A hospital 5G smart LED wristband and its monitoring and positioning system

By integrating a variety of sensors and positioning modules on the medical wristband, real-time monitoring and early warning of patient vital signs and locations is achieved, solving the problem of single functions in the existing technology, and improving the quality of medical care and patient safety.

CN118203436BActive Publication Date: 2024-11-15SECOND MEDICAL CENT OF CHINESE PLA GENERAL HOSPITAL
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410524612.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-29
Publication Date
2024-11-15
Estimated Expiration
2044-04-29

AI Technical Summary

Technical Problem

The existing medical wristbands only record basic information such as the patient's name, and cannot monitor the patient's vital signs and location in real time, resulting in a decrease in the quality of medical care and the inability to fully ensure the safety of patients.

Method used

Design a hospital 5G intelligent LED wristband that integrates temperature sensors, heart rate sensors, breathing frequency sensors, pulse oxygen saturation sensors, motion sensors, GPS positioning modules and 5G communication modules to monitor patients' vital signs in real time through these sensors and modules. and position, promptly warn of abnormal status and risk of falls in patients.

Benefits of technology

It has improved the efficiency of monitoring the patients' basic status, location and fall risks by clinical medical staff, timely perceived and warned of abnormal status of patients, reduced the risk of patients leaving the hospital without authorization, improved the quality of medical care, and fully guaranteed the safety of patients.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118203436B_ABST
    Figure CN118203436B_ABST
Patent Text Reader

Abstract

The present invention discloses a 5G smart LED wristband for hospitals and its monitoring and positioning system, which belongs to the field of medical health monitoring technology, and includes an outer shell, on which an LED display screen is arranged, and the LED display screen is used to display the patient's vital signs information. The present invention solves the problem that the existing wristband, when in use, has no other function except recording the basic information such as the patient's name, has a single function, cannot monitor the patient's vital signs and position in real time, reduces the quality of medical care, and cannot fully guarantee the safety of patients. The present invention will greatly improve the monitoring efficiency of clinical medical care on the basic status, position and risk of falling of patients, etc., can perceive and warn patients of abnormal status more timely, can improve the real-time monitoring ability of medical staff on the status of patients, reduce the risk of patients leaving the hospital without authorization, and timely warn patients of safety events such as falling, which can improve the quality of medical care and fully guarantee the safety of patients.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of medical health monitoring technology, and specifically to a hospital 5G smart LED wristband and a monitoring and positioning system thereof. Background Art

[0002] In the medical field, wristbands are used to provide patient information, displaying routine information such as the patient's name, gender, age, admission diagnosis, and drug allergy information. They are an effective tool for hospitals to identify patients.

[0003] The Chinese patent with publication number CN210931943U discloses a medical wrist band and a medical wrist band alarm system, wherein the medical wrist band comprises a wrist band seat, a left wrist band, a right wrist band and an adjusting piece, wherein the wrist band seat, the left wrist band, the adjusting piece and the right wrist band are sequentially connected to form a ring; a paper with patient information is provided on the wrist band seat; the adjusting piece comprises a moving rod, a sleeve and a locking cylinder, wherein one end of the moving rod is fixed on the left wrist band, and the other end of the moving rod is embedded in one end of the sleeve; the other end of the sleeve is fixed on the right wrist band; the locking cylinder is sleeved on the sleeve and is threadedly connected with the sleeve; a plurality of notches are provided at one end of the sleeve; when worn, the locking cylinder can be rotated at any time to loosen the moving rod, and the length of the wrist band can be adjusted by changing the length of the moving rod embedded in the sleeve to adapt to changes in the patient's condition; however, the patent has the following defects:

[0004] Existing wristbands, when in use, only record basic information such as the patient's name and have no other function. They are single-function and cannot monitor the patient's vital signs and location in real time, which reduces the quality of medical care and cannot fully ensure patient safety. Summary of the invention

[0005] The purpose of the present invention is to provide a hospital 5G smart LED wristband and its monitoring and positioning system, which will greatly improve the monitoring efficiency of clinical medical staff on the basic conditions, positions and fall risks of patients, can perceive and warn patients' abnormal conditions more timely, and can improve the real-time monitoring ability of medical staff on patients' conditions, reduce the risk of patients leaving the hospital without authorization, and timely warn patients of safety incidents such as falls, which can improve the quality of medical care, fully ensure patient safety, and solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A 5G smart LED wristband for hospitals comprises an outer shell, one end of which is provided with a first connecting belt, the other end of which is provided with a second connecting belt, the first connecting belt and the second connecting belt are connected by a connecting piece, and an LED display screen is provided on the outer shell, and the LED display screen is used to display the patient's vital signs information.

