignment on the raw contact data, contact duration statistics, contact quality parameters, and vital sign data, generating contact context monitoring data. Subsequently, the system correlates changes in contact quality and fluctuations in vital signs according to the physiological response time range following changes in contact state, generating contact-related monitoring events, and then generating multi-parameter monitoring results for the mother and infant based on these events. Thus, skin contact state is no longer merely treated as independent statistical data, but rather participates in the analysis as contact context information in multi-parameter monitoring judgment.

[0017] Through the above technical solutions, this application can improve the fusion accuracy of maternal and infant skin contact data and vital sign data, enhance the accuracy of the temporal correlation between changes in contact status and fluctuations in vital signs, and improve the timeliness of identifying complex risk scenarios such as abnormal contact accompanied by physiological abnormalities. Simultaneously, this application also enables monitoring results to simultaneously reflect contact status, contact duration, contact quality, changes in vital signs, and their correlation criteria, providing more complete data support for refined monitoring during maternal and infant skin contact care.

[0018] The specific components and functions of a maternal and infant multi-parameter monitoring system integrating skin contact monitoring provided in this application will be described in detail below with reference to the accompanying drawings and specific embodiments. Figure 1 This is a schematic diagram of the structural composition of a maternal and infant multi-parameter monitoring system integrating skin contact monitoring, as provided in an embodiment of this application. Figure 1 As shown, the system may specifically include the following components: The contact sensing module 101 is used to collect pressure sensing data and identity sensing data during the skin contact process between mother and baby, and generate raw contact data based on the continuous changes in pressure sensing data and the matching status of identity sensing data. The vital signs acquisition module 102 is used to collect vital signs data of the mother and newborn, and to configure the acquisition time information for the vital signs data; The contact state modeling module 103 is used to analyze the continuity, intensity change and contact range of the original contact data, generate mother-infant contact state segments, and generate contact duration statistics and contact quality parameters based on the mother-infant contact state segments. The time calibration module 104 is used to time-align the original contact data, contact duration statistics, contact quality parameters and vital sign data with the state switching nodes in the mother-infant contact state segment as the calibration benchmark, and generate contact context monitoring data. The association discrimination module 105 is used to perform association discrimination on changes in contact quality and fluctuations in vital signs based on contact context monitoring data and according to the physiological response time range after changes in contact status, and generate contact-related monitoring events. The monitoring result output module 106 is used to generate multi-parameter monitoring results for mothers and infants based on contact-related monitoring events.

[0019] In some embodiments, the contact sensing module is used to establish a binding relationship between the pressure acquisition channel and the identity sensing channel based on the maternal and infant monitoring session; Baseline calibration, continuous sampling, and extraction of change segments are performed on pressure sensing data to obtain contact intensity change information; The matching status between the current contact person and the maternal and infant monitoring session is determined based on the identity sensing data; The contact intensity change information, matching status, and corresponding acquisition time sequence are associated and encapsulated to generate raw contact data.

[0020] Specifically, before the mother and baby enter the skin-to-skin contact monitoring system, a mother-baby monitoring session is created, and the mother's wristband identifier, the newborn's wristband identifier, the contact pad device identifier, and the bedside monitoring terminal identifier are written into the same session record. Based on this mother-baby monitoring session, the contact sensing module assigns channel numbers to the pressure acquisition channels located in the contact area on the mother's chest and assigns sensing numbers to the identity sensing channels located on the edge of the contact pad or the newborn patch. Then, a binding relationship is established between the pressure acquisition channels, the identity sensing channels, and the mother-baby monitoring session. After binding is complete, the system assigns all subsequently collected pressure sensing data and identity sensing data to this mother-baby monitoring session, avoiding data confusion caused by temporary contact between different beds, different newborns, or medical staff.

[0021] During the pressure sensing data acquisition process, the contact sensing module first acquires a segment of unloaded pressure data in the initial state before mother-infant skin contact is established. This unloaded pressure data is used as the baseline data for the current pressure acquisition channel, and the baseline data is calibrated by considering the placement of the contact pad, changes in the mother's body position, and environmental disturbances. Subsequently, the contact sensing module continuously samples the pressure acquisition channel according to a preset sampling period, obtaining pressure sensing data arranged in chronological order. For the acquired pressure sensing data, the contact sensing module removes short-term spike interference and outliers exceeding the device's range, and calculates the pressure increment, duration, and trend corresponding to each sampling moment based on the calibrated pressure baseline. When the pressure increment at multiple consecutive sampling points is higher than the contact determination benchmark, the system marks the corresponding data segment as a contact enhancement segment; when the pressure increment decreases and the duration does not reach the complete disengagement determination condition, the system marks the corresponding data segment as a contact weakening segment; when the pressure increment is lower than the contact determination benchmark for a long time, the system marks the corresponding data segment as a disengagement candidate segment. Through the above segment extraction, the contact sensing module generates contact intensity change information that reflects the contact establishment, contact maintenance, contact weakening, and contact disappearance processes.

[0022] Meanwhile, the identity sensing channel continuously acquires identity sensing data corresponding to the current contact object. This identity sensing data can originate from wristband sensing, near-field recognition, or device pairing recognition, and is used to determine whether the mother and newborn near the current contact area belong to an established maternal-infant monitoring session. The contact sensing module matches the identity sensing data with the mother's and newborn's identifiers in the session record. If the sensing result simultaneously satisfies both the mother's and newborn's identifiers, the current contact object is determined to match the maternal-infant monitoring session; if only the mother's identifier, only the newborn's identifier, or a non-session object identifier is detected, it is determined to be an incomplete match or a mismatch. For pressure changes caused by medical staff adjusting the contact pad, blankets pressing on the contact area, or other personnel approaching the device, the contact sensing module can distinguish these changes from actual maternal-infant skin contact data based on the matching status of the identity sensing data.

