A method and system for inter-facility patient transfer monitoring

Abstract

The application discloses a kind of methods and systems for inter-hospital patient transfer monitoring, it is related to medical monitoring technical field, the system includes intelligent monitoring wristband, server platform and application terminal, intelligent monitoring wristband integrates main processor unit, optical heart rate blood oxygen sensor module, temperature sensor, GPS module etc., server platform is deployed based on LSTM long short-term memory network AI early warning engine and intelligent dispatch algorithm, application terminal includes nurse station management platform and medical staff mobile APP, the system realizes the continuous vital sign monitoring of whole course of inter-hospital transfer, multi-modal fusion positioning, AI trend early warning, intelligent scheduling and one-key emergency response, can predict the risk of disease deterioration in advance 5 to 10 minutes, after examination is completed, automatically dispatching vehicle realizes accurate connection, after SOS triggers, it is simultaneously pushed to multiple responsible persons to build collaborative rescue network, effectively guarantee the safety of patient transfer and improve medical service efficiency.

Description

Technical Field

[0001] This invention relates to the field of medical monitoring technology, and more particularly to a method and system for monitoring patient transfer between hospitals. Background Technology

[0002] Currently, there are several technical challenges in patient transfers between hospitals. Transfer vehicles lack continuous vital sign monitoring equipment like those in wards. For cardiovascular patients whose conditions are prone to sudden changes, it is impossible to monitor key indicators such as heart rate, blood oxygen, and body temperature in real time, resulting in significant monitoring blind spots. Existing medical wristbands primarily focus on identification or single in-hospital positioning functions, failing to meet the need for switching between external GPS positioning and precise in-hospital positioning in inter-hospital transfer scenarios. Insufficient positioning accuracy makes it difficult for patients to quickly locate examination departments in unfamiliar hospital areas. Furthermore, the hospital vehicle dispatch system and patient monitoring system are independent, making it difficult for patients to return after completing examinations due to unfamiliarity with the environment. The existing system lacks an intelligent linkage mechanism with the hospital vehicle dispatch system, failing to achieve precise pick-up and dispatch immediately after examinations, leading to long patient waiting times and low hospital vehicle utilization efficiency.

[0003] In the event of an emergency during transport, patients may find it difficult to call for help promptly and effectively, and the request for help may not include precise location and real-time physiological data, delaying rescue opportunities. Current physiological parameter monitoring technologies mostly use static threshold alarms, triggering an alarm only when parameters exceed normal ranges. This is a reactive alarm mechanism and cannot provide early warning of deteriorating conditions. The transport process involves multiple stages, including the transferring department, transport personnel, examination department, and receiving department. Information transmission relies on manual handover, which is error-prone and inefficient, lacking digital end-to-end traceability and quality control methods. Currently, there are many patented technologies in the medical wristband field, for example… For example, CN211402746U discloses an intelligent positioning wristband that only has GPS positioning and basic calling functions, lacking complete closed-loop management for inter-hospital transport. CN42357450A discloses a clinical transport closed-loop system that mainly solves the problem of intra-hospital transport scheduling, but does not cover the special scenarios of inter-hospital transport. CN44975506A discloses an emergency transport monitoring system that focuses on pre-hospital emergency care and is not applicable to inter-hospital examination transport scenarios. These existing technologies cannot solve the problems of closed-loop monitoring, intelligent scheduling and emergency linkage in complex inter-hospital scenarios, which seriously affects the safety and efficiency of inter-hospital transport. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for monitoring patient transfer between hospitals, in order to solve the problems existing in the prior art.

[0005] This invention provides a patient transport monitoring system for inter-hospital use, comprising a smart monitoring wristband, a server platform, and an application terminal. The smart monitoring wristband includes a main processor unit, a blood pressure monitoring module, an optical heart rate and oxygenation sensor module, a body temperature sensor, a communication module, a display screen, multi-function buttons, a red SOS emergency call button, a buzzer vibration motor, and a rechargeable battery. The communication module integrates a GPS module, a Bluetooth 5.0 module, an NB-IoT module, and a UWB tag. The optical heart rate and oxygenation sensor module integrates a green LED, a red LED, an infrared LED, and a photodiode. The server platform includes a data receiving service. The system includes a device, an AI early warning engine server, a scheduling management server, and a database server. The AI ​​early warning engine server deploys a physiological parameter trend prediction model based on an LSTM (Long Short-Term Memory) network. The LSTM model adopts a bidirectional LSTM architecture. The input layer receives heart rate, blood oxygen, and body temperature data for 60 consecutive time steps, the hidden layer contains 128 neuron units, and the output layer outputs the probability value of arrhythmia occurring within the next 10 minutes through a Sigmoid activation function. The scheduling management server implements an intelligent vehicle dispatch algorithm based on the shortest distance and vehicle status. The application terminals include a nurse station management platform and a mobile APP for medical staff.

[0006] Through the above technical solution, utilizing an integrated smart monitoring wristband, server platform, and application terminal, continuous vital sign monitoring and multimodal fusion positioning are achieved throughout the patient transport process. The smart monitoring wristband integrates an optical heart rate and blood oxygen sensor module, a blood pressure monitoring module, a body temperature sensor, a GPS module, a Bluetooth 5.0 module, an NB-IoT module, and a UWB tag. It can achieve positioning via GPS during the out-of-hospital vehicle journey and precise positioning within 3-5 meters and 30 centimeters within the hospital via Bluetooth beacons and UWB base stations. The LSTM (Long Short-Term Memory) network model deployed on the server platform analyzes heart rate, blood oxygen, and body temperature data over 60 consecutive time steps to output the probability of arrhythmia occurring within the next 10 minutes, realizing a shift from traditional post-event alarms to trend-based early warning. The system can predict the risk of a patient's condition worsening 5 to 10 minutes in advance. Based on the shortest distance and vehicle status, the dispatch management server uses an intelligent vehicle dispatch algorithm to automatically select available hospital vehicles and calculate the shortest arrival time after the patient's examination. This solves the problems of patients' difficulty in returning home and the low efficiency of hospital vehicle dispatch. When the SOS emergency call button is triggered, an emergency alarm can be pushed to the nurse station, medical staff's mobile APP, on-duty emergency doctor, and 120 emergency center at the same time. The emergency data package includes the patient's identity, location, physiological parameters, and medical record summary, building a multi-party collaborative emergency response network, which significantly shortens the emergency response time. The entire system realizes digital monitoring, intelligent dispatch, and emergency rescue of the entire process of inter-hospital transfer, effectively ensuring the safety of patient transfer and improving the efficiency of medical services.

[0007] Furthermore, the main processor unit adopts the ARM Cortex-M4 architecture with a working frequency of 168MHz, and has 512KB of flash memory and 128KB of RAM. The optical heart rate and blood oxygen sensor module has a sampling frequency of 100Hz. The body temperature sensor uses a high-precision thermistor with a measurement range of 35 to 42 degrees Celsius and an accuracy of ±0.1 degrees Celsius. The GPS module in the communication module uses the ublox NEO-M8N chip, which supports four major satellite systems: Beidou, GPS, GLONASS, and Galileo, with a positioning accuracy of 2.5 meters. The rechargeable battery uses a polymer lithium battery with a capacity of 300mAh and a rated voltage of 3.7V.