[0008] Preferably, a temperature sensor, a heart rate sensor, a respiratory rate sensor, a pulse oxygen saturation sensor and a battery power management module are arranged on the inner side of the outer shell, and the temperature sensor, the heart rate sensor, the respiratory rate sensor and the pulse oxygen saturation sensor are all electrically connected to the battery power management module, wherein:

[0009] Temperature sensor, used to dynamically sense the patient's body temperature in real time;

[0010] Heart rate sensor, used to dynamically sense the patient's heart rate in real time;

[0011] Respiratory rate sensor, used to dynamically sense the patient's respiratory rate in real time;

[0012] Pulse oxygen saturation sensor, used to dynamically sense the patient's pulse oxygen saturation in real time;

[0013] Based on the real-time dynamic perception of the patient's body temperature, heart rate, respiratory rate and pulse oxygen saturation, the patient's vital signs information is determined.

[0014] Preferably, a motion sensor, a GPS positioning module and a 5G communication module are also provided on the inner side of the outer shell, and the motion sensor, the GPS positioning module and the 5G communication module are all electrically connected to the battery power management module, wherein:

[0015] Motion sensors are used to dynamically sense the patient's gait while walking, and to determine changes in the patient's body position in real time, and to warn of the occurrence of a patient fall based on changes in the patient's body position;

[0016] The GPS positioning module is used to dynamically sense the patient's location in real time and locate the patient in real time, and issue an early warning when the patient's location deviates, making it easier to find the patient;

[0017] The 5G communication module is used to establish a network connection between the wristband and the monitoring terminal, and transmit the wristband monitoring and positioning information to the monitoring terminal in real time through the 5G network.

[0018] Preferably, the motion sensor warns of the occurrence of a patient falling based on a change in the patient's body position, including:

[0019] Determine a preset database, perform data processing on the preset database, and obtain a target matrix; the preset database includes p pieces of data, and each piece of data contains the values ​​of n indicators included in the physical status indicators of the wearer of the smart LED wristband, forming a matrix Q, then the matrix Q contains p rows and n columns, and each piece of data corresponds to whether a fall occurs, forming a judgment vector Y;

[0020] Perform equalization processing on the matrix Q to obtain the processing matrix Z;

[0021]

[0022]

[0023] Among them, ZL(Q i ,Q X ) is the value after equalization of the difference between the i-th column and the x-th column of the matrix Q; Qjx is the value of the j-th row and x-th column of the matrix Q; Q ji is the value of the jth row and ith column of the matrix Q; i = 1, 2, 3...n, x = 1, 2, 3...n, j = 1, 2, 3...p;

[0024] Calculate the characteristic coefficients of the processing matrix Z;

[0025] |Z-λE|=0

[0026] C = ranked(λ)

[0027]

[0028] Where D is the characteristic coefficient of the processing matrix Z; E is the unit matrix, λ is the transformation coefficient solved by Z-λE|=0; λ is a vector containing n values; ranked(λ) is the data after sorting the vector λ from large to small; C is the conversion vector, Ci is the i-th value of the conversion vector C, i=1, 2, 3...n;

[0029] Determine whether the characteristic coefficient of the processing matrix Z is within a preset range. When it is determined that the characteristic coefficient of the processing matrix Z is not within the preset range, perform equalization processing on the processing matrix Z. When it is determined that the characteristic coefficient of the processing matrix Z is within the preset range, use the processing matrix Z as a qualified target matrix.

[0030] The patient's body position changes are matched with the data in the target matrix, and the corresponding values ​​in the judgment vector Y corresponding to the matched data are used to determine whether the patient has fallen, and an early warning is issued for the occurrence of a patient fall.

[0031] According to another aspect of the present invention, a monitoring and positioning system for a 5G smart LED wristband for a hospital is provided, which is built into the above-mentioned 5G smart LED wristband for a hospital, and includes:

[0032] A data collection module is used to collect the gait information and patient position information of the patient when walking that are dynamically sensed by the motion sensor and the GPS positioning module in real time, and determine the real-time data of the patient's gait position based on the gait information and patient position information of the patient when walking;

[0033] A data processing module is used to perform comprehensive processing on the real-time data of the patient's gait position based on retrieval, feature extraction and calculation to determine the patient's gait position representation data;

[0034] A data analysis module, used to comprehensively analyze the patient's gait position representation data and determine the patient monitoring and positioning analysis results based on the patient's gait position;

[0035] The early warning and control module is used to carry out early warning and control of patient monitoring and positioning analysis results.

[0036] Preferably, the data collection module includes:

[0037] A data acquisition unit, used to acquire the patient's walking gait information and patient position information dynamically sensed by the motion sensor and GPS positioning module in real time;

[0038] The data integration unit is used to integrate the gait information and patient position information obtained when the patient is walking, and determine the real-time gait position data of the patient.

[0039] Preferably, the data processing module includes:

[0040] A data retrieval unit, used to retrieve real-time data of the patient's gait position;

[0041] Acquire the real-time data of the patient's gait position, retrieve the real-time data of the patient's gait position one by one based on the sequential retrieval method, filter out the repeated or missing related data contained in the real-time data of the patient's gait position, and determine the real-time data of the patient's gait position that is valuable for monitoring and positioning;

[0042] A feature extraction unit, used for extracting features from real-time data of the patient's gait position;

[0043] Acquire the real-time data of the patient's gait position, and perform feature extraction on the real-time data of the patient's gait position to determine the patient's gait position feature data;

[0044] A data calculation unit, used for calculating the patient's gait position characteristic data;

[0045] The patient's gait position characteristic data is obtained, and the patient's gait position characteristic data is calculated to determine the patient's gait position representation data.