[0023] When generating raw contact data, the contact sensing module uses the acquisition time sequence as the organizational basis, associating and encapsulating contact intensity changes, matching status, and channel source information within the same time period. Each segment of raw contact data includes the segment start and end time, pressure baseline, pressure change amplitude, pressure duration characteristics, identity matching status, and session binding identifier. For example, when a newborn is placed on the mother's chest, if the pressure acquisition channel detects a continuous increase in pressure increment, and the identity sensing channel simultaneously confirms that both the mother's identifier and the newborn's identifier belong to the current session, the contact sensing module encapsulates this time period as contact establishment data matching the current maternal and infant monitoring session; if the pressure signal subsequently drops briefly but the identity sensing status remains matched, it is encapsulated as contact weakening data; if the pressure signal continuously disappears and the newborn's identity sensing signal leaves the contact area, it is encapsulated as departure candidate data.

[0024] Through the above implementation method, the contact sensing module can establish the data attribution relationship between the pressure acquisition channel and the identity sensing channel at the initial stage of data acquisition, and encapsulate the pressure change process and the contact object matching status synchronously into contact raw data, thereby providing a data foundation with session attribution, temporal continuity and object credibility for subsequent mother-infant contact status modeling, contact duration statistics and contact quality analysis.

[0025] In some embodiments, the vital signs acquisition module is used to establish the association between different monitored objects and corresponding vital signs acquisition channels according to the maternal and infant monitoring session; The vital signs data of the corresponding monitored object are obtained based on the vital signs collection channel, and the vital signs data are marked with the channel source and the valid data segment is identified. Based on the time reference of the acquisition device and the data reception sequence, the acquisition time information is configured for the vital signs data, and the vital signs data, channel source markers, and acquisition time information are associated and output.

[0026] Specifically, after a maternal-infant monitoring session is established, the vital signs acquisition module first reads the mother's identification, the newborn's identification, and the corresponding bed information from the session. Based on the nursing configuration, it then determines the maternal and neonatal vital signs acquisition channels. The maternal vital signs acquisition channel can connect to a maternal wristband monitoring device, a chest patch sensor, or a bedside monitor. The neonatal vital signs acquisition channel can connect to a neonatal pulse oximeter probe, a temperature patch, a respiratory monitoring patch, or a movement acquisition device. The vital signs acquisition module binds different acquisition channels to their corresponding monitored subjects, ensuring that the data collected by each channel has a clear subject attribution, preventing the mixing of maternal and neonatal data, or errors in data attribution due to switching between different bedside devices.

[0027] During the data collection process, the vital signs acquisition module continuously acquires vital sign data of the corresponding monitored individual based on the device type, sampling cycle, and data format of each vital signs acquisition channel. For newborns, data reflecting their current physiological state, such as heart rate, blood oxygen saturation, respiratory status, body temperature changes, and body movement, can be collected. For mothers, data reflecting their contact and cooperation status, such as heart rate, respiratory rhythm, body position changes, and local body temperature, can be collected. After receiving data from each channel, the vital signs acquisition module first generates a channel source marker based on the device access number and channel binding relationship, enabling the vital signs data to be identified as maternal data, newborn data, or auxiliary monitoring data.

[0028] The vital signs acquisition module also identifies valid data segments. Specifically, it determines whether the current data belongs to a valid data segment based on the signal quality from the acquisition device, the probe wearing status, the continuity of data, and the reasonableness of numerical changes. When the neonatal blood oxygen probe is stably worn, the signal quality meets the acquisition requirements, and blood oxygen data is continuously output, the system marks that time period as a valid blood oxygen data segment; when the probe is loose, the signal is interrupted, or the value shows a non-physiological jump in a very short time, the system marks the corresponding time period as an invalid or data segment to be verified. For maternal heart rate data, validity can also be identified based on continuous sampling status, device contact status, and abnormal jumps. Through this processing, the subsequent correlation analysis module can distinguish between real vital signs fluctuations and data distortion caused by abnormal device contact.

[0029] When configuring the acquisition time information, the vital signs acquisition module does not simply use the time when the data arrives at the system as the acquisition time, but rather processes it by combining the acquisition device's own time base and the data reception sequence. For acquisition devices with a device clock, the vital signs acquisition module reads the device's output time and calculates the transmission delay by combining it with the system's reception time; for acquisition devices without a complete timestamp, the vital signs acquisition module deduces the corresponding acquisition time according to the data reception order, channel sampling period, and buffer length. If there are batch uploads or interval uploads on the same vital signs acquisition channel, the system splits the batch data into continuous sampling points according to the channel sampling period and configures corresponding acquisition time information for each sampling point.

[0030] For example, during kangaroo care, the neonatal oxygen probe outputs oxygen saturation and pulse rate data every 2 seconds, while the maternal wristband outputs heart rate data every 5 seconds. The vital signs acquisition module establishes channel source markers for both the neonatal oxygen channel and the maternal heart rate channel, and configures acquisition time information according to device time and system receiving order. When the neonatal oxygen saturation data drops from 96% to 92% within 15 seconds of contact establishment, and the signal quality of that data segment is marked as valid, the system outputs the corresponding oxygen change along with the neonatal side channel source marker and acquisition time information. If, during the same period, the maternal wristband experiences a brief signal loss due to arm movement, the system marks that data segment as invalid to prevent it from directly participating in subsequent contact context analysis.

[0031] The vital signs acquisition module ultimately correlates and outputs vital signs data, channel source markers, valid data segment identifiers, and acquisition time information to form a vital signs data set with object attribution, channel source, and time reference. Through the above implementation method, this application can improve the accuracy of attribution and timing of maternal and neonatal vital signs data, reduce misjudgments caused by equipment delays, channel confusion, or invalid data in the analysis, and provide a reliable data foundation for subsequent temporal fusion and correlation discrimination between contact state segments and vital signs changes.

[0032] In some embodiments, the contact state modeling module is used to extract the stable range and abrupt range of the contact intensity change over time from the original contact data, and to determine candidate contact segments in combination with the matching state of the current contact object. Based on the baseline of contact intensity, the duration of the segment, and the trajectory of changes in the contact range, a state transition constraint analysis is performed on the candidate contact segments to generate mother-infant contact state segments with segment boundaries and state identifiers. Based on the duration, switching relationship, and stability of the contact range of the mother-infant contact state segments, statistical results of contact duration and contact quality parameters are generated.

