A centralized intelligent triage and dispatch system for multiple-bed fetal heart monitoring data in a delivery room
Patent Information
- Application Number
- CN202611082819.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-21
- Publication Date
- 2026-10-30
AI Technical Summary
[0007]因此，本发明的目的是提供一种产房多床位胎心监护数据集中智能分诊调度系统，通过对现有的监护系统进行改进，能够解决上述提出现有技术中现有的监护系统仅依据胎心风险高低安排医护处置顺序，不考虑当班助产、医师的连续工作时长、持续处置床位带来的生理疲劳，长期将高危产妇全部指派给同一组医护，极易因人员疲劳造成处置疏漏，调度过程不会自动兼顾产房急救设备占用、急诊新入院产妇优先级，全靠护士人工登记排班分诊，主观误差大，高危产妇容易排队等待的问题
[0024]1、本发明中，通过采集医护连续在岗时长、处置床位数量等多项生理负荷指标生成均衡系数，和床位风险、急救设备余量、急诊优先级一同参与推演计算，自动均衡分配高危处置任务，不会出现同一组医护持续接收高危产妇、疲劳作业的情况，降低人为处置疏漏风险，同时省去护士人工登记排班的工作量，分诊排序客观统一，不会因人为判断产生偏差，高危产妇能够得到优先、合理的医护资源分配。
Smart Images
Figure CN123017390A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fetal heart rate monitoring technology, specifically to a centralized intelligent triage and scheduling system for fetal heart rate monitoring data in multiple beds in the delivery room. Background Technology
[0002] The delivery room is a core medical area in the hospital's obstetrics department that is specially designated and meets aseptic control standards. It is divided into functional areas such as labor waiting room, independent delivery room, isolation delivery room, and neonatal resuscitation area. It is equipped with professional equipment such as multi-bed fetal heart monitor, multi-functional delivery bed, neonatal resuscitation table, and emergency equipment. A dedicated team of obstetricians, midwives, anesthesiologists, and neonatal medical staff is on duty to monitor the entire labor process after the onset of labor, assist with natural childbirth, implement painless childbirth, and handle emergency treatment for dystocia and high-risk pregnant women.
[0003] Fetal heart rate monitoring in the delivery room is a standardized monitoring method used in obstetrics to monitor changes in fetal heart rate and maternal uterine contractions in real time during labor and delivery. It continuously captures the fetal heart rate curve and records the uterine contraction waveform through a dedicated fetal heart rate probe. Medical staff rely on the changes in the waveform to judge whether there are dangerous conditions such as fetal hypoxia or distress in utero, and promptly identify and intervene in the risks of labor. It covers the entire process from labor to delivery and is a core basic monitoring method to ensure maternal and infant safety and avoid critical complications during childbirth.
[0004] However, existing monitoring systems only arrange the order of medical care based on the level of fetal heart rate risk, without taking into account the continuous working hours of midwives and physicians on duty, or the physiological fatigue caused by continuous bed care. In the long run, all high-risk pregnant women are assigned to the same medical team, which can easily lead to oversights due to staff fatigue. The scheduling process does not automatically take into account the occupancy of emergency equipment in the delivery room or the priority of newly admitted pregnant women in the emergency room. It all relies on nurses to manually register, schedule and triage patients, which has a large subjective error and high-risk pregnant women are prone to waiting in line.
[0005] To address this, we propose a centralized intelligent triage and scheduling system for fetal heart rate monitoring data in multi-bed delivery rooms. Summary of the Invention
[0006] In view of the problems existing in the above and / or the existing multi-bed fetal heart rate monitoring data centralized intelligent triage and scheduling system in the delivery room, the present invention is proposed.
