Patient transfer intelligent monitoring platform based on Internet of Things and intelligent wearing technology

Abstract

The invention discloses a patient transfer intelligent monitoring platform based on the Internet of Things and the intelligent wearing technology, and relates to the technical field of transfer intelligent monitoring, and the platform comprises an intelligent wearing equipment module which is used for carrying out the time sequence collection of physical sign data in a transfer process, and obtaining a physical sign monitoring data sequence; the medical equipment interconnection module is used for carrying out simultaneous multi-terminal data acquisition; the data transmission network module is used for determining an infusion pump infusion speed analysis result, a physical sign abnormal mode analysis result, a monitoring terminal abnormal mode analysis result and an oxygen consumption reserve analysis result; and the alarm and display module is used for graded alarm and shared display of the electronic handover bill. The technical problems that in the prior art, continuous dynamic monitoring cannot be achieved in the transferring process of the invasive mechanical ventilation patient, the early warning capability is lacked, and handover communication is not smooth are solved, and the technical effects that physical sign data and medical equipment parameters in the transferring process of the patient are collected in real time, and transferring monitoring continuity and handover efficiency are improved are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a patient transfer intelligent monitoring platform based on Internet of Things and intelligent wearing technology. BACKGROUND

[0002] At present, there is a high risk of in-hospital transfer of patients with invasive mechanical ventilation. The carrying and body position change may cause hemodynamic fluctuation or oxygenation deterioration. The device-related risk is high, such as disconnection of ventilator pipeline, insufficient oxygen supply, battery depletion, etc. At the same time, the existing transfer mode has limitations. The vital signs are observed and recorded by medical staff during transfer. Although continuous monitoring of vital signs can be achieved through a transfer monitor, there is a lack of effective linkage mechanism between various medical monitoring systems, and intelligent early warning cannot be triggered based on real-time physiological parameter changes of the patient. The handover between the transfer team and the receiving department relies on oral report or written record, and the visualization and traceability of the whole transfer process cannot be realized.

[0003] The prior art has the technical problems that the patient with invasive mechanical ventilation cannot be continuously and dynamically monitored during transfer, lacks early warning capability, and the handover communication is not smooth. SUMMARY

[0004] The present application provides a patient transfer intelligent monitoring platform based on Internet of Things and intelligent wearing technology, which is used to solve the technical problems that the patient with invasive mechanical ventilation cannot be continuously and dynamically monitored during transfer, lacks early warning capability, and the handover communication is not smooth in the prior art.

[0005] In view of the above problems, the present application provides a patient transfer intelligent monitoring platform based on Internet of Things and intelligent wearing technology.

[0006] The present application provides a patient transfer intelligent monitoring platform based on Internet of Things and intelligent wearing technology, which comprises:

[0007] The system includes a smart wearable device module for time-series data acquisition of vital signs during patient transport, worn on the patient's body surface, to obtain a vital sign monitoring data sequence; a medical device interconnection module for communication with the transport ventilator, smart infusion pump, and monitoring terminal, and for simultaneous multi-terminal data acquisition of the transport ventilator, smart infusion pump, and monitoring terminal using the same monitoring frequency as the smart wearable device module, to obtain ventilator parameter sequences, infusion pump infusion rate sequences, and monitoring terminal multi-dimensional parameter sequences; a data transmission network module for transmitting the vital sign monitoring data sequence, ventilator parameter sequence, infusion pump infusion rate sequence, and monitoring terminal multi-dimensional parameter sequence to the monitoring and analysis module for analysis, determining the analysis results of infusion pump infusion rate, abnormal vital sign patterns, abnormal monitoring terminal patterns, and oxygen consumption reserves; and an alarm and display module for tiered alarms based on the analysis results of infusion pump infusion rate, abnormal vital sign patterns, abnormal monitoring terminal patterns, and oxygen consumption reserves, triggering alarm indicator lights for audible and visual alarms, tracking the transport route via an electronic map, and displaying the electronic handover form in real time on the display terminal.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The system includes a smart wearable device module for real-time acquisition of vital sign data during patient transport, obtaining a sequence of vital sign monitoring data. A medical device interconnection module is used for simultaneous multi-terminal data acquisition, obtaining ventilator parameter sequences, infusion pump infusion rate sequences, and multi-dimensional parameter sequences from the monitoring terminal. A data transmission network module transmits these vital sign monitoring data sequences, ventilator parameter sequences, infusion pump infusion rate sequences, and monitoring terminal multi-dimensional parameter sequences to a monitoring and analysis module for analysis, determining the analysis results for infusion pump infusion rate, abnormal vital sign patterns, abnormal monitoring terminal patterns, and oxygen consumption / reservoir capacity. An alarm and display module provides tiered alarms, triggering alarm indicator lights for audible and visual alarms, tracking the transport route via an electronic map, and displaying a shared electronic handover form in real-time on the display terminal. This achieves real-time acquisition, transmission, and analysis of vital sign data and medical device parameters during patient transport, improving the continuity of transport monitoring and the efficiency of handover. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1This application provides a schematic diagram of the structure of a smart patient transport monitoring platform based on Internet of Things and smart wearable technology for embodiments of the present application;

[0012] Figure 2 This application provides a schematic diagram of the alarm and display module structure in a smart patient transport monitoring platform based on Internet of Things and smart wearable technology, which is an embodiment of the present application.