[0046] Preferably, the data analysis module includes:

[0047] A data storage unit, used to store pre-set patient gait position standard data for monitoring and positioning the patient;

[0048] A data indexing unit, used to index pre-set patient gait position standard data;

[0049] Acquire patient gait position representation data, and index pre-set patient gait position standard data based on the patient gait position representation data;

[0050] A data retrieval unit, used to retrieve pre-set standard data of patient gait position;

[0051] Obtaining the indexed pre-set patient gait position standard data, and retrieving the patient gait position standard data;

[0052] A comparison and analysis unit, used for comparing and analyzing the patient's gait position representation data;

[0053] Acquire patient gait position representation data and patient gait position standard data, compare and analyze the patient gait position representation data based on the patient gait position standard data, and determine the patient monitoring and positioning analysis results based on the patient gait position;

[0054] In the case where the patient's gait position representation data is within the patient's gait position standard data range, the patient monitoring and positioning analysis result is that the patient monitoring and positioning are normal;

[0055] In the case where the patient's gait position characterization data is not within the patient's gait position standard data range, the patient monitoring and positioning analysis result is that the patient monitoring and positioning is abnormal, and the patient needs to be monitored and positioned early warning.

[0056] Preferably, the early warning control module includes:

[0057] The early warning management unit is used to monitor, locate and warn patients, and to manage patients intelligently. It can remotely transmit early warning information to the monitoring terminal, monitor patient position changes and locate patients in real time to facilitate finding patients.

[0058] Preferably, the feature extraction unit comprises:

[0059] A classification module is used to classify the real-time data of gait position to obtain text data and image data;

[0060] The first extraction module is used to:

[0061] Determine the dimension of the high-dimensional sparse vector based on the data length of the text data; map the text data to a vector corresponding to the dimension based on a pre-built dictionary to obtain a high-dimensional sparse vector of characters corresponding to the text data;

[0062] Decompose the high-dimensional sparse vector of the text into high and low frequency bands to obtain a group of data in different frequency bands; select b text features from the data in different frequency bands in group a respectively, and obtain a*b eigenvalues ​​by statistics; sort the n eigenvalues ​​according to the correlation criterion, and determine the text features based on the sorting results;

[0063] The second extraction module is used to:

[0064] Determining the dimension of the high-dimensional sparse vector based on the resolution of the image data;

[0065] Mapping the image data to a vector corresponding to the dimension based on a pre-built dictionary to obtain a high-dimensional sparse vector of the image corresponding to the image data;

[0066] A target vector is randomly selected from the high-dimensional sparse vectors of the image and determined as the center of the cluster; the high-dimensional sparse vectors of the image that are within a preset distance range from the target vector of the center of the cluster are clustered to obtain a cluster set; the centers of several clusters are determined according to the number of high-dimensional sparse vectors of the image that are outside the preset distance range from the target vector of the center of the cluster; based on the centers of several clusters, the target vectors are clustered to determine several cluster sets; the target vectors in each cluster set are respectively connected to determine an association graph;

[0067] Inputting the association graph into a pre-trained image encoder to output image features;

[0068] The determination module is used to determine the patient's gait position feature data based on text features and image features.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] 1. The present invention can dynamically sense the gait of patients while they are walking in real time through motion sensors, can detect the gait of patients and warn of falls, judge the changes in the patient's body position in real time and warn of the occurrence of patient falls, etc. The patient's position can be located in real time through the GPS positioning module. Adverse events caused by occasional unauthorized departure of hospitalized patients occur from time to time. The GPS positioning module can warn medical staff of the deviation of the patient's position in the hospital and locate the patient's position for easy search. Compared with the simple wristbands used in previous hospitals, which only record basic information such as the patient's name and have no other function, the 5G smart LED wristband will greatly improve the efficiency of clinical medical staff in monitoring the basic status, position and risk of falls of patients, and can sense and warn of abnormal status of patients more timely.

[0071] 2. The present invention collects the gait information and patient position information of the patient while walking that are dynamically sensed in real time by the motion sensor and the GPS positioning module, determines the real-time data of the patient's gait position based on the gait information and patient position information of the patient while walking, performs comprehensive processing on the real-time data of the patient's gait position based on retrieval, feature extraction and calculation, determines the patient's gait position representation data, performs comprehensive analysis on the patient's gait position representation data, determines the patient monitoring and positioning analysis results based on the patient's gait position, performs early warning control on the patient monitoring and positioning analysis results, remotely transmits early warning information to the monitoring terminal, monitors the patient's body position changes and locates the patient in real time, facilitates the search for patients, can improve the medical staff's real-time monitoring ability of the patient's status, reduce the risk of patients leaving the hospital without authorization, and promptly warn patients of safety incidents such as falls, thereby improving the quality of medical care and fully ensuring the safety of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 This is a forward schematic diagram of the hospital 5G smart LED wristband of the present invention;

[0073] Figure 2 This is a module structure diagram of the hospital 5G smart LED wristband of the present invention;

[0074] Figure 3 This is a module schematic diagram of the monitoring and positioning system of the hospital 5G smart LED wristband of the present invention;

[0075] Figure 4 This is an algorithm flow chart of the monitoring and positioning system of the hospital 5G smart LED wristband of the present invention.