[0007] Therefore, the purpose of this invention is to provide a centralized intelligent triage and scheduling system for fetal heart rate monitoring data in multi-bed delivery rooms. By improving the existing monitoring system, it can solve the problems mentioned above. The existing monitoring system only arranges the order of medical care based on the level of fetal heart rate risk, without considering the continuous working hours of midwives and physicians on duty, and the physiological fatigue caused by continuous bed management. In the long run, all high-risk pregnant women are assigned to the same group of medical staff, which is very easy to cause oversights due to staff fatigue. The scheduling process does not automatically take into account the occupancy of emergency equipment in the delivery room and the priority of newly admitted pregnant women in the emergency room. It relies entirely on nurses to manually register and schedule triage, which has large subjective errors and makes it easy for high-risk pregnant women to wait in line.
[0008] To address the aforementioned technical problems, according to one aspect of the present invention, the present invention provides the following technical solution:
[0009] A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room includes a cluster of multi-bed fetal heart rate acquisition terminals, an edge centralized processing gateway, a cloud-based intelligent triage and scheduling platform, a ward terminal display unit, an audio-visual graded early warning unit, and an obstetric mobile interactive terminal.
[0010] The multi-bed fetal heart rate acquisition terminal cluster is deployed in each labor bed to collect and encrypt raw monitoring data of fetal heart rate, uterine contractions, and maternal vital signs in real time and upload them.
[0011] The edge centralized processing gateway communicates with all acquisition terminals and has a built-in time-series alignment correction module, multi-source data fusion unit, fetal distress preliminary judgment model, local network outage cache library, second encrypted transmission module, and fetal heart signal distortion tracing and diagnosis module. The time-series alignment correction module unifies the time axis of multi-terminal data streams. The multi-source data fusion unit generates a standardized monitoring time-series dataset. The fetal distress preliminary judgment model outputs a bed-based preliminary risk label. The fetal heart signal distortion tracing module introduces an industrial vibration frequency domain hierarchical decomposition algorithm to separate four types of waveform distortion components and output the waveform's effective confidence level. The edge centralized processing gateway caches the standardized dataset, risk labels, and distortion confidence levels and synchronously uploads them to the cloud.
[0012] The cloud-based intelligent triage and scheduling platform stores maternal medical records, monitoring time-series data, and obstetric resource information. It includes a multi-dimensional risk grading review module, an obstetric resource scheduling simulation model, a dynamic triage allocation module, a scheduling instruction issuance module, a dynamic diversion scheduling module for medical staff physiological load, and a monitoring waveform feature watermarking traceability and anti-tampering module. The multi-dimensional risk grading review module combines medical records and waveform confidence levels to conduct a secondary review of bed risk. The medical staff physiological load diversion module introduces a rail transit balanced diversion algorithm, integrates medical staff fatigue indicators to adjust scheduling weights, and balances manpower allocation. The waveform feature watermarking module extracts the inherent physiological characteristics of fetal heart rate as a steganographic watermark to achieve traceability of local waveform tampering points. The obstetric resource scheduling simulation model combines bed risk and medical staff load to simulate the optimal triage plan, generate scheduling instructions, and issue them.
[0013] The ward terminal display unit visually displays bed monitoring, risk, and medical staff workload information; the audio-visual graded early warning unit outputs differentiated audio-visual prompts according to risk; and the obstetrics mobile interactive terminal receives scheduling and early warning information and transmits on-site handling records back.
[0014] As a preferred embodiment of the multi-bed fetal heart rate monitoring data centralized intelligent triage and scheduling system for delivery rooms described in this invention, each terminal in the multi-bed fetal heart rate acquisition terminal cluster is equipped with a fetal heart rate signal acquisition module, a uterine contraction pressure acquisition module, a maternal vital signs acquisition module, a local short-term buffer unit, and a first encrypted transmission module. The edge centralized processing gateway is also equipped with a bed load statistics module to count the online status of each acquisition terminal and the number of beds waiting for monitoring in real time.
[0015] As a preferred embodiment of the multi-bed fetal heart rate monitoring data centralized intelligent triage and scheduling system in the delivery room described in this invention, the local network outage cache adopts a cyclic overwrite storage mechanism, continuously caching monitoring data during network outages and automatically retransmitting in batches after communication is restored. The four types of distortion components separated by the fetal heart rate signal distortion tracing and diagnosis module are: maternal motion interference, poor electrode contact, equipment circuit noise, and fetal physiological heart rate fluctuations.