[0033] Specifically, after receiving the raw contact data, the contact state modeling module first sorts the data according to the session identifier and collection sequence of the maternal and infant monitoring session, and reads the contact intensity change information, identity matching status, and channel source information. For continuously collected data within the pressure acquisition channel, the module uses the calibrated contact intensity baseline as a reference to calculate the intensity change amplitude, direction of change, and duration between adjacent sampling points. When the pressure change remains continuously within the preset fluctuation range, the corresponding time period is extracted as the stable interval; when the pressure change exceeds the abrupt change judgment range, or the contact intensity changes from below the contact judgment benchmark to above the contact judgment benchmark, or changes from above the contact judgment benchmark to a continuous decrease, the corresponding time period is extracted as the abrupt change interval.

[0034] When determining candidate contact segments, the contact state modeling module verifies the contact status in conjunction with the current matching status of the contact object. If the pressure intensity consistently meets the contact conditions within a stable interval, and the identity sensing result matches the mother-infant monitoring session, then the stable interval is determined as a valid candidate contact segment. If the pressure intensity experiences a short-term drop, but the identity matching status remains uninterrupted, then the interval is determined as a contact disturbance candidate segment. If the pressure intensity remains below the contact determination benchmark, and the identity sensing signal leaves the contact range, then the interval is determined as a departure candidate segment. For data segments where pressure signals exist but the identity matching status is inconsistent, the module excludes them or marks them as segments to be verified, preventing non-mother-infant contact from being identified as genuine contact.

[0035] When generating mother-infant contact state segments, the contact state modeling module performs state transition constraint analysis on candidate contact segments based on the contact intensity baseline, segment duration, and contact range change trajectory. The contact intensity baseline is used to determine the effective pressure change of each candidate segment relative to the no-load state, the segment duration is used to filter extremely short-term disturbances, and the contact range change trajectory is used to determine whether the contact area continuously covers or has undergone significant contraction or displacement. Based on the previous segment state, the current segment pressure change, and the contact range change, the module determines whether adjacent candidate segments satisfy the succession relationship of contact establishment, stable maintenance, short-term interruption, resumption of contact, or complete separation, and corrects the segment boundaries to generate mother-infant contact state segments with start and end times and state identifiers.

[0036] For example, in an early kangaroo care procedure, during the first 20 seconds after the newborn is placed on the mother's chest, the pressure intensity gradually increases from baseline while the identity matching status remains consistent; this phase is classified as the contact establishment segment. Subsequently, the pressure intensity fluctuates within a small range, and the contact area remains stable for 18 minutes; this is classified as the stable contact segment. When healthcare workers adjust the swaddle, the pressure area briefly contracts for 8 seconds, but the identity matching status is not lost; this is classified as the short-term interruption segment. If the pressure area subsequently expands again and the contact conditions are met, it is classified as the recovery contact segment. Based on these segments, the module accumulates the stable contact time, short-term interruption time, and number of state transitions, and generates contact quality parameters by combining the stability of the contact area.

[0037] Through the above implementation methods, the contact state modeling module can integrate continuous pressure changes, identity matching status, and contact range changes into the contact state modeling process, forming a mother-infant contact state segment with boundaries and status identifiers, and further generating contact duration statistics and contact quality parameters, thereby improving the accuracy of contact process statistics and the reliability of contact quality evaluation.

[0038] In some embodiments, based on the contact intensity baseline, fragment duration, and contact range change trajectory, a state transition constraint analysis is performed on candidate contact fragments to generate mother-infant contact state fragments with fragment boundaries and state identifiers, including: The contact determination threshold corresponding to the candidate contact segment is determined based on the contact strength baseline, and the candidate contact segment is filtered for short-term perturbation based on the segment duration. The contact range stability state of the candidate contact segment is determined based on the contact range change trajectory, and the contact judgment threshold, short-term disturbance filtering results and contact range stability state are input into the preset state transition rule; Based on preset state transition rules, the state succession relationship between adjacent candidate contact segments is determined, and mother-infant contact state segments with segment boundaries and state identifiers are generated.

[0039] Specifically, after obtaining candidate contact segments, the contact state modeling module first reads the contact intensity baseline corresponding to the maternal-infant monitoring session. This contact intensity baseline can be determined by the contact pad's unloaded state, the mother's resting posture, and the sensor's current zero-point offset. The module calculates the relative pressure increment of each candidate contact segment based on the contact intensity baseline and, combined with the newborn's weight range, contact pad placement, and sensor sensitivity corresponding to the current session, generates a contact determination threshold suitable for the current monitoring session. Therefore, the contact determination threshold is not a fixed value but is adjusted according to the current acquisition environment and the individual state of the mother and infant, enabling subsequent state judgment to adapt to different maternal-infant contact scenarios.

[0040] After determining the contact threshold, the contact state modeling module verifies the duration of candidate contact segments. For segments whose duration is less than a preset disturbance time and whose identity matching status remains consistent before and after, the module does not directly identify them as contact state transitions, but instead treats them as short-term disturbances for merging. For example, when a medical staff member slightly adjusts a blanket or when the mother's chest undergoes a short-term displacement due to breathing, the pressure signal may drop for 2 to 5 seconds, but the contact object still matches the maternal and infant monitoring session. In this case, the module merges the dropping segment into the consecutive contact segments before and after and retains the disturbance marker. For segments whose duration meets the state transition condition and whose pressure change direction is stable, the module retains them as segments to be transferred for judgment.

[0041] The contact state modeling module further determines the stable contact state of candidate contact segments based on the contact range change trajectory. Specifically, it can determine whether the contact range is in a stable coverage, local contraction, offset change, or continuous disappearance state based on the area change of the effective contact area in the pressure acquisition channel, the direction of contact center movement, and the degree of contact boundary contraction. For example, when a newborn is stably attached to the mother's chest, the effective contact area remains continuously distributed, and the contact center only changes with breathing within a small range; when the newborn's body slides or partially leaves the contact pad, the effective contact area gradually shrinks, and the contact center shifts in the same direction; when completely leaving the contact area, the effective contact area continuously disappears. The module inputs the above stable contact state, contact determination threshold, and short-term disturbance filtering results into a preset state transition rule.