[0076] In the figure: 1. outer shell; 11. first connecting belt; 12. second connecting belt; 13. connecting piece; 14. LED display screen. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] In order to solve the problem that the existing wristbands only record the basic information such as the patient's name when in use, have no other functions, are single-function, cannot monitor the patient's vital signs and location in real time, reduce the quality of medical care, and cannot fully protect the patient's safety, please refer to Figure 1-Figure 4 , this embodiment provides the following technical solutions:

[0079] A 5G smart LED wristband for hospitals comprises an outer shell 1, a first connecting belt 11 is arranged at one end of the outer shell 1, a second connecting belt 12 is arranged at the other end of the outer shell 1, and the first connecting belt 11 and the second connecting belt 12 are connected by a connecting piece 13.

[0080] It should be noted that the first connecting belt 11 and the second connecting belt 12 are connected by a connecting piece 13, and the 5G smart LED wristband can be worn on the patient's wrist for real-time dynamic perception of the patient's vital signs information.

[0081] An LED display screen 14 is provided on the outer shell 1, and the LED display screen 14 is used to display the patient's vital sign information.

[0082] In this embodiment, a temperature sensor, a heart rate sensor, a respiratory rate sensor, a pulse oxygen saturation sensor and a battery power management module are arranged on the inner side of the outer shell 1, and the temperature sensor, the heart rate sensor, the respiratory rate sensor and the pulse oxygen saturation sensor are all electrically connected to the battery power management module, wherein:

[0083] Temperature sensor, used to dynamically sense the patient's body temperature in real time;

[0084] Heart rate sensor, used to dynamically sense the patient's heart rate in real time;

[0085] Respiratory rate sensor, used to dynamically sense the patient's respiratory rate in real time;

[0086] Pulse oxygen saturation sensor, used to dynamically sense the patient's pulse oxygen saturation in real time;

[0087] Based on the real-time dynamic perception of the patient's body temperature, heart rate, respiratory rate and pulse oxygen saturation, the patient's vital signs information is determined.

[0088] It should be noted that the temperature sensor, heart rate sensor, respiratory rate sensor and pulse oxygen saturation sensor can respectively sense the patient's body temperature, heart rate, respiratory rate and pulse oxygen saturation in real time and dynamically to determine the patient's vital signs information.

[0089] In this embodiment, a motion sensor, a GPS positioning module and a 5G communication module are also provided on the inner side of the outer shell 1, and the motion sensor, the GPS positioning module and the 5G communication module are all electrically connected to the battery power management module, wherein:

[0090] Motion sensors are used to dynamically sense the patient's gait while walking, and to determine changes in the patient's body position in real time, and to warn of the occurrence of a patient fall based on changes in the patient's body position;

[0091] The GPS positioning module is used to dynamically sense the patient's location in real time and locate the patient in real time, and issue an early warning when the patient's location deviates, making it easier to find the patient;

[0092] The 5G communication module is used to establish a network connection between the wristband and the monitoring terminal, and transmit the wristband monitoring and positioning information to the monitoring terminal in real time through the 5G network.

[0093] It should be noted that motion sensors can be used to dynamically perceive the patient's gait while walking in real time, detect the patient's gait and fall warning, judge the patient's position changes in real time, and warn the patient of falls.

[0094] It should be noted that the GPS positioning module can be used to locate the patient's position in real time. Inpatients occasionally leave the hospital without permission, resulting in adverse events. The GPS positioning module can warn medical staff of deviations in the position of in-hospital patients and locate the patient's position for easy search.

[0095] Compared with the simple wristbands used in previous hospitals, which only record basic information such as the patient's name and have no other function, the 5G smart LED wristband will greatly improve the efficiency of clinical medical staff in monitoring the patient's basic conditions, position, and fall risks, and can perceive and warn patients of abnormal conditions more timely.

[0096] Preferably, the motion sensor warns of the occurrence of a patient falling based on a change in the patient's body position, including:

[0097] Determine a preset database, perform data processing on the preset database, and obtain a target matrix; the preset database includes p pieces of data, and each piece of data contains the values ​​of n indicators included in the physical status indicators of the wearer of the smart LED wristband, forming a matrix Q, then the matrix Q contains p rows and n columns, and each piece of data corresponds to whether a fall occurs, forming a judgment vector Y;

[0098] Perform equalization processing on the matrix Q to obtain the processing matrix Z;

[0099]

[0100]

[0101] Among them, ZL(Q i ,Q X ) is the value after equalization of the difference between the i-th column and the x-th column of the matrix Q; Qjx is the value of the j-th row and x-th column of the matrix Q; Q ji is the value of the jth row and ith column of the matrix Q; i = 1, 2, 3...n, x = 1, 2, 3...n, j = 1, 2, 3...p;