[0008] Furthermore, the smart monitoring wristband achieves multimodal fusion positioning. During the out-of-hospital driving phase, the GPS module in the communication module acquires latitude and longitude coordinates every 5 seconds and uploads them to the server via the NB-IoT module in the communication module. When the GPS signal indicates that it has entered within 500 meters of the target hospital area, the system automatically activates the Bluetooth module in the communication module to scan for Bluetooth beacon devices deployed within the hospital. The Bluetooth beacon devices use the iBeacon protocol and use three-level identifiers (UUID, Major, and Minor) to label building, floor, and area information. The smart monitoring wristband calculates its relative position within the hospital by receiving the RSSI signal strength value of the beacon and combining it with a three-point positioning algorithm, with a positioning accuracy of 3 to 5 meters. More than 4 UWB base stations are deployed around the examination departments. The UWB tag in the communication module sends a ranging signal to the base station every 100 milliseconds. The base station calculates the distance using the time-of-flight ranging method and uses the Chan algorithm for position calculation, with a positioning accuracy within 30 centimeters.

[0009] Furthermore, the red SOS emergency call button adopts a recessed design to prevent accidental touches. When pressed, at least 3N of pressure must be applied and held for more than 1 second. After triggering, the buzzer in the buzzer vibration motor emits a continuous alarm sound. The vibration motor in the buzzer vibration motor vibrates continuously at a frequency of 300Hz. The main processor unit packages and generates an emergency data package, which includes patient identity information, current location coordinates, trigger timestamp, real-time heart rate, blood oxygen, and body temperature values, physiological parameter trend curve data for the most recent 30 minutes, patient medical record summary, and allergy history information. After the emergency data package is uploaded to the server through the NB-IoT module in the communication module, the server simultaneously pushes emergency alarms to the nurse station management platform, the mobile APP of medical staff related to the transfer task, the mobile terminals of the on-duty emergency doctors and security personnel in the patient's current area, and the hospital's 120 emergency center.

[0010] Furthermore, the smart monitoring wristband adopts a dynamic power consumption management strategy. In normal monitoring mode, the optical heart rate and blood oxygen sensor module collects data every 10 seconds, the body temperature sensor collects data every 30 seconds, the GPS module in the communication module locates the device every 5 seconds, the Bluetooth module in the communication module scans for beacons at 100-millisecond intervals, the NB-IoT module in the communication module establishes a connection with the server every 30 seconds to upload data and then immediately disconnects to enter power-saving mode, and the display screen automatically turns off after 30 seconds of inactivity. In standby mode, the optical heart rate and blood oxygen sensor module reduces its sampling frequency to once every 30 seconds, the GPS module in the communication module reduces its location frequency to once every 30 seconds, the Bluetooth module in the communication module extends its scanning interval to 500 milliseconds, the NB-IoT module in the communication module extends its connection to once every 60 seconds, and the display screen remains off.

[0011] A method for monitoring patient transport between hospitals includes the following steps: S101 The nurse creates a transport task sheet in the HIS system, filling in patient information and examination items; S102 The nurse puts a smart monitoring wristband on the patient and scans the wristband's QR code to bind the wristband device ID with the patient information and the transport task sheet; S103 The system sends initialization configuration parameters to the wristband, including the patient's normal physiological parameter threshold, early warning trigger conditions, and personalized parameters of the AI ​​early warning model; S104 The wristband starts monitoring mode and continuously collects the patient's heart rate, blood oxygen, and body temperature data; S105 During transport, the GPS module in the communication module acquires location coordinates every 5 seconds, the optical heart rate and blood oxygen sensor module collects data every 10 seconds, and all data is packaged and uploaded to the server every 30 seconds; S106 The server pushes the physiological parameters to the AI ​​early warning engine, and the LSTM model performs trend analysis on the most recent 60 data points to calculate the risk score for the next 10 minutes. When the risk score exceeds 0.7, the server generates an early warning message. The patient is transported to the nurses' station and to the medical staff's mobile app; S107 After the vehicle arrives at the target hospital area, the GPS signal indicates it is within 500 meters of the hospital area. The wristband automatically switches to Bluetooth positioning mode, scans Bluetooth beacons within the hospital, and calculates the relative position within the hospital using RSSI signal strength values ​​combined with a three-point positioning algorithm; S108 After arriving at the examination department area, the wristband activates UWB ultra-wideband positioning to measure distances with four surrounding base stations; S109 After the patient completes the examination, the wristband sends a return dispatch request to the server; S110 The server dispatch management module queries the list of all available vehicles, calculates the distance and estimated travel time between each vehicle and the patient's location, selects the vehicle with the shortest estimated arrival time, generates a dispatch instruction, and pushes it to the driver's mobile terminal; S111 After the vehicle arrives at the pick-up point, the driver confirms that the patient has been picked up. During the return trip, the wristband continues to monitor the patient's physiological parameters and upload location information; S112 After the vehicle arrives at the patient's ward, the driver confirms delivery, and the system automatically generates an electronic transport record and synchronizes it to the HIS system.

[0012] Furthermore, the workflow of the AI ​​early warning engine in S106 is as follows: each time the server receives a new set of physiological parameter data, it adds the data to the sliding window queue. When the queue accumulates 60 data points, the heart rate sequence, blood oxygen sequence, and body temperature sequence are extracted as input features of the LSTM model. After forward propagation, the model outputs a risk score between 0 and 1, which represents the probability of the patient experiencing arrhythmia in the next 10 minutes. When the score exceeds 0.7, the system generates an early warning message, which includes the patient's name, bed number, current location, risk score value, current heart rate, blood oxygen, and body temperature values, and a trend curve image of the last 5 minutes.

[0013] Furthermore, in any step from S105 to S111, when the patient presses and holds the red SOS emergency call button for more than 1 second, the main processor unit initiates the emergency response process. The buzzer in the buzzer vibration motor emits a continuous, rapid alarm sound, the vibration motor in the buzzer vibration motor vibrates continuously at maximum intensity, the display screen switches to the emergency interface, the main processor unit packages and generates an emergency data packet, and uploads it to the server through the NB-IoT module in the communication module. After receiving the emergency request, the server triggers a multi-party collaborative response mechanism. The first path pushes the emergency alarm to the nurse station management platform, the second path pushes it to the mobile APP of all medical staff related to the patient transfer task, the third path pushes it to the mobile terminals of the on-duty emergency doctors and security personnel in the area based on the patient's current location, and the fourth path sends an emergency notification to the hospital's 120 emergency center and automatically dials the voice communication channel between the patient's wristband and the emergency center.

[0014] The beneficial effects of this invention are: 1. This invention enables continuous vital sign monitoring and intelligent early warning throughout the entire inter-hospital transport process. The optical heart rate and blood oxygen sensor module 6 continuously monitors the patient's heart rate and blood oxygen at a sampling frequency of 100Hz. The blood pressure and body temperature sensor 7 collect body temperature data every 30 seconds. All data is uploaded to the server every 30 seconds via the NB-IoT module 9. The LSTM long short-term memory network model deployed on the server adopts a bidirectional LSTM architecture. The input layer receives physiological parameter data for 60 consecutive time steps, the hidden layer contains 128 neuron units, and the output layer calculates the probability value of arrhythmia occurring within the next 10 minutes through the Sigmoid activation function. When the probability value exceeds 0.7, an early warning is triggered. Compared with the traditional static threshold alarm method, this invention can predict the trend of disease deterioration 5 to 10 minutes in advance, significantly reducing the risk of unexpected transport for high-risk groups such as cardiovascular patients, making up for the blind spots of the lack of monitoring equipment in the transport vehicle, and buying valuable intervention time for medical staff.