[0042] Preset state transition rules are used to constrain the state continuity relationship between adjacent candidate contact segments. The module determines whether the current segment represents contact establishment, stable contact, short-term interruption, contact recovery, or complete detachment based on the state identifier of the previous segment, the pressure increment of the current segment, its duration, and the stability of the contact range. If the previous segment represents stable contact, and the pressure increment of the current segment is briefly below the contact determination threshold but the contact range has not completely disappeared, it is determined to be a short-term interruption. If the pressure increment of the current segment rises above the contact determination threshold again and the contact range stabilizes, it is determined to be a contact recovery. If the pressure increment remains below the contact determination threshold and the contact range continues to disappear, it is determined to be a complete detachment. The module corrects the segment start and end boundaries based on the state continuity relationship and assigns a corresponding state identifier to each segment, generating mother-infant contact state segments.

[0043] Through the above implementation methods, the system can perform continuous constraint analysis on candidate contact segments based on dynamic contact determination threshold, short-term disturbance filtering, and stable contact range status. This reduces erroneous segmentation caused by breathing displacement, swaddling adjustments, or instantaneous pressure fluctuations, making the generated mother-infant contact status segments more accurate in terms of segment boundaries and status identification. This provides a reliable basis for subsequent contact duration statistics, contact quality evaluation, and identification of contact-related monitoring events.

[0044] In some embodiments, the time calibration module is used to extract state switching nodes from mother-infant contact state segments and establish a contact-side calibration time reference based on the state switching nodes. Based on the contact side calibration time reference, segment boundary correction is performed on the original contact data and contact duration statistics, and corresponding segment affiliation relationships are configured for the contact quality parameters; Based on the collection time information and channel delay characteristics of vital sign data, the vital sign data is mapped to the corresponding mother-infant contact state segment to generate contact context monitoring data that includes contact state, contact duration, contact quality and changes in vital signs.

[0045] Specifically, after receiving segments of mother-infant contact status, the time calibration module first reads the start and end times, status identifiers, and switching relationships between adjacent segments according to the session identifier of the mother-infant monitoring session, and extracts the status switching nodes. These status switching nodes can include contact establishment nodes, stable contact transitioning to short-term interruption nodes, short-term interruption transitioning to contact resumption nodes, and stable contact transitioning to complete disengagement nodes. The time calibration module uses these nodes as the contact-side calibration time benchmark, establishing a reference position for contact status changes on a unified monitoring time axis. This ensures that subsequent vital sign data are not simply arranged based on device reporting time, but rather aligned and analyzed around changes in contact status.

[0046] After establishing a calibration time reference on the contact side, the time calibration module corrects the segment boundaries of the raw contact data. Specifically, the module traces back the corresponding pressure sensing data and identity sensing data based on the contact state switching node to confirm the pressure change trend, identity matching status, and sampling continuity before and after the state switch. If there is a slight deviation between the boundary point of the contact state segment and the pressure change abrupt change point, the segment boundary is corrected using the combined position of the pressure change abrupt change point and the identity matching change point. If there is a time period in the contact state segment that has been merged after being filtered by short-term disturbances, this time period is retained as an internal disturbance marker for the segment and is not separately split into a new contact state segment. For the contact duration statistics, the time calibration module recalculates the duration of each state segment according to the corrected segment boundaries and updates the statistical results such as effective contact duration, short-term interruption duration, and complete disengagement duration.

[0047] The time calibration module also configures segment affiliations for contact quality parameters. Specifically, each contact quality parameter can be associated with a corresponding mother-infant contact state segment based on the pressure range, contact range stability, and segment time boundary upon which its calculation is based. For example, the contact range stability, pressure distribution uniformity, and contact center drift formed within a stable contact segment are assigned to that stable contact segment; the degree of contact range contraction and recovery time formed within a short-term interruption segment are assigned to that short-term interruption segment. Thus, contact quality parameters can establish a clear correspondence with specific contact state segments, preventing the incorrect use of quality parameters after contact recovery in the analysis of the contact interruption phase.

[0048] For vital sign data, the time calibration module performs mapping processing based on the data acquisition time information and channel delay characteristics. Different vital sign acquisition channels may have different sampling periods and transmission delays. For example, the neonatal blood oxygenation channel uploads data every 2 seconds, the maternal heart rate channel uploads data every 5 seconds, and the respiratory monitoring channel may upload data after batch caching. The time calibration module corrects the vital sign data to the actual acquisition time based on the acquisition time, reception sequence, buffer length, and preset delay compensation value of each channel, and maps it to the corresponding mother-infant contact state segment. If the acquisition time of a certain blood oxygenation data point is within the physiological response observation range after a stable contact transitions to a short interruption, then the blood oxygenation data is assigned to that short interruption segment and its subsequent response window.

[0049] For example, in a mother-infant skin contact monitoring session, the system detected a stable contact transitioning to a short interruption at 10:15:20 and a return to stable contact at 10:15:32. The time calibration module uses this state transition node as a reference to correct the boundaries of the raw pressure data and assigns the contact quality parameters corresponding to the period from 10:15:20 to 10:15:32 to the short interruption segment. Simultaneously, it maps the neonatal blood oxygenation, heart rate, and respiratory data collected within 30 seconds after 10:15:20 to the physiological response window of this short interruption segment, generating contact context monitoring data that includes contact status, contact duration, contact quality, and changes in vital signs.

[0050] Through the above implementation methods, the time calibration module can complete the unified time alignment of contact data, contact quality data and vital sign data with the contact state switching node as the core, improve the temporal consistency between multi-source monitoring data, reduce the correlation misjudgment caused by channel delay and segment boundary deviation, and provide an accurate data foundation for the subsequent correlation judgment between contact quality changes and vital sign fluctuations.

[0051] In some embodiments, the association discrimination module is used to determine the contact state change node from the contact context monitoring data, and configure the corresponding physiological response time range based on the contact state change node; Within the physiological response time range, the characteristics of contact quality change and vital sign fluctuation are extracted and differentially matched with the stable monitoring baseline before the contact state change node. Based on the difference matching results, the temporal correlation strength between changes in contact and fluctuations in vital signs is determined, and contact-related monitoring events with event attribute information and correlation discrimination criteria are generated.