[0016] As a preferred embodiment of the multi-bed fetal heart rate monitoring data centralized intelligent triage and scheduling system for delivery rooms described in this invention, the physiological load indicators collected by the dynamic triage and scheduling module for medical staff include continuous working hours, real-time number of beds treated, on-duty mileage, and fatigue heart rate. The audible and visual graded early warning unit is equipped with three levels of differentiated audible and visual output logic: low-risk constant illumination prompt, medium-risk intermittent audible and visual prompt, and high-risk high-frequency audible and visual pop-up dual reminder.
[0017] As a preferred embodiment of the centralized intelligent triage and scheduling system for fetal heart rate monitoring in multi-bed delivery rooms described in this invention, the cloud-based intelligent triage and scheduling platform is further configured with an automatic monitoring report generation module, and the generated electronic medical records can be connected to the hospital's HIS system.
[0018] As a preferred embodiment of the multi-bed fetal heart rate monitoring data centralized intelligent triage and scheduling system in the delivery room described in this invention, the timing alignment correction module uses a unified standard timestamp in the cloud as a reference to perform interpolation correction processing on the asynchronous monitoring data streams transmitted by each acquisition terminal, thereby eliminating the waveform misalignment problem caused by asynchronous sampling of different hardware devices.
[0019] As a preferred embodiment of the centralized intelligent triage and scheduling system for multi-bed fetal heart rate monitoring in the delivery room described in this invention, the multi-dimensional risk grading and verification module retrieves the past pregnancy medical records and historical similar fetal heart rate monitoring cases of the mothers in the database, and performs a secondary correction on the initial risk judgment output by the edge gateway in combination with the effective confidence of the waveform, so as to distinguish between physiological fetal heart rate fluctuations and pathological fetal distress waveforms.
[0020] As a preferred embodiment of the centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room as described in this invention, the obstetric resource scheduling and deduction model simultaneously incorporates two types of scheduling influencing factors: the occupancy status of emergency equipment in the delivery room and the priority of newly admitted emergency mothers. It also comprehensively outputs the triage and treatment order by combining bed risk and the balance coefficient of medical and nursing physiological load.
[0021] As a preferred embodiment of the multi-bed fetal heart rate monitoring data centralized intelligent triage and scheduling system of the present invention, the monitoring waveform feature watermark traceability and anti-tampering module extracts the unique physiological features of a single segment of the fetal heart rate curve as a hidden watermark. The watermark is synchronously bound to the bed number, the acquisition device, and the operator's identifier, which can locate the position of the local modification of the waveform and the operating subject.
[0022] As a preferred embodiment of the multi-bed fetal heart rate monitoring centralized intelligent triage and scheduling system for the delivery room described in this invention, the obstetric mobile interactive terminal has a local offline caching function. When the network is disconnected, it temporarily stores scheduling reminders and monitoring thumbnail data. After the network is restored, it automatically uploads the on-site treatment records to the cloud platform to complete the status update.
[0023] Compared with the prior art, the beneficial effects achieved by the present invention are:
[0024] 1. In this invention, a balance coefficient is generated by collecting multiple physiological load indicators such as the continuous on-duty time of medical staff and the number of beds treated. This coefficient, along with bed risk, emergency equipment availability, and emergency priority, is used in the simulation calculation to automatically and evenly allocate high-risk treatment tasks. This prevents the same group of medical staff from continuously receiving high-risk pregnant women or working under fatigue, reducing the risk of human error. At the same time, it saves nurses the workload of manual registration and scheduling. The triage and sorting are objective and consistent, and there will be no deviation due to human judgment. High-risk pregnant women can receive priority and reasonable allocation of medical and nursing resources.