[0015] 2. An innovative multimodal fusion positioning and intelligent dispatching mechanism was implemented. During the out-of-hospital journey, the GPS module 9 acquires latitude and longitude coordinates every 5 seconds to achieve a positioning accuracy of 2.5 meters. When entering the target hospital area within 500 meters, the system automatically switches to Bluetooth positioning mode. By receiving the RSSI signal strength value of the Bluetooth beacon and combining it with a three-point positioning algorithm, it achieves in-hospital positioning with an accuracy of 3 to 5 meters. In the examination department area, UWB ultra-wideband positioning technology is used. The UWB tag 9 sends a ranging signal to the base station every 100 milliseconds. The base station uses time-of-flight ranging and the Chan algorithm to achieve accurate positioning within 30 centimeters. After the patient completes the examination, the wristband 1 automatically sends a return dispatch request to the server. The dispatch management server uses an intelligent dispatching algorithm based on the shortest distance and vehicle status to select available hospital vehicles and calculate the estimated arrival time, selecting the vehicle with the shortest arrival time for dispatch. This solves the problems of patients having difficulty returning after completing the examination and hospital vehicles running empty, reducing patient waiting time by more than 60% and improving hospital vehicle utilization efficiency by more than 40%, achieving precise shuttle service with dispatching vehicles immediately after examination completion.

[0016] 3. A one-click triggering multi-party collaborative emergency response mechanism and a full-process digital management system have been established. The SOS emergency call button 5 adopts a recessed design to prevent accidental activation, requiring at least 3N of pressure to be applied for more than 1 second to trigger. After triggering, the buzzer 8 emits an 85-decibel alarm sound, and the vibration motor 8 vibrates continuously at a frequency of 300Hz. The main processor 2 packages and generates a complete emergency data package containing patient identity information, current location coordinates, real-time physiological parameters, trend curves of the last 30 minutes, medical record summary, and allergy history. This data is uploaded to the server via the NB-IoT module 9 through the highest priority channel. The server simultaneously pushes an emergency alarm to the nurses. The system, comprised of a patient station management platform, a mobile app for medical staff involved in the transfer mission, mobile terminals for on-duty emergency physicians and security personnel in the patient's current area, and the hospital's 120 emergency center, has constructed a multi-party collaborative rescue network that enables information-based patient tracing. This reduces emergency response time by more than 50%. The system automatically generates a complete electronic transfer record that includes the start and end times of the transfer, the route, physiological parameter trend curves, warning records, and emergency events. This record is automatically synchronized to the hospital's HIS system and archived with the patient's electronic medical record via the HL7 interface. This enables digital traceability and quality control of the entire transfer process between hospitals, providing data support for continuous improvement in medical quality. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1This is a schematic diagram of the overall structure of the intelligent monitoring wristband of the present invention; Figure 2 This is a schematic diagram of the internal structure of the intelligent monitoring wristband of the present invention; Figure 3 This is a system architecture block diagram of the present invention; Figure 4 This is a flowchart of the software system workflow of the present invention.

[0019] In the diagram: 1. Smart monitoring wristband; 2. Main processor unit; 3. Display screen; 4. Multifunction button; 5. Red SOS emergency call button; 6. Optical heart rate and blood oxygen sensor module; 7. Body temperature sensor; 8. Buzzer vibration motor; 9. Communication module; 10. Rechargeable battery. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0022] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0023] like Figures 1-2As shown, the intelligent monitoring wristband 1 of the present invention includes a medical-grade silicone wristband body 1, a main processor unit 2, a display screen 3, a multi-function button 4, a red SOS emergency call button 5, an optical heart rate and blood oxygen sensor module 6, a body temperature sensor 7, a buzzer 8, a vibration motor 8, a communication module 9, and a rechargeable battery 10. The main processor unit 2 adopts a low-power ARM Cortex-M4 architecture with a working frequency of 168MHz, and has 512KB of flash memory and 128KB of RAM. It is responsible for coordinating sensor data acquisition, location information acquisition, encrypted data transmission, and local AI algorithm calculation. The display screen 3 is a 1.3-inch OLED display. The screen is a monochrome display with a resolution of 128x64 pixels. It communicates with the main processor 2 via an I2C interface and displays the current time, real-time heart rate, blood oxygen saturation percentage, and simple navigation arrow prompts. The multi-function buttons 4 include a power button and a menu button, featuring a waterproof tactile switch design. Long presses turn the device on / off, short presses wake the screen, and switch display interfaces. The red SOS emergency call button 5 has a recessed design to prevent accidental touches; pressing it requires at least 3N of pressure and must be held for more than 1 second. Upon triggering, the buzzer 8 immediately emits a continuous alarm sound. The vibration motor 8 vibrates continuously at a frequency of 300Hz. An optical heart rate and blood oxygen sensor module 6 is also included. It integrates three light sources—green LED, red LED, and infrared LED—and a high-sensitivity photodiode. It uses photoplethysmography (PPG) to measure heart rate and blood oxygen saturation at a sampling frequency of 100Hz, achieving medical-grade accuracy. The body temperature sensor 7 employs a high-precision thermistor, with a measurement range of 35 to 42 degrees Celsius and an accuracy of ±0.1 degrees Celsius. Body temperature is sampled every 30 seconds. The buzzer 8 operates at a frequency of 2.7kHz, achieving a sound pressure level of 85 dB. The vibration motor 8 uses an eccentric rotor design, and the vibration intensity can be adjusted via PWM. The communication module 9 integrates a GPS module 9, a Bluetooth 5.0 module 9, and an NB-IoT module 9. The IoT module 9 and GPS module 9 use the ublox NEO-M8N chip, supporting four major satellite systems: BeiDou, GPS, GLONASS, and Galileo, with a positioning accuracy of 2.5 meters. The Bluetooth 5.0 module 9 supports low-power broadcast and connection modes, with a transmission distance of up to 50 meters. The NB-IoT module 9 operates in the B5 and B8 frequency bands, supporting power-saving mode and extended discontinuous reception mode. The rechargeable battery 10 uses a polymer lithium battery with a capacity of 300mAh and a rated voltage of 3.7V, supporting USB Type-C fast charging. The charging time is 1.5 hours, and the full charge provides up to 48 hours of battery life.

[0024] like Figure 3As shown, the system architecture comprises four layers: a wristband terminal layer, a network transmission layer, a server platform layer, and an application terminal layer. The wristband terminal layer is responsible for collecting patient physiological data and location information. The network transmission layer uploads data via Bluetooth gateways, NB-IoT base stations, and WiFi hotspots. The server platform layer includes a data receiving server, an AI early warning engine server, a scheduling management server, and a database server. The application terminal layer includes a nurse station management platform, a mobile app for medical staff, and an interface to the hospital's HIS system. The AI ​​early warning engine server deploys a physiological parameter trend prediction model based on an LSTM (Long Short-Term Memory) network. This model employs a bidirectional LSTM architecture, with the input layer... The system receives heart rate, blood oxygen, and body temperature data for 60 consecutive time steps. The hidden layer contains 128 neuron units, and the output layer outputs the probability value of arrhythmia occurring within the next 10 minutes through the Sigmoid activation function. When the probability value exceeds 0.7, an early warning is triggered. The scheduling management server implements an intelligent vehicle dispatch algorithm based on the shortest distance and vehicle status. This algorithm first filters out the list of currently idle hospital vehicles, then calculates the straight-line distance between each hospital vehicle and the patient's current location, comprehensively considers the actual road conditions and historical travel time, selects the hospital vehicle with the shortest estimated arrival time for dispatch, and pushes the dispatch instruction to the driver's mobile terminal in real time via the WebSocket protocol.