[0013] Explanation of reference numerals in the attached diagram: Smart wearable device module 10, medical device interconnection module 20, data transmission network module 30, alarm and display module 40. Detailed Implementation

[0014] This application provides a smart monitoring platform for patient transport based on the Internet of Things and smart wearable technology, which is used to address the technical problems of the inability to continuously and dynamically monitor invasive mechanical ventilation patients during transport, the lack of early warning capabilities, and the poor communication during handover in existing technologies.

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0016] Examples, such as Figure 1 As shown, this application provides a smart monitoring platform for patient transport based on the Internet of Things and smart wearable technology, the platform comprising:

[0017] The intelligent wearable device module 10 is used to collect vital sign data in a time sequence during the patient's transport process by wearing it on the patient's body surface, thereby obtaining a vital sign monitoring data sequence.

[0018] Specifically, the smart wearable device module 10 is worn on the body surface of patients undergoing invasive mechanical ventilation to collect vital sign data in a time sequence during transport, thereby obtaining a vital sign monitoring data sequence. This module includes a flexible chest patch and a wristband sensor. The flexible chest patch is used to collect ECG signals and body temperature, while the wristband sensor is used to collect blood oxygen saturation and blood pressure. Each collected vital sign monitoring data point includes at least ECG, SpO2, blood pressure, and body temperature. The vital sign monitoring data sequence is processed through multi-scale convolution to determine a multi-scale vital sign monitoring feature set. These features are then randomly and non-overlappingly combined pairwise to obtain a multi-scale vital sign monitoring feature combination set. After iterative analysis and mean processing, target vital sign monitoring features are extracted and approximately matched with multiple sample vital sign abnormality monitoring feature templates in a vital sign abnormality pattern recognizer. The sample vital sign abnormality monitoring feature template corresponding to the maximum matching similarity is used as the matching template. Furthermore, the aggregated sample vital sign abnormality type corresponding to the matching template is used as the vital sign abnormality pattern parsing result, providing a data foundation for subsequent abnormality warning and decision support.

[0019] The medical device interconnection module 20 is used to communicate with the transport ventilator, the smart infusion pump and the monitoring terminal, and to simultaneously collect multi-terminal data from the transport ventilator, the smart infusion pump and the monitoring terminal using the same monitoring frequency as the smart wearable device module, so as to obtain the ventilator parameter sequence, the infusion pump infusion rate sequence and the monitoring terminal multi-dimensional parameter sequence.

[0020] Specifically, the medical device interconnection module 20 establishes communication connections with the transport ventilator, intelligent infusion pump, and monitoring terminal. It simultaneously collects multi-device data from these devices at the same monitoring frequency as the intelligent wearable device module, thereby obtaining ventilator parameter sequences, infusion pump infusion rate sequences, and monitoring terminal multi-dimensional parameter sequences. This module extracts the infusion rate threshold judgment unit (containing preset infusion rate thresholds) from the monitoring and analysis module, traverses the infusion pump infusion rate sequence for anomaly identification, and utilizes a vital sign anomaly pattern recognizer and a monitoring terminal anomaly pattern recognizer to perform pattern recognition on the vital sign monitoring data sequence and the monitoring terminal multi-dimensional parameter sequence, respectively. Furthermore, it can estimate oxygen consumption based on transport ventilator parameters, transport distance, and time, providing early warnings when oxygen reserves are low. This achieves synchronous acquisition and integration of data from multiple devices, providing comprehensive data support for subsequent graded alarms and decision support.

[0021] The data transmission network module 30 is used to transmit the vital sign monitoring data sequence, ventilator parameter sequence, infusion pump infusion rate sequence and monitoring terminal multidimensional parameter sequence to the monitoring and analysis module for analysis, and to determine the analysis results of infusion pump infusion rate, vital sign abnormality mode, monitoring terminal abnormality mode and oxygen consumption reserve.