[0052] Specifically, after receiving the contact context monitoring data, the association and discrimination module first reads the status identifier, segment boundary, contact quality parameters, and corresponding vital sign data of each mother-infant contact state segment according to the mother-infant monitoring session, and then determines the contact state change nodes. Contact state change nodes can be nodes such as stable contact turning into short-term interruption, short-term interruption turning into contact resumption, stable contact turning into complete detachment, and contact quality changing from stable to declining. The module configures the physiological response time range according to the contact change type corresponding to different nodes. For example, for short-term interruption nodes, it configures a blood oxygen and heart rate response range of several seconds to tens of seconds after the node; for nodes with continuously declining contact quality, it configures a longer respiratory and body movement response range before and after the node, so that changes in vital signs can be analyzed within a time period that conforms to the physiological response pattern.

[0053] Within the physiological response time range, the correlation discrimination module extracts contact quality change features and vital sign fluctuation features, respectively. Contact quality change features can be formed by the degree of contact area contraction, the magnitude of decrease in contact intensity, the trend of contact center shift, and the duration of the segment; vital sign fluctuation features can be formed by the magnitude of neonatal blood oxygen saturation decrease, the magnitude of heart rate changes, respiratory rhythm changes, and body movement changes. The module also extracts a stable monitoring baseline from the stable contact segment prior to the contact state change point. This stable monitoring baseline represents reference data when the contact state is stable, the contact quality is stable, and vital signs are in a relatively stable state. Subsequently, the module performs difference matching between the contact quality change features and vital sign fluctuation features within the physiological response time range and the stable monitoring baseline to determine the magnitude, direction, and duration of the changes.

[0054] For example, during a mother-infant skin-to-skin contact monitoring session, the system detected a transition from stable contact to a short interruption at 10:20:10, with the contact area significantly shrinking within 8 seconds. The correlation discrimination module uses this state change node as a baseline, extracting the stable monitoring baseline for the 60 seconds preceding the node, and then reads the newborn's blood oxygen, heart rate, and respiratory data within the 30-second physiological response time range following the node. If the newborn's blood oxygen was stable at around 96% before the node, and then dropped to 92% within 15 seconds after the node, while the heart rate fluctuation increased, and this change was sequential in time with the shrinking contact area and decreased contact intensity, the module determined that there was a high temporal correlation between the contact change and the vital sign fluctuations. If the vital sign fluctuations occurred before the node, or occurred significantly beyond the physiological response time range, the module reduced the correlation strength between this change and the contact state change.

[0055] When determining the temporal correlation strength, the correlation discrimination module calculates the strength by comprehensively considering the order of changes, magnitude of changes, duration, data reliability, and type of contact state change in the difference matching results. For cases where a decline in contact quality occurs first, followed by fluctuations in vital signs within the physiological response time range, and the corresponding vital sign data is within a valid data segment, the module increases the temporal correlation strength. For data segments with probe loosening, missing data, or bodily movement interference, the module decreases the discrimination weight of the corresponding vital sign fluctuations. Subsequently, the module generates contact-related monitoring events based on the temporal correlation strength and preset event judgment conditions, and configures event attribute information and correlation discrimination criteria for each event. Event attribute information may include event type, trigger time, correlation level, and involved monitored objects. Correlation discrimination criteria may include contact state change nodes, contact quality change characteristics, vital sign fluctuation characteristics, stable monitoring baseline, and physiological response time range.

[0056] Through the above implementation methods, the association discrimination module can establish an association analysis process with physiological response constraints around the contact state change node, transforming the simple parallel display of contact quality changes and vital sign fluctuations into event discrimination results with temporal relationships, thereby improving the accuracy and timeliness of identifying contact abnormalities accompanied by physiological abnormalities, and providing traceable discrimination basis for the subsequent output of maternal and infant multi-parameter monitoring results.

[0057] In some embodiments, the temporal correlation strength between changes in contact and fluctuations in vital signs is determined based on the difference matching results, and contact-related monitoring events with event attribute information and correlation discrimination criteria are generated, including: Based on the difference matching results, the temporal consistency relationship among contact change characteristics, contact quality change characteristics, and vital sign fluctuation characteristics was determined. Based on the order of changes, duration, and data reliability within the physiological response time range, a weighted calculation of temporal consistency relationships is performed to obtain the temporal correlation strength. Based on the temporal correlation strength and preset event judgment conditions, contact-related monitoring events with event attribute information and correlation judgment criteria are generated.

[0058] Specifically, after obtaining the difference matching results, the correlation discrimination module first performs temporal processing on the contact change characteristics, contact quality change characteristics, and vital sign fluctuation characteristics. Contact change characteristics can be determined by state switching nodes in the mother-infant contact state segment; contact quality change characteristics can be determined by shrinkage of the contact area, decrease in contact intensity, shift of the contact center, and changes in contact stability; and vital sign fluctuation characteristics can be determined by changes in neonatal blood oxygenation, heart rate, respiration, and body movement data relative to a stable monitoring baseline. The module compares the occurrence time, direction of change, and duration of the above characteristics according to a unified monitoring timeline to determine whether there is a sequential relationship among the three types of characteristics, generating a temporally consistent relationship.

[0059] When determining temporal consistency, if the contact state first changes from stable contact to a short-term interruption, followed by a contraction in the contact range or a decrease in contact intensity in the contact quality parameters, and then vital sign fluctuations occur within the corresponding physiological response time range, it is determined to be positive temporal consistency. If the vital sign fluctuations occur before the contact change, or if the interval between the contact change and the vital sign fluctuations exceeds the corresponding physiological response time range, it is determined to be low consistency or inconsistent. If the contact quality change and vital sign fluctuations occur simultaneously, but there is probe loosening, missing data, or significant body movement interference, the relationship is marked as consistency to be verified. Through the above processing, the module can distinguish between associated fluctuations caused by contact changes, physiological fluctuations caused by non-contact factors, and spurious correlations caused by abnormal monitoring data.

[0060] When calculating the temporal correlation strength, the correlation discrimination module weights the temporal consistency relationship based on the order of changes, duration, and data reliability within the physiological response time range. The order of changes characterizes whether changes in contact quality, vital sign fluctuations, and other factors conform to a preset physiological response logic; duration characterizes whether the decline in contact quality and vital sign fluctuations persist for the time required for discrimination; and data reliability characterizes the acquisition quality, channel validity, and time stamp accuracy of the corresponding vital sign data. The module forms a comprehensive weight based on these factors and performs a weighted calculation on the temporal consistency relationship to obtain the temporal correlation strength.