[0025] The multimodal fusion positioning technology of this invention adopts a layered switching strategy. During the out-of-hospital driving phase, the GPS module 9 acquires latitude and longitude coordinates every 5 seconds and transmits the NMEA format data to the main processor 2 via serial port. After parsing, the main processor 2 uploads the data to the server via the NB-IoT module 9. The server performs reverse address parsing on the Gaode Map API based on the latitude and longitude coordinates to obtain readable street address information. When the GPS signal indicates that the wristband 1 has entered within 500 meters of the target hospital area, the system automatically starts the Bluetooth scanning function to search for Bluetooth beacon devices deployed within the hospital. These beacon devices use the iBeacon protocol and are identified by three levels of identifiers: UUID, Major, and Minor. The wristband 1 marks specific building, floor, and area information. By receiving the RSSI signal strength value of the beacon and combining it with the three-point positioning algorithm, the wristband 1 calculates its relative position within the hospital, achieving a positioning accuracy of 3 to 5 meters. For scenarios requiring higher accuracy, the system enables UWB ultra-wideband positioning technology, deploying more than 4 UWB base stations around the examination department. The UWB tag 9 built into the wristband 1 sends a ranging signal to the base station every 100 milliseconds. The base station calculates the distance using the time-of-flight ranging method and uses the Chan algorithm or Taylor series expansion method for position calculation, achieving a positioning accuracy within 30 centimeters. The switching of positioning modes is automatically managed by the state machine in the main processor 2, requiring no manual intervention.

[0026] The physiological parameter monitoring process of this invention is as follows: the optical heart rate and blood oxygen sensor 6 acquires photoelectric signals every 10 milliseconds; the main processor 2 filters out DC components and high-frequency noise through a bandpass filter, retaining the effective pulse wave signal of 0.5 to 5 Hz; a peak detection algorithm is used to identify the peak point of the pulse wave; the time interval between adjacent peaks is calculated to obtain the instantaneous heart rate; the moving average of the instantaneous heart rate over 60 consecutive seconds is used to obtain the average heart rate; blood oxygen saturation is obtained by calculating the ratio R of the absorbance of red light and infrared light; the percentage of blood oxygen saturation is calculated according to the empirical formula SpO2 = 110 minus 25 times R; the body temperature sensor 7... Temperature is calculated by measuring the resistance change of a thermistor, and the Steinhart-Hart equation is used for temperature conversion to obtain an accurate body temperature value. All physiological parameter data are temporarily stored locally in a circular buffer of the main processor 2. The buffer can store continuous data for the most recent 10 minutes. Data is packaged and encrypted every 30 seconds and uploaded to the server. The encryption uses the AES-128 algorithm, and the key is the session key negotiated with the server when the device is activated. After receiving the data, the server first decrypts and verifies its integrity, and then stores it in the time series database. At the same time, the data is pushed to the AI ​​early warning engine in real time for trend analysis.

[0027] The AI ​​early warning engine works as follows: Each time the server receives a new set of physiological parameter data, it adds the data to a sliding window queue. When the queue accumulates 60 data points, the heart rate, blood oxygen, and body temperature sequences are extracted as input features for the LSTM model. After forward propagation, the model outputs a risk score between 0 and 1, representing the probability of the patient experiencing arrhythmia within the next 10 minutes. When the score exceeds 0.7, the system determines it to be a high-risk state and immediately generates an early warning message. The message includes the patient's name, bed number, current location, risk score, current heart rate, blood oxygen, and body temperature values, and a trend curve image for the last 5 minutes. The early warning message is simultaneously sent to the nurse station management platform and the mobile apps of relevant medical staff via push service. Upon receiving the warning, the nurse station platform pops up a prominent red alert window and plays an alarm sound. Upon receiving the warning, the medical staff's app is notified via system notification and vibrates. Medical staff can click on the warning message to view detailed patient information and physiological parameter trends, and can directly call the patient or accompanying person for confirmation. The early warning record is automatically archived in the database for later review and model optimization.

[0028] The intelligent dispatching process of this invention is as follows: After the patient completes the examination, they can enter the menu mode by briefly pressing the multi-function button 4 on the wristband 1, select the "Prepare to Return" option, and confirm. The wristband 1 then sends a return dispatch request to the server. The request data packet includes the patient ID, current precise location coordinates, timestamp, and battery level information. After receiving the request, the server's dispatching management module first queries the database for the status information of all hospital vehicles, filters out the list of vehicles with an idle status, and then calls the map API to calculate the path distance and estimated travel time between each idle vehicle and the patient's location. After considering the road condition information, the vehicle with the shortest estimated arrival time is selected as the dispatch target, a dispatch instruction is generated, and it is pushed to the driver's mobile terminal via a WebSocket long connection. The instruction content includes the patient's name, pick-up details, and other information. The system displays the detailed address and navigation route, the patient's destination ward, estimated arrival time, and contact number. After receiving the instruction, the driver clicks "Accept Task," and the system updates the vehicle's status to "Executing" and sends a message to the patient's wristband 1 showing the estimated arrival time. When the vehicle's GPS location shows it is within 50 meters of the pick-up point, the system sends a reminder message to the patient's wristband 1 and vibrates to alert them. The patient can view the vehicle's license plate number and driver's name on the wristband 1 display screen 3 for confirmation. After boarding, the driver clicks "Confirm Pickup" on their mobile terminal, and the system records the pick-up completion time and begins the return monitoring process. After the vehicle arrives at the destination ward, the driver clicks "Confirm Delivery" again. The system records the complete timeline and trajectory data of the entire transfer process, automatically generates an electronic transfer record, and synchronizes it to the HIS system.

[0029] The one-button emergency call mechanism of this invention allows patients to press and hold the red SOS button 5 for more than one second whenever they feel unwell or encounter an emergency. Upon detecting the button trigger, the main processor 2 of the wristband 1 immediately initiates the highest-priority emergency response process. First, the buzzer 8 emits a continuous, rapid alarm sound, the vibration motor 8 vibrates continuously at maximum intensity, and the display screen 3 switches to the emergency interface, showing a message that emergency assistance is being called. Simultaneously, the main processor 2 generates an emergency data package, which includes the patient's complete identity information, current precise location coordinates, trigger timestamp, real-time heart rate, blood oxygen, and body temperature values, physiological parameter trend curves for the past 30 minutes, a summary of the patient's medical records, and allergy history information. The data package is uploaded to the server via the NB-IoT module 9 through the highest-priority channel. Upon receiving the emergency request, the server immediately triggers a multi-party collaborative response mechanism, pushing the emergency alarm to the nurse station management platform. A full-screen red alarm window pops up on the platform interface, displaying detailed patient information and real-time location, and plays a loud alarm. Upon the alarm ringing, the second route pushes the emergency alert to the mobile apps of all medical staff involved in the patient transfer, including the responsible nurse in the transferring department, the accompanying medical staff, and the on-duty doctor in the receiving department. The third route determines the patient's current location and automatically pushes the alert to the mobile terminals of the on-duty emergency room doctor and security personnel in that area. The fourth route sends an emergency notification to the hospital's 120 emergency center and automatically establishes a voice communication channel between the patient's wristband 1 and the emergency center. The emergency center doctor can understand the patient's condition in real time through voice and guide the initial on-site treatment. All personnel who receive the emergency alert can view the patient's real-time location map, physiological parameter curves, and medical record information on their mobile terminals and can quickly reach the scene using the one-click navigation function. The first rescuer to arrive at the scene clicks the arrival button on their mobile terminal, the system records the response time, and notifies other personnel. After the rescue is completed, medical staff fill in the emergency treatment record on their mobile terminals, and the system automatically archives the complete information chain of the entire emergency event for post-event analysis and quality improvement.