[0022] Specifically, the data transmission network module 30 is responsible for transmitting the vital sign monitoring data sequences (including ECG, SpO2, blood pressure, body temperature, etc.) collected by the smart wearable device module, the ventilator parameter sequences (such as tidal volume, airway pressure, FiO2, etc.) obtained by the medical device interconnection module, the infusion pump infusion rate sequences, and the multi-dimensional parameter sequences of the monitoring terminal to the monitoring and analysis module in real time via the Internet of Things communication network. In the monitoring and analysis module, an infusion rate threshold judge traverses the infusion pump infusion rate sequences, compares them with preset thresholds to identify anomalies, and generates infusion pump infusion rate analysis results. Simultaneously, a vital sign anomaly pattern recognizer performs multi-scale convolutional feature extraction and template matching on the vital sign data to determine the vital sign anomaly pattern analysis results. Then, the monitoring terminal anomaly pattern recognizer analyzes the monitoring parameter sequences to identify anomalies such as human-machine asynchrony and tubing leaks. Furthermore, by combining transport ventilator parameters, transport distance, and time, oxygen consumption is estimated and oxygen consumption reserve analysis results are generated. Ultimately, this achieves real-time transmission and in-depth analysis of multi-source data, providing a data foundation for graded alarms and intelligent decision-making.

[0023] The alarm and display module 40 is used to perform graded alarms based on the analysis results of the infusion pump infusion rate, the analysis results of abnormal vital signs, the analysis results of abnormal monitoring terminal, and the analysis results of oxygen consumption and storage. It triggers alarm indicator lights to perform audible and visual alarms, tracks the transfer route through an electronic map, and displays the electronic handover form in real time on the display terminal.

[0024] Specifically, the alarm and display module 40 is used to implement a graded alarm mechanism based on the analysis results of the infusion pump infusion rate, abnormal vital signs, abnormal monitoring terminal, and oxygen consumption. A red audible and visual alarm is triggered when each analysis result meets the first-level alarm rule (such as ventilator dislodgement, severe hypoxemia, etc.), and a yellow audible and visual alarm is triggered when it meets the second-level alarm rule (such as mild parameter abnormalities like increased FiO2 demand). Simultaneously, this module achieves meter-level positioning and displays the transfer route via an electronic map, automatically records key information such as elevator waiting time and transfer handover time for post-event review, and can share structured electronic transfer records containing time stamps, route markers, and event markers in real time on the display terminal. This enables visualized tracking of the transfer process and real-time sharing and display of electronic handover documents, improving the response speed to abnormal events.

[0025] In one possible implementation, the smart wearable device module 10 further includes:

[0026] The smart wearable device module includes a flexible chest patch and a wristband sensor. The flexible chest patch is used to collect electrocardiogram signals and body temperature, and the wristband sensor is used to collect blood oxygen saturation and blood pressure.

[0027] Specifically, the smart wearable device module includes a flexible chest patch and a wristband sensor. The flexible chest patch is attached to the chest surface of the patient undergoing invasive mechanical ventilation to collect electrocardiogram (ECG) signals and body temperature data in real time. The wristband sensor is worn on the wrist of the patient undergoing invasive mechanical ventilation to collect blood oxygen saturation (SpO2) and blood pressure parameters. Through these two types of devices, a sequence of vital sign monitoring data containing key indicators such as ECG, SpO2, blood pressure, and body temperature is acquired in a time-series manner. The data sequence undergoes multi-scale convolution processing, feature combination, iterative analysis, and mean calculation to extract target vital sign features. These features are then approximated and matched with pre-stored sample feature templates in a vital sign abnormality pattern recognizer. This enables continuous monitoring of vital signs and abnormal pattern recognition during the transport of patients undergoing invasive mechanical ventilation, providing basic data support for subsequent graded alarms and intelligent decision-making.

[0028] In one possible implementation, the data transmission network module 30 further includes:

[0029] The threshold determination unit is used to extract the infusion rate threshold determination unit of the monitoring and analysis module, wherein the infusion rate threshold determination unit includes a preset infusion rate threshold.

[0030] The infusion rate analysis result extraction unit is used to use the infusion rate threshold judge to traverse the infusion rate sequence of the infusion pump for anomaly identification based on the preset infusion rate threshold, and obtain the infusion rate analysis result of the infusion pump.

[0031] Specifically, the threshold judgment unit is used to extract the infusion rate threshold judgment unit from the monitoring and analysis module. The judgment unit has a built-in parameter preset infusion rate safety threshold range. By extracting the threshold judgment unit, a benchmark is provided for subsequent identification of infusion rate anomalies, ensuring that the traversal check of the infusion pump infusion rate sequence has reasonableness and safety standards.

[0032] The infusion rate analysis result extraction unit calls the infusion rate threshold judge extracted from the monitoring and analysis module. Based on a preset infusion rate threshold (which is pre-set by factors such as drug type and dosage), it performs a point-by-point traversal analysis of the infusion rate sequence collected in real time during the transport process by the infusion pump. When the actual infusion rate is detected to exceed the preset threshold range, abnormal data points are automatically marked, and combined with the time series, an infusion pump infusion rate analysis result including the abnormality type, the time period of occurrence, and the degree of deviation is generated, providing an accurate basis for assessing the infusion status for subsequent graded alarms.