[0061] For example, in an early skin contact monitoring session, the system determined that a stable contact transitioned to a short-term interruption at 10:32:05. From 10:32:05 to 10:32:12, the contact area continuously contracted. From 10:32:18, the neonatal's blood oxygen saturation dropped from 96% to 92%, and heart rate fluctuations increased. Because the vital signs fluctuations were within the physiological response time range corresponding to this short-term interruption, and the blood oxygenation channel signal quality was valid and channel delay had been corrected, the module assigned a high weight to the sequence of events and data reliability. If a persistent contact center shift was also detected during the same period, the contribution of contact quality changes to the temporal correlation strength was increased, ultimately generating a higher temporal correlation strength.

[0062] The association discrimination module generates contact-related monitoring events based on the temporal association strength and preset event judgment conditions. When the temporal association strength reaches the corresponding judgment condition, the module configures the event attribute information as a contact-related physiological fluctuation event and records the trigger time, the monitored object involved, the contact state type, the association level, and the event duration. When the temporal association strength does not reach the judgment condition but there is a decline in contact quality, the module generates a contact quality concern event. When the data credibility is insufficient, the module generates a monitoring event pending review. The association discrimination criteria include state switching nodes, contact quality change characteristics, vital sign fluctuation characteristics, stable monitoring baseline, physiological response time range, and temporal association strength.

[0063] Through the above implementation methods, the system can further transform the difference matching results into quantifiable temporal correlation strength, and generate contact-related monitoring events with event attribute information and correlation discrimination criteria based on the temporal correlation strength, thereby improving the accuracy of correlation discrimination between contact changes and vital sign fluctuations, and reducing misjudgments caused by occasional fluctuations, equipment interference and time mismatch.

[0064] In some embodiments, the monitoring result output module is used to merge and map the levels of contact-related monitoring events to generate corresponding monitoring risk identifiers. Based on the monitoring risk markers, the corresponding mother-infant contact status segments, contact duration statistics, contact quality parameters, changes in vital signs, and related judgment criteria are linked and organized to generate monitoring evidence data; Based on the monitoring evidence data, generate maternal and infant multi-parameter monitoring results and output the maternal and infant multi-parameter monitoring results to the corresponding monitoring terminal.

[0065] Specifically, upon receiving contact-related monitoring events, the monitoring result output module first merges events within the same monitoring period according to the mother-infant monitoring session, event occurrence time, and event attribute information. For events that occur consecutively within adjacent time ranges and are triggered by the same change in contact status, the module merges them into the same monitoring event chain. In cases where decreased contact quality, fluctuating vital signs, and reduced data reliability occur simultaneously, the module determines the primary and secondary events based on the order of event triggering and correlation criteria to avoid repeated alerts for the same abnormal process. After event merging, the module performs a level mapping based on event type, temporal correlation strength, magnitude of vital sign changes, duration, and data reliability, generating corresponding monitoring risk identifiers.

[0066] When generating monitoring risk indicators, the module can map events with short-term contact disturbances and stable vital signs to alert-level indicators; events with declining contact quality but whose vital sign changes have not yet reached the warning threshold to attention-level indicators; events with changes in contact status followed by fluctuations in effective vital signs and high temporal correlation to warning-level indicators; and events with contact disengagement or persistent abnormal contact quality accompanied by significant abnormal vital signs to intervention-level indicators. For example, during a kangaroo care session, the system identifies a newborn's contact transitioning from stable to a short-term interruption, followed by a continuous contraction of the contact area, a drop in blood oxygen from 96% to 92% within 20 seconds, and increased heart rate fluctuations. After the correlation discrimination module generates contact-related monitoring events, the monitoring result output module maps this event to a warning-level monitoring risk indicator based on high temporal correlation strength and effective blood oxygen data.

[0067] Subsequently, the monitoring results output module organizes monitoring evidence data based on monitoring risk markers. For each monitoring risk marker, the module extracts the corresponding mother-infant contact status segment, including the status marker, segment start and end times, and status transition nodes; extracts contact duration statistics, including the effective contact duration, short interruption duration, and cumulative number of interruptions; extracts contact quality parameters, including the stability of the contact range, changes in contact intensity, and contact center offset; extracts vital sign changes, including changes in neonatal blood oxygenation, heart rate, and respiration relative to a stable monitoring baseline; and extracts correlation criteria, including the physiological response time range, temporal correlation strength, and data reliability. The above data is organized and correlated according to the same event chain to form monitoring evidence data that can explain the source of risk and the judgment process.

[0068] Based on the monitoring evidence data, the monitoring result output module generates multi-parameter monitoring results for mother and infant. These results may include the current contact status, contact quality level, contact duration statistics, vital sign trends, monitoring risk indicators, event trigger time, and nursing prompts, and are output to the bedside monitoring terminal, nurse station terminal, or nursing information system. Healthcare staff can view the risk level and corresponding evidence data on the terminal, such as the time of short-term interruption, the time of blood oxygen decline, the trend of contact area contraction, and the correlation criteria provided by the system, thereby distinguishing between different scenarios of ordinary contact disturbance, decreased contact quality, and contact accompanied by physiological abnormalities.

[0069] Through the above implementation methods, the monitoring result output module can transform contact-related monitoring events into multi-parameter monitoring results for mothers and infants with risk levels and evidence support, reducing duplicate alarms and isolated alarms, and improving the readability, traceability and timeliness of clinical treatment of monitoring results.

[0070] The above embodiments have described in detail the specific components and functions of the maternal and infant multi-parameter monitoring system integrating skin contact monitoring of this application. The implementation process of the maternal and infant multi-parameter monitoring method integrating skin contact monitoring of this application will be described in detail below with reference to specific embodiments. Figure 2 This is a flowchart illustrating a multi-parameter monitoring method for mothers and infants that integrates skin contact monitoring, as provided in an embodiment of this application. Figure 2 As shown, the method may specifically include the following steps: S201, collects pressure sensing data and identity sensing data during the skin-to-skin contact process between mother and baby, and generates raw contact data based on continuous changes and matching status; S202, collect vital sign data of the mother and newborn, and configure the collection time information; S203, Model the contact status of the raw contact data to generate mother-infant contact status segments, contact duration statistics and contact quality parameters; S204, using the state switching node in the mother-infant contact state segment as the calibration benchmark, map the vital signs data to the corresponding mother-infant contact state segment to generate contact context monitoring data; S205. Based on the physiological response time range after the change in contact state, the contact context monitoring data is correlated and identified to generate contact-related monitoring events, and maternal and infant multi-parameter monitoring results are generated accordingly.