[0025] 2. In this invention, by embedding a fetal heart signal distortion tracing and diagnosis module in the edge gateway, an industrial vibration frequency domain hierarchical decomposition algorithm is introduced to split four types of waveform interference signals, and the effective confidence level of the waveform is output synchronously. Combined with a cloud-based multi-dimensional risk review module, the risk of fetal distress is determined a second time. This can automatically distinguish between false abnormalities such as maternal turning over and electrode loosening and the actual hypoxia waveform of the fetus, reducing invalid false alarms from the equipment. It eliminates the need for medical staff to manually verify the authenticity of the waveforms one by one. At the same time, it unifies the time axis of all monitoring terminals and generates a standardized time-series dataset. Waveforms from multiple beds can be compared horizontally and synchronously, improving the accuracy of risk identification and reducing the risk of missed diagnosis and misdiagnosis.
[0026] 3. In this invention, a matching monitoring waveform feature watermark traceability and anti-tampering module is used, combined with an edge network outage caching, multi-terminal hierarchical early warning, and automatic medical record generation system. Based on the fetal heart rate's own physiological curve, a hidden watermark is generated and bound to the device and operator. This can accurately locate the local tampering points of the waveform, meet the needs of medical dispute evidence and in-hospital medical record quality control, automatically retain all monitoring data when the network is out of service, and automatically retransmit the data when the network is restored without losing monitoring records. The system provides three-level synchronous push of hierarchical early warnings through the nurse station screen, bedside audio and light, and medical staff handheld terminals. It can also automatically generate monitoring reports that can be connected to the in-hospital HIS. The entire system integrates data retention, early warning push, and medical record archiving, and is adapted to the standardized quality control management requirements of the delivery room. Attached Figure Description
[0027] Figure 1 This is a schematic diagram of the system architecture of the present invention;
[0028] Figure 2 This is a schematic diagram of the system workflow of the present invention;
[0029] Figure 3 This is a schematic diagram of the cloud-based intelligent triage and scheduling platform of the present invention;
[0030] Figure 4 This is a schematic diagram illustrating the anti-tampering properties of the monitoring waveform feature watermark for tracing the source of the present invention;
[0031] Figure 5 This is a schematic diagram of the sound and light graded early warning system of the present invention;
[0032] Figure 6This is a schematic diagram of the multi-bed fetal heart rate acquisition method of the present invention;
[0033] Figure 7 This is a schematic diagram illustrating the fetal heart rate signal distortion tracing and diagnosis method of the present invention. Detailed Implementation
[0034] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
[0035] This invention provides a centralized intelligent triage and scheduling system for fetal heart rate monitoring data in multi-bed delivery rooms. It generates a balance coefficient by collecting multiple physiological load indicators such as the continuous on-duty time of medical staff and the number of beds to be treated. This coefficient, along with bed risk, emergency equipment availability, and emergency priority, is used in the extrapolation and calculation to automatically and evenly allocate high-risk treatment tasks. This prevents the same group of medical staff from continuously receiving high-risk pregnant women or working under fatigue, reducing the risk of human error. At the same time, it saves nurses the workload of manual registration and scheduling. The triage and sorting are objective and consistent, without deviations due to human judgment. High-risk pregnant women can receive priority and reasonable allocation of medical and nursing resources.
[0036] Please see Figure 1-7 A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room includes a multi-bed fetal heart rate acquisition terminal cluster, an edge centralized processing gateway, a cloud-based intelligent triage and scheduling platform, a ward terminal display unit, an audio-visual graded early warning unit, and an obstetric mobile interactive terminal. The multi-bed fetal heart rate acquisition terminal cluster is deployed in each delivery bed to collect and encrypt raw monitoring data of fetal heart rate, uterine contractions, and maternal vital signs in real time. The edge centralized processing gateway communicates with all acquisition terminals and has a built-in time sequence alignment correction module, a multi-source data fusion unit, a fetal distress preliminary judgment model, a local network outage cache library, a second encrypted transmission module, and a fetal heart rate signal distortion tracing and diagnosis module. The time sequence alignment correction module unifies the time axis of the multi-terminal data stream, the multi-source data fusion unit generates a standardized monitoring time sequence dataset, the fetal distress preliminary judgment model outputs a bed preliminary judgment risk label, and the fetal heart rate signal distortion tracing and diagnosis module introduces an industrial vibration frequency domain hierarchical decomposition algorithm to separate four types of waveform distortion components and output the waveform effective confidence level. The edge centralized processing gateway caches the standardized dataset, risk labels, and distortion confidence levels and synchronously uploads them to the cloud.