[0033] In one possible implementation, the data transmission network module 30 further includes:

[0034] An abnormal pattern recognition extraction unit is used to extract the abnormal pattern recognition of vital signs and the abnormal pattern recognition of the monitoring terminal from the monitoring and analysis module.

[0035] The pattern recognition unit is used to perform pattern recognition on the vital sign monitoring data sequence using the vital sign abnormality pattern recognizer to obtain the vital sign abnormality pattern analysis result.

[0036] The abnormal pattern parsing result acquisition unit is used to perform pattern recognition on the multi-dimensional parameter sequence of the monitoring terminal using the abnormal pattern recognizer of the monitoring terminal, and obtain the abnormal pattern parsing result of the monitoring terminal.

[0037] Specifically, the abnormal pattern recognition unit is used to extract abnormal pattern recognizers for vital signs and abnormal pattern recognizers for monitoring terminals from the monitoring and analysis module. The abnormal pattern recognizer for vital signs is obtained by acquiring multiple abnormal monitoring features and types of abnormal vital signs from samples, performing union processing on similar abnormal types to obtain aggregated abnormal types, mapping and aggregating the abnormal monitoring features to obtain a feature set, and then generating a template for abnormal monitoring features through mean processing. This template is then mapped and associated one-to-one with the aggregated abnormal types to construct the abnormal pattern recognizer. The abnormal pattern recognizer for monitoring terminals pre-stores abnormal feature templates for multi-dimensional parameters of the monitoring terminal (such as tidal volume and airway pressure), which can be used to identify abnormal parameter patterns corresponding to equipment malfunctions or changes in the condition of patients undergoing invasive mechanical ventilation, providing a core analytical tool for subsequent abnormal pattern recognition of vital signs and monitoring terminal data.

[0038] The pattern recognition unit utilizes a vital sign abnormality pattern recognizer to perform pattern recognition on vital sign monitoring data sequences to obtain abnormality pattern analysis results. This unit first performs multi-scale convolution on the vital sign monitoring data sequences (including ECG, SpO2, blood pressure, body temperature, etc.) to determine a multi-scale vital sign monitoring feature set. Then, it randomly and non-overlapping pairwise combinations of this set are performed to obtain a multi-scale vital sign monitoring feature combination set. After iterative analysis and mean processing, the target vital sign monitoring feature is extracted. Subsequently, the target vital sign monitoring feature is approximated and matched with multiple sample vital sign abnormality monitoring feature templates pre-stored in the vital sign abnormality pattern recognizer. The sample vital sign abnormality monitoring feature template corresponding to the maximum matching similarity is used as the matching template. The aggregated sample vital sign abnormality type corresponding to this template is then used as the vital sign abnormality pattern analysis result, achieving accurate identification of vital sign abnormality patterns during the transport of patients under invasive mechanical ventilation.

[0039] The abnormal pattern analysis result acquisition unit utilizes the monitoring terminal abnormal pattern recognizer to perform pattern recognition on the multi-dimensional parameter sequence of the monitoring terminal to obtain the abnormal pattern analysis result. First, it constructs the monitoring terminal abnormal pattern recognizer by performing similar union processing and mapping aggregation on multiple sample monitoring terminal abnormal monitoring features and sample abnormal types. This recognizer pre-stores abnormal feature templates for multi-dimensional parameters such as tidal volume, airway pressure, and FiO2 (e.g., parameter fluctuation patterns when the patient and machine are out of sync, and pressure drop features when there is a leak in the tubing). Multi-scale convolution processing is then performed on the monitoring terminal multi-dimensional parameter sequence to extract target monitoring features. These features are then approximated and matched with the sample abnormal monitoring feature templates in the recognizer. The aggregated sample abnormal type corresponding to the template with the highest similarity is taken as the monitoring terminal abnormal pattern analysis result, thereby enabling the identification of abnormalities in the operating status of equipment such as transport ventilators and changes in the condition of patients undergoing invasive mechanical ventilation.

[0040] In one possible implementation, the smart wearable device module 10 further includes:

[0041] Each vital sign monitoring data sequence includes at least ECG, SpO2, blood pressure, and body temperature.