[0071] Specifically, during step S201, the system first establishes a maternal-infant monitoring session, binding the mother's identity identifier, the newborn's identity identifier, the pressure acquisition device identifier, and the identity sensing device identifier, ensuring that all subsequently collected data belongs to the same maternal-infant monitoring session. The contact sensing module acquires the pressure baseline in the initial state before skin contact is established, and continuously samples the pressure sensing data during the monitoring process, extracting information such as pressure increment, duration, trend of change, and changes in contact range.

[0072] Simultaneously, the contact sensing module acquires identity sensing data and matches it with the identity identifier in the maternal and infant monitoring session to determine whether the current contact object belongs to the target maternal and infant pair. The system associates and encapsulates the continuous changes in pressure sensing data, the matching status of identity sensing data, and the corresponding acquisition sequence to generate raw contact data, thereby avoiding misjudging touch by medical staff, pressure from bedding, or approach by non-target objects as genuine maternal and infant skin contact.

[0073] During step S202, the system establishes a connection between the mother, newborn, and corresponding vital sign acquisition channels based on the maternal and infant monitoring session, and acquires vital sign data of the mother and newborn through each vital sign acquisition channel. For the newborn, data such as blood oxygen saturation, heart rate, respiration, body temperature, and body movement can be collected; for the mother, data such as heart rate, respiration, body position, and local body temperature can be collected.

[0074] The system assigns channel source tags to each vital sign data point based on device access information and channel binding relationships, and identifies valid data segments by considering signal quality, probe wearing status, data continuity, and the reasonableness of numerical changes. For data uploaded by the acquisition devices, the system further configures acquisition time information based on the device's own time reference, system reception timing, sampling period, and channel delay characteristics, ensuring that the vital sign data has a clear object attribution, channel source, and time reference.

[0075] During step S203, the system performs contact state modeling on the raw contact data. The contact state modeling module uses a pressure baseline as a reference to extract stable and abrupt intervals of contact intensity variation over time from the raw contact data, and combines this with the matching status of the identity sensing data to determine candidate contact segments. Intervals where the pressure intensity consistently meets the contact conditions and the identity matching status is consistent are identified as valid candidate contact segments; intervals where the pressure intensity drops briefly but the identity matching status remains uninterrupted are identified as contact disturbance candidate segments; and intervals where the pressure intensity consistently falls below the contact determination benchmark and the identity sensing signal leaves the contact range are identified as departure candidate segments.

[0076] Subsequently, the system determines a dynamic contact determination threshold based on the contact intensity baseline, performs short-term disturbance filtering in conjunction with the segment duration, and then determines the stable state of the contact range based on the contact range change trajectory. The system inputs the contact determination threshold, the short-term disturbance filtering results, and the stable state of the contact range into preset state transition rules to determine the state succession relationship between adjacent candidate contact segments, generating mother-infant contact state segments with segment boundaries and state identifiers. Based on the duration, switching relationship, and contact range stability of each mother-infant contact state segment, the system further generates contact duration statistics and contact quality parameters.

[0077] During step S204, the system uses the state switching nodes in the mother-infant contact state segment as the calibration benchmark to map vital sign data to the corresponding mother-infant contact state segment. The time calibration module extracts state switching nodes such as contact establishment, short interruption, contact resumption, and complete separation from the mother-infant contact state segment, and establishes a contact-side calibration time benchmark based on these state switching nodes. Based on the contact-side calibration time benchmark, the system performs segment boundary correction on the original contact data and contact duration statistics, and configures the contact quality parameters to the corresponding mother-infant contact state segment.

[0078] For vital sign data, the system corrects the vital sign data to the actual acquisition time based on the acquisition time information, channel sampling cycle, data reception sequence and channel delay characteristics, and maps it to the corresponding mother-infant contact state segment and its physiological response observation range, thereby generating contact context monitoring data that includes contact state, contact duration, contact quality and changes in vital signs.

[0079] In step S205, the system performs correlation discrimination on the contact context monitoring data according to the physiological response time range after the change in contact state. The correlation discrimination module identifies the contact state change node from the contact context monitoring data and configures the corresponding physiological response time range according to different contact change types. Within this physiological response time range, the system extracts contact quality change characteristics and vital sign fluctuation characteristics, and performs difference matching with the stable monitoring baseline before the contact state change node to determine whether there is a sequential relationship between contact change, contact quality change, and vital sign fluctuation. Subsequently, the system performs a weighted calculation of the temporal consistency relationship based on the order of change, duration, and data reliability to obtain the temporal correlation strength, and generates contact-related monitoring events based on the temporal correlation strength and preset event judgment conditions.

[0080] After generating contact-related monitoring events, the system further merges and maps the events to different levels, generating corresponding monitoring risk identifiers. It then correlates and organizes fragments of mother-infant contact status, statistical results of contact duration, contact quality parameters, changes in vital signs, and correlation criteria to form monitoring evidence data. Finally, the system generates multi-parameter monitoring results for mother and infant based on the monitoring evidence data and outputs them to bedside monitoring terminals, nurse station terminals, or nursing information systems. Through this method, this application can fuse mother-infant skin contact monitoring data with vital sign data in a unified time series, improving the accuracy of correlation judgment between changes in contact status and fluctuations in vital signs, and enhancing the timeliness of identifying complex risk scenarios such as abnormal contact accompanied by physiological abnormalities.