[0037] The cloud-based intelligent triage and scheduling platform stores maternal medical records, monitoring time-series data, and obstetric resource information. It includes a multi-dimensional risk grading review module, an obstetric resource scheduling simulation model, a dynamic triage allocation module, a scheduling instruction issuance module, a dynamic diversion scheduling module for medical staff physiological load, and a monitoring waveform feature watermarking traceability and anti-tampering module. The multi-dimensional risk grading review module combines medical records and waveform confidence to conduct a secondary review of bed risk. The medical staff physiological load diversion module introduces a rail transit balanced diversion algorithm, integrates medical staff fatigue indicators to adjust scheduling weights, and balances manpower allocation. The waveform feature watermarking module extracts the inherent physiological characteristics of fetal heart rate as a steganographic watermark to achieve traceability of local waveform tampering points. The obstetric resource scheduling simulation model combines bed risk and medical staff load to simulate the optimal triage plan, generate scheduling instructions, and issue them.
[0038] The ward terminal display unit visually displays bed monitoring, risk, and medical staff workload information. The audio-visual tiered early warning unit outputs differentiated audio-visual prompts according to risk. The obstetrics mobile interactive terminal receives dispatch and early warning information and transmits on-site treatment records. By collecting multiple physiological load indicators such as the continuous on-duty time of medical staff and the number of beds treated, a balance coefficient is generated. This coefficient, along with bed risk, emergency equipment availability, and emergency priority, participates in the extrapolation calculation to automatically and evenly allocate high-risk treatment tasks. This prevents the same group of medical staff from continuously receiving high-risk pregnant women or working under fatigue, reducing the risk of human error. At the same time, it saves nurses the workload of manual registration and scheduling. The triage and sorting are objective and consistent, without deviations due to human judgment. High-risk pregnant women can receive priority and reasonable allocation of medical and nursing resources.
[0039] Each terminal in the multi-bed fetal heart rate monitoring terminal cluster has a built-in fetal heart rate signal acquisition module, uterine contraction pressure acquisition module, maternal vital signs acquisition module, local short-term cache unit, and first encrypted transmission module. The edge centralized processing gateway also has a built-in bed load statistics module to count the online status of each acquisition terminal and the number of monitoring waiting beds in real time. The local network outage cache adopts a cyclic overwrite storage mechanism to continuously cache monitoring data during network outages and automatically retransmit it in batches after communication is restored. The fetal heart rate signal distortion tracing and diagnosis module separates four types of distortion components: maternal motion interference, poor electrode contact, equipment circuit noise, and fetal physiological heart rate fluctuations.
[0040] The dynamic triage and scheduling module for medical staff's physiological workload collects indicators such as continuous working hours, real-time number of beds under treatment, on-duty mileage, and fatigue heart rate. The audible and visual graded early warning unit is equipped with three levels of differentiated audible and visual output logic: constant light for low risk, intermittent audible and visual alerts for medium risk, and dual alerts with high-frequency audible and visual pop-ups for high risk. The cloud-based intelligent triage and scheduling platform is also equipped with an automatic monitoring report generation module, and the generated electronic medical records can be connected to the hospital's HIS system.