[0030] The low-power management strategy of this invention is as follows: the main processor 2 dynamically adjusts the power consumption of each module according to the working state of the wristband 1. In normal monitoring mode, the main processor 2 operates in medium-frequency mode, the heart rate and blood oxygen sensor 6 collects data every 10 seconds, and enters standby mode after 5 seconds of data collection. The body temperature sensor 7 collects data every 30 seconds, the GPS module 9 locates the device every 5 seconds, the Bluetooth module 9 scans beacons at 100-millisecond intervals, and the NB-IoT module 9 establishes a connection with the server every 30 seconds, uploads data, and then immediately disconnects to enter power-saving mode. The display screen 3 automatically turns off after 30 seconds of inactivity. In this mode, the average power consumption of the entire device is approximately 15 mA. In static standby mode, when the system detects that the patient remains still for a long time, it automatically enters deep power-saving mode, and the heart rate and blood oxygen sensor 6 reduces the data collection frequency. The sampling frequency is reduced to once every 30 seconds, the GPS module 9 reduces its positioning frequency to once every 30 seconds, the Bluetooth scanning interval is extended to 500 milliseconds, the NB-IoT module 9 extends its connection to once every 60 seconds, and the display screen 3 remains off. In this mode, the average power consumption of the whole device is reduced to 8 mA. In charging mode, after the main processor 2 detects USB power supply, all modules return to normal operating frequency, the display screen 3 remains on to show the charging progress, and the charging chip adopts a constant current and constant voltage two-stage charging strategy. The first 80% of the power is charged with a constant current of 300 mA, and the last 20% is switched to constant voltage charging to gradually reduce the current to cut off. After it is fully charged, it automatically stops charging to prevent overcharging. Through the above dynamic power consumption management, the 300 mA battery 10 can support 48 hours of continuous normal monitoring work, meeting the needs of inter-hospital transfer and long-term standby.

[0031] like Figure 4 As shown, the complete workflow of this invention includes the following steps: S101: Before patient transfer, the nurse in the transferring department creates a transfer task form in the HIS system, filling in the patient's basic information, the transferring department, the target examination department, the examination items, the appointment time, the patient's condition summary, and special precautions.

[0032] S102: The nurse takes out the fully charged smart monitoring wristband 1 from the charging cabinet, checks that the wristband 1's battery level is 100%, puts the wristband 1 on the patient, and adjusts the tightness to a comfortable level that is not easy to fall off.

[0033] S103: The nurse uses a mobile terminal to scan the QR code on the back of wristband 1. The system automatically binds and associates the device ID of wristband 1 with the patient information and the transfer task order, completing the patient identity authentication and task activation.

[0034] S104: The system sends initialization configuration parameters to wristband 1, including the patient's normal physiological parameter threshold, early warning trigger conditions, the location coordinates of the target examination department, and the personalized parameters of the AI ​​early warning model.

[0035] S105: After receiving the configuration, the wristband 1 starts the monitoring mode and begins to continuously collect the patient's heart rate, blood oxygen and body temperature data. The display screen 3 displays the current physiological parameter values.

[0036] S106: The patient is escorted by an accompanying person to the hospital vehicle stop. After boarding, the hospital vehicle driver clicks the departure button on the mobile terminal, and the system records the start time of the transfer.

[0037] S107: During transport, wristband 1 continues to work in normal monitoring mode, GPS module 9 obtains location coordinates every 5 seconds, heart rate and blood oxygen sensor 6 collects data every 10 seconds, and all data is packaged and uploaded to the server every 30 seconds.

[0038] S108: After receiving the data, the server pushes the physiological parameters to the AI ​​early warning engine. The LSTM model performs trend analysis on the most recent 60 data points and calculates the risk score for the next 10 minutes.

[0039] S109: When the risk score calculated by the AI ​​model exceeds 0.7, the server immediately generates an early warning message and pushes it to the nurse station and the mobile APP of medical staff to remind them that the patient may be about to have an abnormal heart rate.

[0040] S110: After receiving the alert, medical staff will check the detailed physiological parameter curves, call the accompanying person to confirm the patient's condition, and guide the appropriate treatment measures if necessary.

[0041] S111: If the patient feels unwell at any time during transport, they can press and hold the red SOS button 5 to trigger a one-button emergency call, and the wristband 1 will immediately emit an audible, visual, and vibration alarm.

[0042] S112: After receiving the SOS signal, the server initiates a multi-party collaborative response mechanism and simultaneously pushes an emergency alarm to the nurses' station, transport personnel, target department, on-duty emergency doctor, security guard, and 120 emergency center.

[0043] S113: Rescue personnel from all parties can view the patient's real-time location and physiological parameters through mobile terminals, quickly rush to the scene to provide emergency treatment, and the system records the response time of each step.

[0044] S114: After the rescue is completed, medical staff fill in the emergency treatment record on the mobile terminal, and the system automatically archives the complete information of the emergency event.

[0045] S115: After the hospital vehicle arrives at the target hospital area, the GPS signal shows that it has entered the 500-meter range of the hospital area. The wristband 1 automatically switches to Bluetooth positioning mode and begins scanning Bluetooth beacons within the hospital.

[0046] S116: Wristband 1 calculates its relative position within the hospital by receiving RSSI signal strength from multiple beacons and using a three-point positioning algorithm, achieving a positioning accuracy of 3 to 5 meters.

[0047] S117: The wristband 1 and display screen 3 display simple navigation arrows to guide the patient to the target examination department. The patient follows the navigation prompts to walk to the examination department.

[0048] S118: Upon arrival at the examination department area, wristband 1 activates UWB ultra-wideband positioning to measure distances with four surrounding base stations, improving positioning accuracy to within 30 centimeters.

[0049] S119: Medical staff in the examination department confirm the patient's identity by scanning the QR code on the wristband 1, and the system automatically pushes the patient's medical record summary and examination item information to the examination equipment terminal.

[0050] S120: After the patient completes the examination, the wristband 1 automatically determines that the examination is complete by detecting the patient's prolonged static state, or the patient can actively select the "Prepare to Return" option in the wristband 1 menu.

[0051] S121: Wristband 1 sends a return scheduling request to the server. The request data packet includes the patient ID, current precise location coordinates, and timestamp.

[0052] S122: After receiving the request, the server scheduling and management module queries the list of all currently idle hospital vehicles and calculates the distance between each vehicle and the patient's location and the estimated travel time.

[0053] S123: The system selects the hospital vehicle with the shortest estimated arrival time, generates a dispatch instruction, and pushes it to the driver's mobile terminal. The instruction includes the patient's location, destination, and navigation route.

[0054] S124: After receiving the instruction, the driver clicks to accept the task. The system updates the hospital vehicle status to "in progress" and pushes a message to the patient's wristband 1 to display the estimated arrival time.

[0055] S125: When the hospital vehicle arrives at the pick-up point according to the navigation route and the GPS positioning shows that it is within 50 meters, the system sends an alert message to the patient's wristband 1 and vibrates.

[0056] S126: The patient confirms the vehicle's license plate number and driver's name by viewing the screen 3 on the wristband 1. After getting in the car, the driver confirms the pick-up of the patient on the mobile terminal.

[0057] S127: During the return journey, wristband 1 continues to monitor the patient's physiological parameters and upload location information, while the server continuously performs AI trend analysis and anomaly warnings.

[0058] S128: After the hospital vehicle arrives at the patient's ward, the driver clicks to confirm delivery, and the system records the end time of the transfer and the complete transfer route.