[0102] Calculate the characteristic coefficients of the processing matrix Z;

[0103] |Z-λE|=0

[0104] C = ranked(λ)

[0105]

[0106] Where D is the characteristic coefficient of the processing matrix Z; E is the unit matrix, λ is the transformation coefficient solved by Z-λE|=0; λ is a vector containing n values; ranked(λ) is the data after sorting the vector λ from large to small; C is the conversion vector, Ci is the i-th value of the conversion vector C, i=1, 2, 3...n;

[0107] Determine whether the characteristic coefficient of the processing matrix Z is within a preset range. When it is determined that the characteristic coefficient of the processing matrix Z is not within the preset range, perform equalization processing on the processing matrix Z. When it is determined that the characteristic coefficient of the processing matrix Z is within the preset range, use the processing matrix Z as a qualified target matrix.

[0108] The patient's body position changes are matched with the data in the target matrix, and the corresponding values ​​in the judgment vector Y corresponding to the matched data are used to determine whether the patient has fallen, and an early warning is issued for the occurrence of a patient fall.

[0109] The working principle and beneficial effects of the above technical solution: When the motion sensor warns the patient of the occurrence of falling based on the patient's body position change, the patient's body position change is usually matched with a preset database (including several preset data), and whether the patient has fallen is determined according to the matching result, so as to improve the accuracy of the judgment. However, in the process of matching the patient's body position change with the preset database, it is necessary to process the preset database. The preset database includes p data, and each data contains the values ​​of n indicators included in the physical state indicators of the wearer of the smart LED wristband, forming a matrix Q. Then the matrix Q contains p rows and n columns, and each data corresponds to whether the patient has fallen, forming a judgment vector Y; however, the numerical differences and data indicators in the preset database are too many and different, and it is necessary to perform equalization processing to better display the characteristics of each data, perform equalization processing on the matrix Q, and obtain the processing matrix Z; calculate the characteristic coefficient of the processing matrix Z; determine whether the characteristic coefficient of the processing matrix Z is within the preset range. When it is determined that the characteristic coefficient of the processing matrix Z is not within the preset range, it means that the processing matrix obtained by processing is unqualified and needs to be equalized again. When it is determined that the characteristic coefficient of the processing matrix Z is within a preset range, the processing matrix Z is used as a qualified target matrix; matching is performed based on the patient's body position changes and the data in the target matrix, which facilitates improving the rate and accuracy of data matching, and facilitates rapid determination of whether the patient has fallen based on the corresponding values ​​in the judgment vector Y corresponding to the matched data, thereby providing an early warning of the occurrence of a patient fall.

[0110] In order to better demonstrate the monitoring and positioning process of the hospital 5G smart LED wristband, this embodiment also provides a monitoring and positioning system for the hospital 5G smart LED wristband, which is built into the above-mentioned hospital 5G smart LED wristband, including:

[0111] A data collection module is used to collect the gait information and patient position information of the patient when walking that are dynamically sensed by the motion sensor and the GPS positioning module in real time, and determine the real-time data of the patient's gait position based on the gait information and patient position information of the patient when walking;

[0112] In this embodiment, the data collection module includes:

[0113] A data acquisition unit, used to acquire the patient's walking gait information and patient position information dynamically sensed by the motion sensor and GPS positioning module in real time;

[0114] The data integration unit is used to integrate the gait information and patient position information obtained when the patient is walking, and determine the real-time gait position data of the patient.

[0115] A data processing module is used to perform comprehensive processing on the real-time data of the patient's gait position based on retrieval, feature extraction and calculation to determine the patient's gait position representation data;

[0116] In this embodiment, the data processing module includes:

[0117] A data retrieval unit, used to retrieve real-time data of the patient's gait position;

[0118] Acquire the real-time data of the patient's gait position, retrieve the real-time data of the patient's gait position one by one based on the sequential retrieval method, filter out the repeated or missing related data contained in the real-time data of the patient's gait position, and determine the real-time data of the patient's gait position that is valuable for monitoring and positioning;

[0119] A feature extraction unit, used for extracting features from real-time data of the patient's gait position;

[0120] Acquire the real-time data of the patient's gait position, and perform feature extraction on the real-time data of the patient's gait position to determine the patient's gait position feature data;

[0121] A data calculation unit, used for calculating the patient's gait position characteristic data;

[0122] The patient's gait position characteristic data is obtained, and the patient's gait position characteristic data is calculated to determine the patient's gait position representation data.