[0041] The timing alignment correction module uses a unified cloud-based standard timestamp as a benchmark to perform interpolation correction on the asynchronous monitoring data streams transmitted by each acquisition terminal, eliminating waveform misalignment caused by asynchronous sampling of different hardware devices; the multi-dimensional risk grading review module retrieves the mother's past pregnancy medical records and historical similar fetal heart rate monitoring cases from the database, and performs secondary correction on the initial risk assessment output by the edge gateway based on the effective confidence level of the waveform, distinguishing between physiological fetal heart rate fluctuations and pathological fetal distress waveforms.
[0042] The obstetric resource scheduling simulation model incorporates two scheduling influencing factors: the occupancy status of emergency equipment in the delivery room and the priority of newly admitted emergency mothers. It combines bed risk and the physiological load balance coefficient of medical staff to comprehensively output the triage and treatment order. The monitoring waveform feature watermark traceability and anti-tampering module extracts the unique physiological features of a single segment of the fetal heart rate curve as a hidden watermark. The watermark is simultaneously bound to the bed number, the acquisition device, and the operator's identifier, which can locate the location of local waveform modification and the operator. The obstetric mobile interactive terminal has a local offline caching function. When the network is disconnected, it temporarily stores scheduling reminders and monitoring thumbnail data. After the network is restored, it automatically uploads the on-site treatment records to the cloud platform to complete the status update.
[0043] The workflow of this invention is as follows: After the mother enters the labor room, the staff deploys a multi-bed fetal heart rate monitoring terminal at the corresponding bed. The fetal heart rate, uterine contractions and vital signs monitoring module inside the terminal continuously collects raw monitoring data. After the data collection is completed, it is transmitted to the outside in real time through the first encrypted transmission module. The data from all terminals are uniformly aggregated to the edge centralized processing gateway inside the labor room, and the data processing flow of the entire system is officially started.
[0044] After receiving the raw data streams asynchronously transmitted from each bed, the edge gateway first activates the timing alignment and correction module. Using the unified standard timestamp in the cloud as a benchmark, it interpolates and corrects the data with sampling time differences between different devices, calibrating the monitoring waveforms of all beds to the same time axis, eliminating the problems of waveform misalignment and inability to compare laterally. After the correction is completed, the data is handed over to the multi-source data fusion unit to integrate fetal heart rate, uterine contractions, and maternal vital signs information to generate a standardized monitoring time series dataset in a unified format.
[0045] Subsequently, the gateway simultaneously runs two sets of analysis units: one is the fetal distress preliminary assessment model, which directly provides the first, second, and third-level preliminary risk labels for each bed based on the standardized dataset; the other is the fetal heart signal distortion tracing and diagnosis module, which uses an industrial-grade frequency domain hierarchical decomposition algorithm to split the waveform signal, distinguishing four types of interference components: maternal activity, poor electrode contact, circuit noise, and physiological heart rate fluctuations, and calculating and labeling the effective confidence level of the waveform. At the same time, the bed load statistics module in the gateway counts the online status of each terminal and the total number of beds to be monitored in real time. The local network outage cache library is simultaneously opened for cyclic storage. Regardless of network connectivity, all standardized datasets, risk labels, and waveform confidence levels are retained locally. Once the network cable is disconnected, the monitoring information will not be lost. After communication is restored, all cached content will be automatically re-uploaded to the cloud in batches. After the gateway completes all edge preprocessing work, it uploads the integrated set of encrypted data to the cloud-based intelligent triage and scheduling platform.
[0046] After receiving the data uploaded from the edge, the cloud platform first retrieves the past pregnancy history of the mothers and similar historical fetal heart rate monitoring cases stored in the database through the multi-dimensional risk grading review module. It then recalibrates the initial risk assessment given by the edge based on the effective confidence level of the waveform, distinguishes between simple physiological fluctuations and pathological fetal distress, and outputs the final unified bed risk level. After calibration, the data is sent to the obstetric resource scheduling model. The model reads two types of core parameters at the same time: one is the final risk level of each bed, the priority of mothers admitted to the emergency department, and the occupancy status of emergency equipment in the delivery room; the other is the load indicators such as continuous working time, number of beds treated, on-duty travel mileage, and fatigue heart rate collected in real time by the dynamic triage scheduling module for the physiological load of medical staff. Based on the rail transit balanced triage algorithm, the load balance coefficient of each midwife and physician is calculated. The optimal triage and treatment sequence is deduced by comprehensively considering all parameters, and standardized scheduling instructions are generated.