[0042] Specifically, each vital sign monitoring data point in the sequence includes at least an electrocardiogram (ECG), oxygen saturation (SpO2), blood pressure, and body temperature. The ECG is acquired via a flexible chest patch, reflecting the heart's electrical activity. SpO2 and blood pressure are acquired by wristband sensors to assess oxygenation and circulatory function. Body temperature is monitored in real-time via the flexible chest patch, reflecting the basal metabolic status of patients on invasive mechanical ventilation. These data are acquired sequentially to form a continuous vital sign monitoring data sequence. After multi-scale convolution, feature combination, and iterative analysis, target vital sign features are extracted and matched with pre-stored abnormal pattern templates. This provides crucial data support for identifying abnormal vital signs (such as hypoxemia and arrhythmias) during the transport of patients on invasive mechanical ventilation.

[0043] In one possible implementation, the anomaly pattern recognizer extraction unit further includes:

[0044] The abnormality type acquisition unit is used to acquire multiple sample vital sign abnormality monitoring features and multiple sample vital sign abnormality types.

[0045] An anomaly monitoring feature set acquisition unit is used to perform a union of similar types on the multiple sample abnormality types to obtain multiple aggregated sample abnormality types, and to map and aggregate the multiple sample abnormality monitoring features to obtain multiple sample abnormality monitoring feature sets.

[0046] The abnormal monitoring feature template acquisition unit is used to traverse the multiple sample abnormal monitoring feature sets and perform mean processing to obtain multiple sample abnormal monitoring feature templates.

[0047] An abnormal pattern recognizer construction unit is used to perform a one-to-one mapping association between the multiple aggregated sample abnormality types and the multiple sample abnormality monitoring feature templates to construct the abnormality pattern recognizer.

[0048] Specifically, the abnormality type acquisition unit is used to acquire multiple abnormality monitoring features and corresponding abnormality types of samples from historical clinical data. The abnormality monitoring features include specific data characteristics such as electrocardiogram (ECG) waveform distortion, oxygen saturation (SpO2) below 90%, sudden rise / fall in blood pressure exceeding a preset threshold, and abnormally high body temperature. Abnormality types include severe hypoxemia and hemodynamic instability. By collecting and structurally storing the above sample data, basic data support is provided for subsequent aggregation analysis of abnormality patterns and the construction of a recognition device.

[0049] The abnormal monitoring feature set acquisition unit is used to perform a union of similar abnormality types of samples acquired by the abnormality type acquisition unit. That is, abnormality types with the same clinical nature or similar pathological mechanisms are merged. For example, different degrees of hypoxemia are merged into a single aggregated abnormality type of hypoxemia, thus obtaining multiple aggregated abnormality types of samples. At the same time, the abnormality monitoring features of multiple samples are mapped and aggregated according to the corresponding abnormality type. All features belonging to the same aggregated abnormality type (such as features such as SpO2 consistently below 90% and abnormally increased respiratory rate corresponding to hypoxemia) are integrated into a set, thereby obtaining multiple abnormality monitoring feature sets of samples. This lays the foundation for subsequent feature template extraction and construction of an abnormality pattern recognizer.

[0050] The abnormality monitoring feature template acquisition unit is used to traverse and process multiple sample abnormality monitoring feature sets. For each abnormality monitoring feature in each set (such as SpO2 value and respiratory rate corresponding to hypoxemia), representative typical feature parameters are calculated by averaging, thereby forming multiple sample abnormality monitoring feature templates. For example, the average SpO2 value of all samples in the hypoxemia feature set is calculated to obtain the feature template threshold for this abnormality type. In this way, a corresponding feature template is constructed for each aggregated sample abnormality type, providing a standardized comparison benchmark for subsequent abnormality pattern recognition.

[0051] The abnormal pattern recognizer building unit is used to map and associate multiple aggregated sample abnormality types (such as clinical classifications like severe hypoxemia and hemodynamic disturbances) with sample abnormality monitoring feature templates (such as the SpO2 mean threshold corresponding to hypoxemia and the blood pressure fluctuation feature template corresponding to hemodynamic disturbances) one-to-one. By establishing such mapping relationships, each aggregated abnormality type is bound to its typical feature template, constructing an abnormality pattern recognizer. This allows the recognizer to quickly identify the corresponding abnormality type (such as identifying arrhythmias through ECG waveform feature templates) based on the input vital sign monitoring data sequence, providing an analytical model for real-time abnormality early warning during transport.

[0052] In one possible implementation, the pattern recognition unit further includes:

[0053] The target vital sign monitoring feature acquisition unit is used to extract vital sign monitoring features from the vital sign monitoring data sequence to obtain target vital sign monitoring features.

[0054] An approximate matching unit is used to approximate match the target vital sign monitoring features with multiple sample vital sign abnormality monitoring feature templates in the vital sign abnormality pattern recognizer, take the sample vital sign abnormality monitoring feature template corresponding to the maximum matching similarity as the matching template, and take the aggregated sample vital sign abnormality type corresponding to the matching template as the vital sign abnormality pattern parsing result.