[0123] A data analysis module, used to comprehensively analyze the patient's gait position representation data and determine the patient monitoring and positioning analysis results based on the patient's gait position;

[0124] In this embodiment, the data analysis module includes:

[0125] A data storage unit, used to store pre-set patient gait position standard data for monitoring and positioning the patient;

[0126] A data indexing unit, used to index pre-set patient gait position standard data;

[0127] Acquire patient gait position representation data, and index pre-set patient gait position standard data based on the patient gait position representation data;

[0128] A data retrieval unit, used to retrieve pre-set standard data of patient gait position;

[0129] Obtaining the indexed pre-set patient gait position standard data, and retrieving the patient gait position standard data;

[0130] A comparison and analysis unit, used for comparing and analyzing the patient's gait position representation data;

[0131] Acquire patient gait position representation data and patient gait position standard data, compare and analyze the patient gait position representation data based on the patient gait position standard data, and determine the patient monitoring and positioning analysis results based on the patient gait position;

[0132] In the case where the patient's gait position representation data is within the patient's gait position standard data range, the patient monitoring and positioning analysis result is that the patient monitoring and positioning are normal;

[0133] In the case where the patient's gait position characterization data is not within the patient's gait position standard data range, the patient monitoring and positioning analysis result is that the patient monitoring and positioning is abnormal, and the patient needs to be monitored and positioned early warning.

[0134] The early warning and control module is used to carry out early warning and control of patient monitoring and positioning analysis results.

[0135] In this embodiment, the early warning control module includes:

[0136] The early warning management unit is used to monitor, locate and warn patients, and to manage patients intelligently. It can remotely transmit early warning information to the monitoring terminal, monitor patient position changes and locate patients in real time to facilitate finding patients.

[0137] It should be noted that by collecting the patient's walking gait information and patient position information perceived in real time by the motion sensor and the GPS positioning module, the patient's gait position real-time data is determined based on the patient's walking gait information and patient position information, and the patient's gait position real-time data is comprehensively processed based on retrieval, feature extraction and calculation to determine the patient's gait position representation data, and the patient's gait position representation data is comprehensively analyzed to determine the patient monitoring and positioning analysis results based on the patient's gait position, and early warning control is performed on the patient monitoring and positioning analysis results, and early warning information is remotely transmitted to the monitoring terminal. The patient's body position changes are monitored and located in real time to facilitate finding patients, which can improve the medical staff's real-time monitoring ability of the patient's status, reduce the risk of patients leaving the hospital without authorization, and timely warn patients of safety incidents such as falls, thereby improving the quality of medical care and fully ensuring patient safety.

[0138] Preferably, the feature extraction unit comprises:

[0139] A classification module is used to classify the real-time data of gait position to obtain text data and image data;

[0140] The first extraction module is used to:

[0141] Determine the dimension of the high-dimensional sparse vector based on the data length of the text data; map the text data to a vector corresponding to the dimension based on a pre-built dictionary to obtain a high-dimensional sparse vector of characters corresponding to the text data;

[0142] Decompose the high-dimensional sparse vector of the text into high and low frequency bands to obtain a group of data in different frequency bands; select b text features from the data in different frequency bands in group a respectively, and obtain a*b eigenvalues ​​by statistics; sort the n eigenvalues ​​according to the correlation criterion, and determine the text features based on the sorting results;

[0143] The second extraction module is used to:

[0144] Determining the dimension of the high-dimensional sparse vector based on the resolution of the image data;

[0145] Mapping the image data to a vector corresponding to the dimension based on a pre-built dictionary to obtain a high-dimensional sparse vector of the image corresponding to the image data;

[0146] A target vector is randomly selected from the high-dimensional sparse vectors of the image and determined as the center of the cluster; the high-dimensional sparse vectors of the image that are within a preset distance range from the target vector of the center of the cluster are clustered to obtain a cluster set; the centers of several clusters are determined according to the number of high-dimensional sparse vectors of the image that are outside the preset distance range from the target vector of the center of the cluster; based on the centers of several clusters, the target vectors are clustered to determine several cluster sets; the target vectors in each cluster set are respectively connected to determine an association graph;

[0147] Inputting the association graph into a pre-trained image encoder to output image features;

[0148] The determination module is used to determine the patient's gait position feature data based on text features and image features.

[0149] The working principle of the above technical solution: In this embodiment, machine learning or deep learning algorithms are used to classify data to obtain text data and image data.

[0150] In this embodiment, when constructing a high-dimensional sparse vector based on text data, it is first necessary to determine the dimension of the high-dimensional sparse vector. Specifically, the text data is segmented and the size of the vocabulary is counted. The size of the vocabulary is usually used as the dimension of the high-dimensional sparse vector. For example, if there are 10,000 words in the vocabulary, the dimension of the high-dimensional sparse vector can be set to 10,000 dimensions. Mapping text data to high-dimensional sparse vectors: Pre-construct a dictionary: Build a dictionary based on the training data set, the dictionary contains all the words that appear in the text data, and assign a unique index value to each word. Map text data to high-dimensional sparse vectors: For each text data, each word therein is mapped to the corresponding high-dimensional sparse vector according to the dictionary. Specifically, the text data can be represented as a sparse vector, in which each dimension represents a word in the dictionary. If the text data contains the word, the value of the corresponding dimension is the frequency of the word in the text or other representation. Obtaining a high-dimensional sparse vector corresponding to the text data: After the above mapping process, we can obtain a high-dimensional sparse vector of text corresponding to the text data, whose dimension is consistent with the dimension of the high-dimensional sparse vector determined above. Mapping the text data to a high-dimensional sparse vector space facilitates subsequent text processing and analysis.