[0047] After the dispatch instruction is generated, it is sent to the dispatch instruction distribution module and pushed to various terminals. At the same time, it is sent to the monitoring waveform feature watermark traceability and anti-tampering module. The module extracts the unique physiological extreme value and variation cycle characteristics of a single segment of fetal heart curve as a hidden watermark, and embeds the bed number, acquisition equipment and operator identification into the time series data. Any subsequent local waveform modification can be reversed to locate the tampering location and operator, which is used for medical record quality control and dispute evidence.
[0048] The ward terminal display unit at the ward nurse station synchronously updates the monitoring curves, risk levels, and medical staff load balance coefficients for all beds, facilitating a comprehensive view for nurses. Each labor bed is equipped with an audio-visual grading and early warning unit that outputs differentiated prompts based on the risk level issued by the cloud. Low risk is indicated by constant lighting, medium risk by intermittent audio-visual alerts, and high risk by simultaneous high-frequency lighting plus pop-up reminders on handheld terminals. Midwives and physicians wear obstetric mobile interactive terminals that receive triage reminders and abbreviated monitoring information in real time. The devices have built-in offline caching, which can temporarily save all prompts when the network is disconnected. After medical staff arrive at the corresponding bed and complete the examination and treatment, they enter the treatment record on the terminal, and the data is automatically transmitted back to the cloud. The platform updates the bed risk and medical staff load parameters in real time and enters the next round of scheduling simulation cycle.
[0049] After the daily monitoring work is completed, the cloud-based monitoring report generation module integrates the full-day time-series monitoring data, risk assessment records, and medical treatment records to generate a complete electronic fetal heart rate medical record. It supports direct retrieval and archiving by the hospital's HIS system. The entire process of data collection, preprocessing, intelligent triage, graded early warning, and evidence archiving operates in a closed loop. The entire process does not require a large amount of manual registration, manual waveform identification, or manual arrangement of medical care, making it suitable for the high-intensity work scenario of simultaneous monitoring of multiple mothers in the delivery room.
[0050] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims
1. A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in multi-bed delivery rooms, characterized in that, It includes a multi-bed fetal heart rate monitoring terminal cluster, an edge centralized processing gateway, a cloud-based intelligent triage and scheduling platform, a ward terminal display unit, an audio-visual graded early warning unit, and an obstetric mobile interactive terminal; The multi-bed fetal heart rate acquisition terminal cluster is deployed in each labor bed to collect and encrypt raw monitoring data of fetal heart rate, uterine contractions, and maternal vital signs in real time and upload them. The edge centralized processing gateway communicates with all acquisition terminals and has a built-in time-series alignment correction module, multi-source data fusion unit, fetal distress preliminary judgment model, local network outage cache library, second encrypted transmission module, and fetal heart signal distortion tracing and diagnosis module. The time-series alignment correction module unifies the time axis of multi-terminal data streams. The multi-source data fusion unit generates a standardized monitoring time-series dataset. The fetal distress preliminary judgment model outputs a bed-based preliminary risk label. The fetal heart signal distortion tracing module introduces an industrial vibration frequency domain hierarchical decomposition algorithm to separate four types of waveform distortion components and output the waveform's effective confidence level. The edge centralized processing gateway caches the standardized dataset, risk labels, and distortion confidence levels and synchronously uploads them to the cloud. The cloud-based intelligent triage and scheduling platform stores maternal medical records, monitoring time-series data, and obstetric resource information. It includes a multi-dimensional risk grading review module, an obstetric resource scheduling simulation model, a dynamic triage allocation module, a scheduling instruction issuance module, a dynamic diversion scheduling module for medical staff physiological load, and a monitoring waveform feature watermarking traceability and anti-tampering module. The multi-dimensional risk grading review module combines medical records and waveform confidence levels to conduct a secondary review of bed risk. The medical staff physiological load diversion module introduces a rail transit balanced diversion algorithm, integrates medical staff fatigue indicators to adjust scheduling weights, and balances manpower allocation. The waveform feature watermarking module extracts the inherent physiological characteristics of fetal heart rate as a steganographic watermark to achieve traceability of local waveform tampering points. The obstetric resource scheduling simulation model combines bed risk and medical staff load to simulate the optimal triage plan, generate scheduling instructions, and issue them. The ward terminal display unit visually displays bed monitoring, risk, and medical staff workload information; the audio-visual graded early warning unit outputs differentiated audio-visual prompts according to risk; and the obstetrics mobile interactive terminal receives scheduling and early warning information and transmits on-site handling records back.