[0081] It should be understood that the sequence number of each step in the above method embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0082] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although the technical solutions of this application have been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A maternal and infant multi-parameter monitoring system integrating skin contact monitoring, characterized in that, include: The contact sensing module is used to collect pressure sensing data and identity sensing data during the skin-to-skin contact process between mother and baby, and to generate raw contact data based on the continuous changes in the pressure sensing data and the matching status of the identity sensing data. The vital signs acquisition module is used to collect vital signs data of the mother and newborn, and to configure acquisition time information for the vital signs data; The contact state modeling module is used to analyze the continuity, intensity change and contact range of the original contact data, generate mother-infant contact state segments, and generate contact duration statistics and contact quality parameters based on the mother-infant contact state segments. The time calibration module is used to time-align the original contact data, contact duration statistics, contact quality parameters and vital sign data with the state switching nodes in the mother-infant contact state segment as the calibration benchmark, and generate contact context monitoring data. The association discrimination module is used to perform association discrimination on changes in contact quality and fluctuations in vital signs based on the contact context monitoring data and the physiological response time range after changes in contact state, and generate contact-related monitoring events. The monitoring result output module is used to generate multi-parameter monitoring results for mother and baby based on the contact-related monitoring events.

2. The system according to claim 1, characterized in that, The contact sensing module is used to establish a binding relationship between the pressure acquisition channel and the identity sensing channel based on the maternal and infant monitoring session; Baseline calibration, continuous sampling, and extraction of change segments are performed on the pressure sensing data to obtain contact intensity change information; The matching status between the current contact object and the maternal and infant monitoring session is determined based on the identity sensing data. The contact intensity change information, the matching state, and the corresponding acquisition time sequence are associated and encapsulated to generate the original contact data.

3. The system according to claim 1, characterized in that, The vital signs acquisition module is used to establish the association between different monitored objects and corresponding vital signs acquisition channels according to the maternal and infant monitoring session; Based on the vital signs acquisition channel, the vital signs data of the corresponding monitored object are obtained, and the vital signs data are marked with channel source and identified as valid data segments. Based on the time reference of the acquisition device and the data reception sequence, the acquisition time information is configured for the vital signs data, and the vital signs data, the channel source marker, and the acquisition time information are associated and output.

4. The system according to claim 1, characterized in that, The contact state modeling module is used to extract the stable range and abrupt range of contact intensity change over time from the original contact data, and determine candidate contact segments in combination with the matching state of the current contact object. Based on the baseline of contact intensity, the duration of the segment, and the trajectory of changes in the contact range, a state transition constraint analysis is performed on the candidate contact segments to generate the mother-infant contact state segments with segment boundaries and state identifiers. Based on the duration, switching relationship, and stability of the contact range of the mother-infant contact state segment, the contact duration statistics and contact quality parameters are generated.

5. The system according to claim 4, characterized in that, The process involves performing state transition constraint analysis on the candidate contact segments based on the contact intensity baseline, segment duration, and contact range change trajectory, generating the mother-infant contact state segments with segment boundaries and state identifiers, including: The contact determination threshold corresponding to the candidate contact segment is determined based on the contact strength baseline, and the candidate contact segment is subjected to short-term perturbation filtering based on the segment duration; The contact range stability state of the candidate contact segment is determined based on the contact range change trajectory, and the contact determination threshold, the short-term disturbance filtering result, and the contact range stability state are input into a preset state transition rule; Based on the preset state transition rules, the state succession relationship between adjacent candidate contact segments is determined, and the mother-infant contact state segment with segment boundaries and state identifiers is generated.

6. The system according to claim 1, characterized in that, The time calibration module is used to extract state switching nodes from the mother-infant contact state segments and establish a contact-side calibration time reference based on the state switching nodes. Based on the contact-side calibration time reference, the original contact data and the contact duration statistics are corrected for segment boundaries, and the corresponding segment affiliation relationship is configured for the contact quality parameters; Based on the acquisition time information and channel delay characteristics of the vital signs data, the vital signs data are mapped to the corresponding mother-infant contact state segments to generate the contact context monitoring data, which includes contact state, contact duration, contact quality, and changes in vital signs.

7. The system according to claim 1, characterized in that, The association discrimination module is used to determine the contact state change nodes from the contact context monitoring data, and configure the corresponding physiological response time range based on the contact state change nodes; Within the physiological response time range, contact quality change characteristics and vital sign fluctuation characteristics are extracted and compared with the stable monitoring baseline before the contact state change node; Based on the difference matching results, the temporal correlation strength between changes in contact and fluctuations in vital signs is determined, and the contact-related monitoring events with event attribute information and correlation discrimination criteria are generated.

8. The system according to claim 7, characterized in that, The step of determining the temporal correlation strength between contact changes and vital sign fluctuations based on the difference matching results, and generating the contact-related monitoring event with event attribute information and correlation discrimination criteria, includes: Based on the difference matching results, the temporal consistency relationship between contact change characteristics, contact quality change characteristics, and vital sign fluctuation characteristics is determined. Based on the order of changes, duration, and data reliability within the physiological response time range, the temporal consistency relationship is weighted and calculated to obtain the temporal correlation strength. Based on the temporal correlation strength and preset event determination conditions, the contact-related monitoring event with event attribute information and correlation discrimination criteria is generated.

9. The system according to claim 1, characterized in that, The monitoring result output module is used to merge and map the contact-related monitoring events to generate corresponding monitoring risk identifiers. Based on the aforementioned monitoring risk identifiers, the corresponding mother-infant contact status segments, contact duration statistics, contact quality parameters, changes in vital signs, and related judgment criteria are linked and organized to generate monitoring evidence data. The maternal and infant multi-parameter monitoring results are generated based on the monitoring evidence data, and the maternal and infant multi-parameter monitoring results are output to the corresponding monitoring terminal.

10. A method for multi-parameter monitoring of mother and infant based on skin contact monitoring according to any one of claims 1 to 9, characterized in that, include: Collect pressure sensing data and identity sensing data during the skin-to-skin contact process between mother and baby, and generate raw contact data based on continuous changes and matching status; Collect vital sign data of the mother and newborn, and configure the collection time information; The original contact data is used to model the contact state, generating mother-infant contact state segments, contact duration statistics, and contact quality parameters; Using the state switching node in the mother-infant contact state segment as a calibration benchmark, the vital sign data is mapped to the corresponding mother-infant contact state segment to generate contact context monitoring data; Based on the physiological response time range after the change in contact state, the contact context monitoring data is correlated and identified to generate contact-related monitoring events, and maternal and infant multi-parameter monitoring results are generated accordingly.

Images