[0151] In this embodiment, the high-dimensional sparse vector of the text is decomposed into high and low frequency bands to obtain a group of data of different frequency bands: first, the features in the high-dimensional sparse vector are grouped according to their frequencies in the entire data set and divided into different frequency bands. The features can be divided into high frequency bands and low frequency bands according to the frequency of the features or other evaluation indicators. The high frequency band represents the features with higher frequency in the data set, and the low frequency band represents the features with lower frequency. B text features are selected from the data of different frequency bands in a group, and a*b feature values ​​are obtained by statistics: for each frequency band data, b text features are selected therefrom (can be selected according to specific rules or randomly), and then the values ​​of these features are counted to obtain a*b feature values. According to the correlation criterion, n feature values ​​are sorted, and the text features are determined based on the sorting results: for the obtained a*b feature values, the correlation analysis method (such as Pearson correlation coefficient) can be used to calculate the correlation between the features. According to the size of the correlation, these feature values ​​are sorted, and the sorting results can be used as the selection criteria for text features. Based on the sorting results, the final text features are determined, and the feature engineering processing of the high-dimensional sparse vector of the text is performed to determine the final text features, thereby realizing feature selection and dimensionality reduction of the text data.

[0152] In this embodiment, when constructing a high-dimensional sparse vector based on image data, the dimension of the high-dimensional sparse vector can be determined according to the resolution of the image. The resolution of the image can be expressed in the form of width×height×number of channels, where the number of channels represents the color channels of the image (such as 3 for RGB images). For example, if the resolution of an RGB image is 100×100 pixels, it can be represented as a high-dimensional sparse vector with a dimension of 100×100×3=30,000. Mapping image data to a high-dimensional sparse vector:

[0153] Pre-build a dictionary: Build a dictionary based on the training data set, which contains all possible features in the image (such as pixel values, color distribution, etc.), and assign a unique index value to each feature. Map image data to high-sparse vectors: For each image, map each feature in the image data to the corresponding high-dimensional sparse vector according to the dictionary. The image can be represented as a sparse vector, where each dimension represents a feature in the dictionary. If the image data contains the feature, the value of the corresponding dimension can represent the feature value or the frequency of occurrence of the feature, etc. Obtain a high-dimensional sparse vector corresponding to the image data: Through the above mapping process, a high-dimensional sparse vector corresponding to the image data can be obtained, and its dimension is consistent with the dimension of the high-dimensional sparse vector determined in the step. Through the above steps, the image data can be mapped to a high-dimensional sparse vector space, which is conducive to subsequent image processing and analysis.

[0154] In this embodiment, a target vector is arbitrarily selected from the high-dimensional sparse vectors of the image and determined as the center of the cluster; the high-dimensional sparse vectors of the image that are within a preset distance range from the target vector of the center of the cluster are clustered to obtain a cluster set; the centers of several clusters are determined according to the number of high-dimensional sparse vectors of the image that are outside the preset distance range from the target vector of the center of the cluster; based on the centers of several clusters, the target vectors are clustered to determine several cluster sets; the target vectors in each cluster set are connected respectively to determine an association graph; cluster analysis of the high-dimensional sparse vectors of the image is implemented, and finally an association graph is determined, thereby revealing the intrinsic relationship and similarity between the image data.

[0155] In this embodiment, a method for obtaining a pre-trained image encoder includes inputting an image pre-training set into an image encoder, outputting an image feature vector that can characterize image features, and restoring the image feature vector to an image using a principal component analysis method; using a cross entropy loss function to calculate a loss value between a real image and a restored image, and iteratively updating all parameters of the image encoder using a gradient descent method until the loss value no longer changes, thereby obtaining a pre-trained image encoder. The principal component analysis method is to average all elements in the image feature vector and then calculate the covariance matrix of the image feature vector, perform eigenvalue decomposition on the covariance matrix, obtain a feature vector, and use the feature vector to restore the image.

[0156] The beneficial effects of the above technical solution are: data classification is performed on the real-time data of gait position to obtain text data and image data; feature extraction is performed on the text data and image data based on different methods to obtain text features and image features, and the patient's gait position feature data is accurately determined based on the text features and image features.