2. The centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 1, characterized in that, Each terminal in the multi-bed fetal heart rate monitoring terminal cluster has a built-in fetal heart rate signal acquisition module, a uterine contraction pressure acquisition module, a maternal vital signs acquisition module, a local short-term buffer unit, and a first encrypted transmission module. The edge centralized processing gateway also has a built-in bed load statistics module to count the online status of each acquisition terminal and the number of monitoring waiting beds in real time.
3. The centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 2, characterized in that, The local network outage cache library adopts a cyclic overwrite storage mechanism, continuously caching monitoring data during network outages and automatically retransmitting it in batches after communication is restored. The four types of distortion components separated by the fetal heart signal distortion tracing and diagnosis module are: maternal motion interference, poor electrode contact, equipment circuit noise, and fetal physiological heart rate fluctuations.
4. The centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 3, characterized in that, The physiological load indicators collected by the dynamic diversion and scheduling module for medical staff include continuous working hours, real-time number of beds treated, on-duty mileage, and fatigue heart rate. The audio-visual graded early warning unit is set with three levels of differentiated audio-visual output logic: low-risk constant light prompt, medium-risk intermittent audio-visual prompt, and high-risk high-frequency audio-visual pop-up dual reminder.
5. The centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 4, characterized in that, The cloud-based intelligent triage and scheduling platform is also equipped with an automatic monitoring report generation module, and the generated electronic medical records can be connected to the hospital's HIS system.
6. The centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 5, characterized in that, The timing alignment correction module uses a unified standard timestamp in the cloud as a reference to perform interpolation correction on the asynchronous monitoring data streams transmitted by each acquisition terminal, thereby eliminating waveform misalignment caused by asynchronous sampling of different hardware devices.
7. A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 6, characterized in that, The multi-dimensional risk grading review module retrieves the mother's past pregnancy medical records and historical similar fetal heart rate monitoring cases from the database, and performs a secondary correction on the initial risk assessment output by the edge gateway based on the effective confidence level of the waveform, distinguishing between physiological fetal heart rate fluctuations and pathological fetal distress waveforms.
8. A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 7, characterized in that, The obstetric resource scheduling simulation model incorporates two scheduling influencing factors: the occupancy status of emergency equipment in the delivery room and the priority of newly admitted emergency mothers. It also combines bed risk and the physiological load balance coefficient of medical staff to comprehensively output the triage and treatment order.
9. A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 8, characterized in that, The monitoring waveform feature watermark traceability and anti-tampering module extracts the unique physiological features of a single segment of the fetal heart rate curve as a hidden watermark. The watermark is simultaneously bound to the bed number, acquisition device, and operator identifier, which can locate the position of local waveform modification and the operator.
10. A centralized intelligent triage and scheduling system for fetal heart rate monitoring data in a multi-bed delivery room according to claim 9, characterized in that, The obstetric mobile interactive terminal has a local offline caching function. When the network is disconnected, it temporarily stores scheduling reminders and monitoring abbreviated data. After the network is restored, it automatically uploads the on-site treatment records to the cloud platform to complete the status update.