[0059] S129: The system automatically generates electronic transport records, including transport start and end times, travel routes, physiological parameter trend curves, early warning records, and emergency event records.

[0060] S130: The electronic transport record is automatically synchronized to the hospital's HIS system via an interface, and is linked and archived with the patient's electronic medical record to complete the entire transport process.

[0061] S131: The nurse removes wristband 1 from the patient's wrist, puts it back in the charging cabinet for charging, and the system releases the wristband 1 from the patient's binding relationship. Wristband 1 returns to standby mode to wait for the next use.

[0062] The integration of this invention with the hospital's HIS system adopts the standard HL7 interface protocol. The server interacts with the HIS system via RESTful API. When a transport task is created, the system calls the HIS interface to obtain the patient's complete medical record information, allergy history, diagnostic records, and medication records. During the transport process, physiological parameter data, early warning records, and emergency event records are transmitted back to the HIS system in real time and linked to the patient's electronic medical record. After the transport is completed, the generated electronic transport record is stored in PDF format and accessed through the HIS system's medical record browsing interface. All data interactions are encrypted using HTTPS, and the server deploys digital certificates for identity authentication. Data storage uses the AES-256 encryption algorithm, complying with relevant regulations on medical data security and privacy protection.

[0063] Example 1 (To make the instruction manual easier to understand and avoid touching on sensitive information, the license plate numbers in the following content have been blurred. They are not actual license plate numbers and do not specifically refer to any particular license plate number. They are only used as examples. Actual operation shall prevail in actual use.) A top-tier hospital has two campuses, A and B, approximately 8 kilometers apart, about a 30-minute drive apart. Campus A is a general ward area, while Campus B is a medical technology and examination center equipped with advanced diagnostic equipment such as MRI, CT, and PET-CT. The hospital's cardiology department is located on the 12th floor of Campus A. A 62-year-old male patient, Mr. Zhang, was admitted to the cardiology department of Campus A for "coronary artery disease and unstable angina." After his condition stabilized, he needed to be transferred to the nuclear medicine department of Campus B for myocardial perfusion imaging to assess myocardial ischemia.

[0064] At 8:30 a.m., Li, a nurse in the cardiology department, logged into the HIS system on the computer at the nurses' station, created a transfer task form, filled in the patient's name Zhang, bed number 12, the transferring department cardiology department, the target examination department auxiliary examination room, and the examination item radionuclide myocardial perfusion imaging. The system interface then automatically displayed eye-catching examination instructions in the sidebar: "Note: All metal objects and external electronic devices of the patient must be removed before this examination; appointment time 10:00 a.m.; patient's condition summary: coronary heart disease, unstable angina pectoris, recovery period, NYHA class II heart function; special precautions: allergy to iodine contrast agents." After completing the task, she clicked save.

[0065] Nurse Li took out the No. 6 smart monitoring wristband 1 from the charging cabinet, checked that the display screen 3 showed that the battery was 100%, put it on the left wrist of patient Zhang, and adjusted the tightness of the wristband 1 to the patient's comfort. She used her work phone to open the medical care APP and scanned the QR code on the back of the wristband 1. The system popped up a confirmation binding interface showing the patient's name Zhang, bed number 12, and the device number of wristband 1, WB-006. The nurse clicked to confirm and complete the binding. The display screen 3 of the wristband 1 showed the patient's name and the current heart rate of 72 beats per minute.

[0066] The system sends personalized configuration parameters to the wristband 1. The normal range of heart rate is set to 60 to 100 beats per minute, the normal range of blood oxygen is set to 95% to 100%, and the normal range of body temperature is set to 36.0 to 37.5 degrees Celsius. An immediate alarm is triggered when the heart rate is below 50 or above 120. The AI ​​early warning model adjusts the sensitivity parameters based on the patient's history of myocardial infarction.

[0067] At 9:00 a.m., the caregiver, Wang, escorted the patient, Zhang, down the elevator and onto the hospital vehicle at the designated stop in front of the outpatient building of Hospital A. The driver, Zhao, clicked the departure button on his mobile device to record the transfer start time as 9:02 a.m., and the hospital vehicle then drove away from Hospital A.

[0068] During transport, wristband 1 operates in normal monitoring mode. Heart rate sensor 6 measures the patient's heart rate at 78 beats per minute, blood oxygen saturation at 98%, and body temperature at 36.8 degrees Celsius. GPS module 9 obtains the current longitude 116.3652 and latitude 39.9148, locating the patient at the intersection of a certain road and a certain street. Data is packaged and uploaded to the server every 30 seconds. The server pushes the data to the AI ​​early warning engine. The LSTM model analyzes the heart rate trend of the most recent 60 data points and calculates a heart rate arrhythmia risk score of 0.35 for the next 10 minutes, which is lower than the warning threshold of 0.7. Therefore, the system judges it as a normal state and does not trigger an early warning.

[0069] At 9:15 AM, halfway through the journey, the AI ​​warning engine detected that the patient's heart rate rapidly increased from 78 beats per minute to 105 beats per minute within 3 minutes, and the blood oxygen saturation slightly decreased to 96%. The model recalculated the risk score as 0.78, exceeding the warning threshold. The server immediately generated a warning message stating that the patient, Zhang, bed number 12, was currently located at [location omitted], with a heart rate of 105 beats per minute showing an upward trend, blood oxygen saturation of 96%, and a risk score of 0.78, suggesting attention. The message was also pushed to the cardiology ward nurses' station and to nurse Li's mobile app.

[0070] After receiving the alert, nurse Li immediately checked the detailed heart rate trend chart and found that the heart rate had been rising continuously for the past 3 minutes. She immediately called the caregiver Wang to inquire about the patient's condition. Caregiver Wang said that the patient complained of mild chest tightness. Nurse Li instructed the caregiver to have the patient take deep breaths and relax, and told him to observe closely and press the SOS button immediately if the symptoms worsened.

[0071] Two minutes later, the heart rate began to drop to 90 beats per minute, blood oxygen recovered to 97%, and the AI ​​model risk score dropped to 0.52. Nurse Li recorded the treatment measures on the mobile APP as telephone guidance to the patient to relax and relieve symptoms. The system linked the treatment record to the transfer task order.

[0072] At 9:32 AM, the hospital vehicle arrived at Campus B. GPS positioning indicated that it had entered within 500 meters of Campus B. After detecting the change in GPS coordinates, the main processor 2 of the wristband 1 automatically started Bluetooth scanning and found 5 Bluetooth beacons deployed in Campus B. The UUID is FDA50693-A4E2-4FB1-AFCF-C6EB07647825. The Major value is 2, which identifies Campus B. The Minor values ​​are 101, 102, 103, 104, and 105, which identify different floors and areas. The RSSI values ​​of these beacons received by the wristband 1 are -45dBm, -62dBm, -78dBm, -55dBm, and -70dBm, respectively. Using a three-point positioning algorithm, the current location is calculated to be on the west side of the lobby on the first floor of Campus B, with a positioning accuracy of about 4 meters.

[0073] Patient Zhang, accompanied by caregiver Wang, walked to the auxiliary examination room. The wristband 1 display screen 3 showed an eastward arrow and a navigation prompt indicating a distance of approximately 120 meters. The patient followed the navigation, walked through the corridor into the medical technology building, and took the elevator to the 3rd floor. At this time, the wristband 1 detected a beacon RSSI value of -38dBm with a Minor value of 305, indicating that the patient had arrived at the east side of the 3rd floor and was approximately 30 meters away from the auxiliary examination room.