[0055] Specifically, the target vital sign monitoring feature acquisition unit is used to extract vital sign monitoring features from the vital sign monitoring data sequence (including ECG, SpO2, blood pressure, body temperature, etc.). First, the data sequence is processed by multi-scale convolution to generate a multi-scale vital sign monitoring feature set. Then, the features in this set are randomly and non-overlappingly combined in pairs to form a multi-scale vital sign monitoring feature combination set. After that, the iterative multi-scale vital sign monitoring feature set is determined through iterative analysis. Finally, the set is mean-processed to obtain the target vital sign monitoring features that can accurately characterize the current vital sign status of patients under invasive mechanical ventilation, providing key data support for subsequent abnormal pattern recognition.

[0056] The approximate matching unit is used to calculate the similarity between the target vital sign monitoring features and multiple sample vital sign abnormality monitoring feature templates (such as the blood oxygen saturation threshold corresponding to severe hypoxemia, blood pressure fluctuation features corresponding to hemodynamic disturbances, etc.) pre-stored in the vital sign abnormality pattern recognizer. The degree of feature matching is measured by Euclidean distance. The sample vital sign abnormality monitoring feature template corresponding to the maximum matching similarity is determined as the matching template. Then, the aggregated sample vital sign abnormality type (such as severe hypoxemia, etc.) corresponding to the template is used as the vital sign abnormality pattern parsing result, so as to realize the accurate identification and classification of vital sign abnormalities during the transfer of invasive mechanical ventilation patients, and provide a basis for subsequent graded alarm and clinical intervention.

[0057] In one possible implementation, the target vital sign monitoring feature acquisition unit further includes:

[0058] The monitoring feature set determination unit is used to perform multi-scale convolution on the vital sign monitoring data sequence to determine the multi-scale vital sign monitoring feature set.

[0059] The monitoring feature combination set acquisition unit is used to randomly and non-overlappingly combine the multi-scale vital sign monitoring feature sets in pairs to obtain the multi-scale vital sign monitoring feature combination set.

[0060] The iterative analysis unit is used to perform iterative analysis on the multi-scale vital sign monitoring feature combination set to determine the iterative multi-scale vital sign monitoring feature set.

[0061] The mean processing unit is used to perform mean processing on the iterative multi-scale vital sign monitoring feature set to obtain the target vital sign monitoring features.

[0062] Specifically, the monitoring feature set determination unit is used to perform multi-scale convolution processing on vital sign monitoring data sequences (including ECG, SpO2, blood pressure, body temperature, etc.). By using convolution kernels of different scales to operate on the data sequences, it extracts the features of the signals at different time resolutions (such as heart rate variability features in the high-frequency band and body temperature fluctuation features in the low-frequency band), thereby generating a vital sign feature set covering multiple dimensions of time granularity. This enables multi-scale feature representation of vital sign data of patients undergoing invasive mechanical ventilation, providing basic data support for subsequent feature combination and abnormal pattern recognition.

[0063] The monitoring feature combination set acquisition unit is used to process the multi-scale vital sign monitoring feature set output by the monitoring feature set determination unit. It combines the features in the set in pairs in a random and non-repeating manner (for example, combining the high-frequency features of ECG with the dynamic change features of SpO2, and the time-domain features of blood pressure with the trend features of body temperature). Through this combination method, a multi-scale vital sign monitoring feature combination set covering the correlation between features is generated, thereby exploring the synergistic change patterns of different vital sign parameters at multiple scales, laying the foundation for subsequent screening of key feature combinations through iterative analysis.

[0064] When the iterative analysis unit performs iterative analysis on the multi-scale vital sign monitoring feature combination set, it first calculates the similarity between sub-features for each feature combination using Euclidean distance, normalizes the similarity values ​​to the [0, 1] interval through linear transformation, and then fills the initially empty two-dimensional matrix according to the feature combination order to form the iterative analysis matrix. Then, it uses this matrix as a convolution kernel to perform sliding window convolution operation on any feature vector in the combination. By setting the iteration termination condition (such as the convergence of the loss function or the upper limit of the number of iterations), it optimizes the feature weights multiple times, selects feature combinations that contribute highly to abnormal pattern recognition, and finally determines the iterative multi-scale vital sign monitoring feature set to achieve feature dimensionality reduction and optimization.

[0065] The mean processing unit is used to perform mean processing on the iterative multi-scale vital sign monitoring feature set output by the iterative analysis unit. By using a weighted average method (the weights can be set according to the contribution of the features to abnormality identification), it aggregates and calculates the multi-dimensional feature parameters (such as ECG waveform features at different scales, SpO2 change rate, blood pressure fluctuation amplitude, etc.) in the set, and extracts the typical feature parameters that can comprehensively characterize the current vital sign status of patients under invasive mechanical ventilation, thereby obtaining the target vital sign monitoring features and providing standardized feature input for subsequent matching with the template in the vital sign abnormality pattern recognizer.