[0074] Four UWB base stations are deployed at the entrance of the auxiliary examination room. The UWB tag 9 built into the wristband 1 starts ranging with the base stations. The base station numbers are UWB-B301, UWB-B302, UWB-B303, and UWB-B304 respectively, and the ranging values are 8.52 meters, 6.31 meters, 5.18 meters, and 9.74 meters respectively. The wristband 1 uses the Chan algorithm to calculate the precise coordinates as X = 15.3 meters and Y = 8.7 meters, corresponding to the position of the third row of seats in the waiting area of the catheterization laboratory, and the positioning accuracy reaches 25 centimeters.

[0075] At 9:58 am, the auxiliary examination nurse scans the QR code of the patient's wristband 1 through the mobile terminal. The system automatically pushes the patient's medical record summary, showing stable angina pectoris in the recovery period and allergy to iodine contrast agent. After the nurse confirms, she guides the patient into the auxiliary examination room to prepare for the examination. The nurse conducts identity verification, confirms the system pop-up reminder, and according to the prompt "Please remove all metal items and external electronic devices of the patient", assists in removing the wristband 1. During the examination, the intelligent monitoring wristband is kept and the patient is guided to prepare for the examination after confirmation. The system records the time when the patient arrives at the examination department as 9:58 am, and the transfer takes 56 minutes.

[0076] The examination process lasts about 45 minutes. During this period, the wristband 1 works in the static standby mode, and the power consumption is reduced to 8 mA. The examination is completed at 10:43. The nurse immediately re-wears the wristband for the patient, and the system resumes monitoring the patient's vital signs. The main processor 2 automatically determines that the examination is completed and sends a return scheduling request to the server. The request data packet includes the patient ID as P202501230015, the current position coordinates as longitude 116.3812 and latitude 39.9087, and the timestamp as 2025-01-23T10:43:25.

[0077] After receiving the request, the server scheduling management module queries the hospital vehicle status table and finds that the status of the hospital vehicle with license plate number Jing A12345 is idle, and its current position is in the parking lot of Area B. The status of the hospital vehicle with license plate number Jing B67890 is idle, and its current position is in the outpatient building of Area A. The scheduling algorithm calculates that Jing A12345 is 200 meters away from the patient's position and is estimated to arrive in 1 minute, and Jing B67890 is 8 kilometers away from the patient's position and is estimated to arrive in 30 minutes. The system selects Jing A12345 as the dispatched object, generates a vehicle dispatch instruction and pushes it to the mobile terminal of the driver Zhao. The instruction shows that the patient Zhang is to be picked up at bed number 12, the pick-up location is at the entrance of the 3rd floor of the医技楼 in Area B, and the destination is the 12th floor of the Department of Cardiology in Area A. Please go immediately.

[0078] After receiving the instruction, the driver Zhao clicks to accept the task. The system updates the status of Jing A****** to in-execution and pushes a message to the wristband 1 of the patient, showing that the hospital vehicle is expected to arrive in 1 minute. Please wait at the entrance of the catheterization laboratory. The license plate number is Jing A******. The screen 3 of the wristband 1 displays this message and vibrates to prompt the patient.

[0079] One minute later, driver Zhao drove the hospital vehicle to the entrance of the auxiliary examination room on the 3rd floor of the medical technology building. The GPS positioning showed that it was 48 meters away from the target point. The system sent a reminder message to wristband 1 and vibrated it. The display screen 3 of wristband 1 showed that your hospital vehicle has arrived, license plate number: Beijing A12345, driver Master Zhao. Patient Zhang walked out of the catheterization room accompanied by the escort Wang and saw the hospital vehicle with license plate number Beijing A12345. After confirmation, they got on the vehicle.

[0080] Driver Zhao clicked the button to confirm receiving the patient on the mobile terminal. The system recorded the connection completion time as 10:46. The hospital vehicle left from Area B and returned to Area A. On the way, wristband 1 resumed the normal monitoring mode, continuously collected the patient's physiological parameters and uploaded them to the server. The AI early warning engine analyzed the data trend in real time and no abnormal situation was found.

[0081] At 11:18, the hospital vehicle arrived at the entrance of the 12th floor ward of the cardiology department in Area A. Driver Zhao clicked the button to confirm delivering the patient on the mobile terminal. The system recorded the transfer end time as 11:18, and the return journey took �2 minutes.

[0082] The system automatically generated an electronic transfer record. The transfer start time was 9:02, the transfer end time was 11:18, the total time was 136 minutes, the outbound journey took 56 minutes, the examination took 45 minutes, the waiting time took 2 minutes, the return journey took 32 minutes. The driving track recorded the complete GPS coordinate tracks of the outbound and return journeys. The physiological parameter trend curve showed that the heart rate range throughout the journey was 72 to 105 beats per minute, the blood oxygen range was 96% to 98%, and the body temperature was kept constant at 36.8 degrees Celsius. The early warning record showed that a heart rate increase warning was triggered at 9:15. After the nurse Li gave telephone guidance, the symptoms were relieved and no emergency occurred. The electronic transfer record was generated in PDF format and automatically uploaded to the HIS system through the HL7 interface and associated with the electronic medical record of patient Zhang.

[0083] Nurse Li removed wristband 1 numbered 6 from patient Zhang's wrist. The inspection of the display screen 3 showed that the remaining power was 73%, which met the demand for the next use. Wristband 1 was put back into slot 6 of the charging cabinet for charging. The system automatically解除 the binding relationship between the device number WB-006 of wristband 1 and patient Zhang. Wristband 1 returned to the standby state, and the entire transfer process was successfully completed.

[0084] From this embodiment, it can be seen that the intelligent monitoring wristband 1 system of the present invention realizes continuous physiological monitoring, precise positioning, intelligent early warning and automatic scheduling throughout the transfer of patients between hospital areas, effectively guarantees the safety of patient transfer, improves the transfer efficiency, optimizes the medical staff work process, and provides reliable technical support for the cross-hospital area medical service of the hospital.

[0085] Note: There is an unclear expression "系统自动解除腕带1设备号WB-006与患者张某的绑定关系" in the original text. I translated it as "The system automatically解除 the binding relationship between the device number WB-006 of wristband 1 and patient Zhang" with "解除" left as it is because it's not clear what the exact word should be here. You may need to check and correct it according to the actual situation.This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0086] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A patient transfer monitoring system for inter-hospital transfers, characterized in that, The system includes a smart monitoring wristband, a server platform, and an application terminal. The smart monitoring wristband includes a main processor unit, an optical heart rate and blood oxygen sensor module, a body temperature sensor, a communication module, a display screen, a multi-function button, a red SOS emergency call button, a buzzer vibration motor, and a rechargeable battery (10). The communication module integrates a GPS module, a Bluetooth 5.0 module, an NB-IoT module, and a UWB tag. The blood pressure monitoring module and the optical heart rate and blood oxygen sensor module integrate green LED, red LED, infrared LED, and photodiode. The server platform includes a data receiving server, an AI early warning engine server, a scheduling management server, and a database server. The AI ​​early warning engine server deploys a physiological parameter trend prediction model based on an LSTM long short-term memory network. The LSTM model adopts a bidirectional LSTM architecture. The input layer receives heart rate, blood oxygen, and body temperature data for 60 consecutive time steps. The hidden layer contains 128 neuron units. The output layer outputs the probability value of heart rate arrhythmia occurring within the next 10 minutes through a Sigmoid activation function. The scheduling management server implements an intelligent vehicle dispatch algorithm based on the shortest distance and vehicle status. The application terminal includes a nurse station management platform and a mobile APP for medical staff.

2. The patient transfer monitoring system between hospitals according to claim 1, characterized in that, The main processor unit adopts the ARM Cortex-M4 architecture, with a working frequency of 168MHz, and has 512KB flash memory and 128KB memory. The sampling frequency of the optical heart rate and blood oxygen sensor module is 100Hz. The body temperature sensor adopts a high-precision thermistor with a measurement range of 35 to 42 degrees Celsius and an accuracy of ±0.1 degrees Celsius. The GPS module in the communication module adopts the ubloxNEO-M8N chip, which supports the four major satellite systems of Beidou, GPS, GLONASS and Galileo, with a positioning accuracy of 2.5 meters. The rechargeable battery (10) adopts a polymer lithium battery with a capacity of 300mAh and a rated voltage of 3.7V.

3. The patient transfer monitoring system between hospitals according to claim 1, characterized in that, The smart monitoring wristband achieves multimodal fusion positioning. During the out-of-hospital driving phase, the GPS module in the communication module acquires latitude and longitude coordinates every 5 seconds and uploads them to the server via the NB-IoT module in the communication module. When the GPS signal indicates that it has entered within 500 meters of the target hospital area, the system automatically activates the Bluetooth module in the communication module to scan for Bluetooth beacon devices deployed within the hospital. The Bluetooth beacon devices use the iBeacon protocol and use three-level identifiers (UUID, Major, and Minor) to mark building, floor, and area information. The smart monitoring wristband calculates its relative position within the hospital by receiving the RSSI signal strength value of the beacon and combining it with a three-point positioning algorithm, with a positioning accuracy of 3 to 5 meters. More than 4 UWB base stations are deployed around the examination departments. The UWB tag in the communication module sends a ranging signal to the base station every 100 milliseconds. The base station calculates the distance using the time-of-flight ranging method and uses the Chan algorithm for position calculation, with a positioning accuracy within 30 centimeters.

4. The patient transfer monitoring system between hospitals according to claim 1, characterized in that, The red SOS emergency call button features a recessed design to prevent accidental activation. When pressed, at least 3N of pressure must be applied and held for more than 1 second. Upon triggering, the buzzer in the buzzer vibration motor emits a continuous alarm sound. The vibration motor in the buzzer vibration motor vibrates continuously at a frequency of 300Hz. The main processor unit packages and generates an emergency data package, which includes patient identification information, current location coordinates, trigger timestamp, real-time heart rate, blood oxygen, and body temperature values, physiological parameter trend curve data for the most recent 30 minutes, patient medical record summary, and allergy history information. The emergency data package is uploaded to the server via the NB-IoT module in the communication module. The server then simultaneously pushes emergency alarms to the nurse station management platform, the mobile APP of medical staff involved in the transfer task, the mobile terminals of on-duty emergency doctors and security personnel in the patient's current area, and the hospital's 120 emergency center.

5. The patient transfer monitoring system between hospitals according to claim 1, characterized in that, The smart monitoring wristband employs a dynamic power consumption management strategy. In normal monitoring mode, the optical heart rate and blood oxygen sensor module collects data every 10 seconds, the body temperature sensor collects data every 30 seconds, the GPS module in the communication module locates the device every 5 seconds, the Bluetooth module in the communication module scans for beacons at 100-millisecond intervals, the NB-IoT module in the communication module establishes a connection with the server every 30 seconds to upload data and then immediately disconnects to enter power-saving mode, and the display screen automatically turns off after 30 seconds of inactivity. In standby mode, the optical heart rate and blood oxygen sensor module reduces its sampling frequency to once every 30 seconds, the GPS module in the communication module reduces its location frequency to once every 30 seconds, the Bluetooth module in the communication module extends its scanning interval to 500 milliseconds, the NB-IoT module in the communication module extends its connection to once every 60 seconds, and the display screen remains off.

6. A method for monitoring patient transfer between hospitals, characterized in that, The system according to claim 1 includes the following steps: S101 The nurse creates a transport task sheet in the HIS system, filling in patient information and examination items; S102 The nurse puts a smart monitoring wristband on the patient and scans the wristband's QR code to bind the wristband device ID with the patient information and the transport task sheet; S103 The system sends initialization configuration parameters to the wristband, including the patient's normal physiological parameter threshold, early warning trigger conditions, and personalized parameters of the AI ​​early warning model; S104 The wristband starts the monitoring mode and continuously collects the patient's heart rate, blood oxygen, and body temperature data; S105 During transport, the GPS module in the communication module acquires location coordinates every 5 seconds, the optical heart rate and blood oxygen sensor module collects data every 10 seconds, and all data is packaged and uploaded to the server every 30 seconds; S106 The server pushes the physiological parameters to the AI ​​early warning engine, the LSTM model performs trend analysis on the most recent 60 data points to calculate the risk score for the next 10 minutes, and when the risk score exceeds 0.7, the server generates an early warning message and pushes it to the system. The system includes mobile apps for nurses' stations and medical staff; S107: After the hospital vehicle arrives at the target hospital area, the GPS signal indicates it is within 500 meters of the hospital area. The wristband automatically switches to Bluetooth positioning mode, scans Bluetooth beacons within the hospital, and calculates the relative position within the hospital using RSSI signal strength values ​​combined with a three-point positioning algorithm; S108: After arriving at the examination department area, the wristband activates UWB ultra-wideband positioning to measure distances with four surrounding base stations; S109: After the patient completes the examination, the wristband sends a return dispatch request to the server; S110: The server dispatch management module queries the list of all available hospital vehicles, calculates the distance and estimated travel time between each vehicle and the patient's location, selects the vehicle with the shortest estimated arrival time, generates a dispatch instruction, and pushes it to the driver's mobile terminal; S111: After the hospital vehicle arrives at the pick-up point, the driver confirms that the patient has been picked up. During the return trip, the wristband continues to monitor the patient's physiological parameters and upload location information; S112: After the hospital vehicle arrives at the patient's ward, the driver confirms delivery, and the system automatically generates an electronic transport record and synchronizes it to the HIS system.

7. The method for monitoring patient transfer between hospitals according to claim 6, characterized in that, The workflow of the AI ​​early warning engine in S106 is as follows: each time the server receives a new set of physiological parameter data, it adds the data to the sliding window queue. When the queue accumulates 60 data points, the heart rate sequence, blood oxygen sequence, and body temperature sequence are extracted as input features of the LSTM model. After forward propagation, the model outputs a risk score between 0 and 1, which represents the probability of the patient experiencing arrhythmia in the next 10 minutes. When the score exceeds 0.7, the system generates an early warning message, which includes the patient's name, bed number, current location, risk score value, current heart rate, blood oxygen, and body temperature values, and a trend curve image of the last 5 minutes.

8. The method for monitoring patient transfer between hospitals according to claim 6, characterized in that, In any step from S105 to S111, when the patient presses and holds the red SOS emergency call button for more than 1 second, the main processor unit initiates the emergency response process. The buzzer in the buzzer vibration motor emits a continuous, rapid alarm sound, the vibration motor in the buzzer vibration motor vibrates continuously at maximum intensity, the display screen switches to the emergency interface, the main processor unit packages and generates an emergency data packet, and uploads it to the server through the NB-IoT module in the communication module. After receiving the emergency request, the server triggers a multi-party collaborative response mechanism. The first path pushes the emergency alarm to the nurse station management platform, the second path pushes it to the mobile APP of all medical staff related to the patient transfer task, the third path pushes it to the mobile terminals of the on-duty emergency doctors and security personnel in the area based on the patient's current location, and the fourth path sends an emergency notification to the hospital's 120 emergency center and automatically dials the voice communication channel between the patient's wristband and the emergency center.

Images