[0066] In one possible implementation, such as Figure 2 As shown, the alarm and display module 40 further includes:

[0067] The Level 1 alarm rule judgment unit is used to determine whether the analysis results of the infusion pump infusion rate, the analysis results of the abnormal vital signs mode, the analysis results of the abnormal monitoring terminal mode, and the analysis results of oxygen consumption and storage meet the corresponding Level 1 alarm rules. If so, a Level 1 audible and visual alarm is triggered.

[0068] The secondary alarm rule judgment unit is used to determine whether the analysis results of the infusion pump infusion rate, the analysis results of the abnormal vital signs mode, the analysis results of the abnormal monitoring terminal mode, and the analysis results of oxygen consumption and storage meet the corresponding secondary alarm rules. If so, the secondary audible and visual alarm is triggered.

[0069] Specifically, when the infusion pump infusion rate analysis results show that the infusion rate exceeds the preset danger threshold (such as an abnormally rapid infusion rate), the abnormal signs analysis results identify severe hypoxemia (SpO2 < 90%) and abnormal vital signs, the abnormal monitoring terminal analysis results indicate the presence of emergency equipment malfunctions such as ventilator dislodgement or complete airway obstruction, and the oxygen consumption analysis results show that the oxygen supply is below the safety threshold and cannot be maintained until the end of the transport, this unit will trigger a level one audible and visual alarm, using a red alarm indicator combined with an audible alarm to prompt the transport team to immediately handle the emergency.

[0070] The secondary alarm rule judgment unit is used to determine whether the analysis results of the infusion pump infusion rate, abnormal vital signs pattern, abnormal monitoring terminal pattern, and oxygen consumption reserve meet the corresponding secondary alarm rules. If they do, a secondary audible and visual alarm is triggered. When the infusion pump infusion rate analysis result shows that the infusion rate is close to the preset warning threshold (such as the infusion rate gradually approaching the danger range), the abnormal vital signs pattern analysis result identifies mild hypoxemia, slight fluctuations in blood pressure, or other vital signs requiring attention, the abnormal monitoring terminal pattern analysis result indicates a lack of synchronization between the patient and the machine, an upward trend in airway pressure, and the oxygen consumption reserve analysis result shows that the oxygen reserve is lower than the usual warning value but can still support the patient to the end of the transport, this unit triggers a secondary audible and visual alarm, using a yellow alarm indicator light combined with an audible alarm to remind the transport team to anticipate and handle potential risks.

[0071] In one possible implementation, the alarm and display module 40 further includes:

[0072] The electronic handover form includes at least a time stamp, a route stamp, and an event stamp.

[0073] Specifically, the electronic handover form includes at least time stamps, route stamps, and event stamps. The time stamps are synchronized with the system clock via the meter-level positioning function of the smart wearable device module, automatically recording the precise time of each transfer node (such as the time of leaving the ward, the time of arriving at the elevator, and the time of entering the receiving department). The route stamps track the transfer path in real time based on an electronic map, recording the specific walking route and the time spent on each segment, facilitating post-transfer efficiency review. The event stamps record key events during the transfer process in a structured manner (such as the trigger time of the first-level alarm, medication adjustment operations, equipment failure handling, etc.). Each stamp is associated with a timestamp and location information, enabling real-time data sharing between the transfer team and the receiving department through a tablet terminal. This automatically generates a structured electronic document containing a timeline, route trajectory, and event list, solving the problem of information omissions in traditional verbal handovers and improving the accuracy and efficiency of transfer handovers.

[0074] The electronic handover form marking types and data are shown in Table 1:

[0075] Table 1: Electronic Handover Form

[0076]

[0077] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0078] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0079] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A patient transport intelligent monitoring platform based on Internet of Things and smart wearable technology, characterized in that, The platform includes: The intelligent wearable device module is used to collect vital sign data in a timely manner during the patient's transport process by wearing it on the patient's body surface, and to obtain a vital sign monitoring data sequence; The medical device interconnection module is used to communicate with the transport ventilator, the smart infusion pump and the monitoring terminal, and to collect data from the transport ventilator, the smart infusion pump and the monitoring terminal simultaneously at the same monitoring frequency as the smart wearable device module, so as to obtain the ventilator parameter sequence, the infusion pump infusion rate sequence and the monitoring terminal multidimensional parameter sequence. The data transmission network module is used to transmit the vital sign monitoring data sequence, ventilator parameter sequence, infusion pump infusion rate sequence, and monitoring terminal multidimensional parameter sequence to the monitoring and analysis module for analysis, and to determine the analysis results of infusion pump infusion rate, vital sign abnormality mode, monitoring terminal abnormality mode, and oxygen consumption reserve. The alarm and display module is used to issue graded alarms based on the analysis results of the infusion pump infusion rate, abnormal vital signs, abnormal monitoring terminal, and oxygen consumption. It triggers alarm indicator lights for audible and visual alarms, tracks the transfer route via an electronic map, and displays the electronic handover form in real time on the display terminal.

2. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 1, characterized in that, The smart wearable device module includes a flexible chest patch and a wristband sensor. The flexible chest patch is used to collect electrocardiogram signals and body temperature, and the wristband sensor is used to collect blood oxygen saturation and blood pressure.

3. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 1, characterized in that, include: A threshold determination unit is used to extract the infusion rate threshold determination unit of the monitoring and analysis module, wherein the infusion rate threshold determination unit includes a preset infusion rate threshold. The infusion rate analysis result extraction unit is used to use the infusion rate threshold judge to traverse the infusion rate sequence of the infusion pump for anomaly identification based on the preset infusion rate threshold, and obtain the infusion rate analysis result of the infusion pump.

4. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 1, characterized in that, include: An abnormal pattern recognition extraction unit is used to extract the abnormal pattern recognition of vital signs and the abnormal pattern recognition of the monitoring terminal of the monitoring and analysis module. The pattern recognition unit is used to perform pattern recognition on the vital sign monitoring data sequence using the vital sign abnormality pattern recognizer to obtain the vital sign abnormality pattern analysis result. The abnormal pattern parsing result acquisition unit is used to perform pattern recognition on the multi-dimensional parameter sequence of the monitoring terminal using the abnormal pattern recognizer of the monitoring terminal, and obtain the abnormal pattern parsing result of the monitoring terminal.

5. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 4, characterized in that, Each vital sign monitoring data sequence includes at least ECG, SpO2, blood pressure, and body temperature.

6. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 5, characterized in that, include: Anomaly type acquisition unit is used to acquire multiple sample vital sign anomaly monitoring features and multiple sample vital sign anomaly types; An anomaly monitoring feature set acquisition unit is used to perform a union of similar types on the multiple sample abnormality types to obtain multiple aggregated sample abnormality types, and to map and aggregate the multiple sample abnormality monitoring features to obtain multiple sample abnormality monitoring feature sets. An anomaly monitoring feature template acquisition unit is used to traverse the multiple sample vital sign anomaly monitoring feature sets and perform mean processing to obtain multiple sample vital sign anomaly monitoring feature templates. An abnormal pattern recognizer construction unit is used to perform a one-to-one mapping association between the multiple aggregated sample abnormality types and the multiple sample abnormality monitoring feature templates to construct the abnormality pattern recognizer.

7. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 6, characterized in that, include: The target vital sign monitoring feature acquisition unit is used to extract vital sign monitoring features from the vital sign monitoring data sequence to obtain target vital sign monitoring features; An approximate matching unit is used to approximate match the target vital sign monitoring features with multiple sample vital sign abnormality monitoring feature templates in the vital sign abnormality pattern recognizer, take the sample vital sign abnormality monitoring feature template corresponding to the maximum matching similarity as the matching template, and take the aggregated sample vital sign abnormality type corresponding to the matching template as the vital sign abnormality pattern parsing result.

8. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 7, characterized in that, include: The monitoring feature set determination unit is used to perform multi-scale convolution on the vital sign monitoring data sequence to determine the multi-scale vital sign monitoring feature set; The monitoring feature combination set acquisition unit is used to randomly and non-overlappingly combine the multi-scale vital sign monitoring feature set into pairs to obtain the multi-scale vital sign monitoring feature combination set. The iterative analysis unit is used to perform iterative analysis on the multi-scale vital sign monitoring feature combination set to determine the iterative multi-scale vital sign monitoring feature set. The mean processing unit is used to perform mean processing on the iterative multi-scale vital sign monitoring feature set to obtain the target vital sign monitoring features.

9. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 1, characterized in that, include: The Level 1 alarm rule judgment unit is used to judge whether the analysis results of the infusion pump infusion rate, the analysis results of the abnormal vital signs mode, the analysis results of the abnormal monitoring terminal mode, and the analysis results of oxygen consumption and storage meet the corresponding Level 1 alarm rules. If so, the Level 1 audible and visual alarm is triggered. The secondary alarm rule judgment unit is used to determine whether the analysis results of the infusion pump infusion rate, the analysis results of the abnormal vital signs mode, the analysis results of the abnormal monitoring terminal mode, and the analysis results of oxygen consumption and storage meet the corresponding secondary alarm rules. If so, the secondary audible and visual alarm is triggered.

10. The intelligent patient transport monitoring platform based on the Internet of Things and smart wearable technology as described in claim 1, characterized in that, The electronic handover form includes at least a time stamp, a route stamp, and an event stamp.

Images