[0157] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0158] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A monitoring and positioning system for a hospital 5G smart LED wristband, built into a hospital 5G smart LED wristband, wherein: A 5G smart LED wristband for a hospital, comprising an outer shell (1), one end of the outer shell (1) being provided with a first connecting belt (11), the other end of the outer shell (1) being provided with a second connecting belt (12), the first connecting belt (11) and the second connecting belt (12) being connected via a connecting member (13), the outer shell (1) being provided with an LED display screen (14), the LED display screen (14) being used to display vital sign information of a patient; the invention is characterized in that it comprises: A data collection module is used to collect the gait information and patient position information of the patient when walking that are dynamically sensed by the motion sensor and the GPS positioning module in real time, and determine the real-time data of the patient's gait position based on the gait information and patient position information of the patient when walking; A data processing module is used to perform comprehensive processing on the real-time data of the patient's gait position based on retrieval, feature extraction and calculation to determine the patient's gait position representation data; A data analysis module, used to comprehensively analyze the patient's gait position representation data and determine the patient monitoring and positioning analysis results based on the patient's gait position; Early warning control module, used for early warning control of patient monitoring and positioning analysis results; The data processing module comprises: A data retrieval unit, used to retrieve real-time data of the patient's gait position; Acquire the real-time data of the patient's gait position, retrieve the real-time data of the patient's gait position one by one based on the sequential retrieval method, filter out the repeated or missing related data contained in the real-time data of the patient's gait position, and determine the real-time data of the patient's gait position that is valuable for monitoring and positioning; A feature extraction unit, used for extracting features from real-time data of the patient's gait position; Acquire the real-time data of the patient's gait position, and perform feature extraction on the real-time data of the patient's gait position to determine the patient's gait position feature data; A data calculation unit, used for calculating the patient's gait position characteristic data; Acquiring patient gait position characteristic data, and calculating the patient gait position characteristic data to determine the patient gait position representation data; The feature extraction unit comprises: A classification module is used to classify the real-time data of gait position to obtain text data and image data; The first extraction module is used to: Determine the dimension of the high-dimensional sparse vector based on the data length of the text data; map the text data to a vector corresponding to the dimension based on a pre-built dictionary to obtain a high-dimensional sparse vector of characters corresponding to the text data; Decompose the high-dimensional sparse vector of the text into high and low frequency bands to obtain a group of data in different frequency bands; select b text features from the data in different frequency bands in group a respectively, and obtain a*b eigenvalues ​​by statistics; sort the n eigenvalues ​​according to the correlation criterion, and determine the text features based on the sorting results; The second extraction module is used to: Determining the dimension of the high-dimensional sparse vector based on the resolution of the image data; Mapping the image data to a vector corresponding to the dimension based on a pre-built dictionary to obtain a high-dimensional sparse vector of the image corresponding to the image data; A target vector is randomly selected from the high-dimensional sparse vectors of the image and determined as the center of the cluster; the high-dimensional sparse vectors of the image that are within a preset distance range from the target vector of the center of the cluster are clustered to obtain a cluster set; the centers of several clusters are determined according to the number of high-dimensional sparse vectors of the image that are outside the preset distance range from the target vector of the center of the cluster; based on the centers of several clusters, the target vectors are clustered to determine several cluster sets; the target vectors in each cluster set are respectively connected to determine an association graph; Inputting the association graph into a pre-trained image encoder to output image features; The determination module is used to determine the patient's gait position feature data based on text features and image features.

2. According to claim 1, a monitoring and positioning system for a hospital 5G smart LED wristband is characterized in that: A temperature sensor, a heart rate sensor, a respiratory rate sensor, a pulse oxygen saturation sensor and a battery power management module are arranged on the inner side of the outer shell (1); the temperature sensor, the heart rate sensor, the respiratory rate sensor and the pulse oxygen saturation sensor are all electrically connected to the battery power management module, wherein: Temperature sensor, used to dynamically sense the patient's body temperature in real time; Heart rate sensor, used to dynamically sense the patient's heart rate in real time; Respiratory rate sensor, used to dynamically sense the patient's respiratory rate in real time; Pulse oxygen saturation sensor, used to dynamically sense the patient's pulse oxygen saturation in real time; Based on the real-time dynamic perception of the patient's body temperature, heart rate, respiratory rate and pulse oxygen saturation, the patient's vital signs information is determined.

3. A monitoring and positioning system for a 5G smart LED wristband in a hospital according to claim 2, characterized in that: A motion sensor, a GPS positioning module and a 5G communication module are also arranged on the inner side of the outer shell (1), and the motion sensor, the GPS positioning module and the 5G communication module are all electrically connected to the battery power management module, wherein: Motion sensors are used to dynamically sense the patient's gait while walking, and to determine changes in the patient's body position in real time, and to warn of the occurrence of a patient fall based on changes in the patient's body position; The GPS positioning module is used to dynamically sense the patient's location in real time and locate the patient in real time, and issue an early warning when the patient's location deviates, making it easier to find the patient; The 5G communication module is used to establish a network connection between the wristband and the monitoring terminal, and transmit the wristband monitoring and positioning information to the monitoring terminal in real time through the 5G network.

4. A monitoring and positioning system for a 5G smart LED wristband in a hospital according to claim 3, characterized in that: The motion sensor warns of the occurrence of a patient falling based on a change in the patient's body position, including: Determine a preset database, perform data processing on the preset database, and obtain a target matrix; the preset database includes p pieces of data, and each piece of data contains the values ​​of n indicators included in the physical status indicators of the wearer of the smart LED wristband, forming a matrix Q, then the matrix Q contains p rows and n columns, and each piece of data corresponds to whether a fall occurs, forming a judgment vector Y; perform equalization processing on the matrix Q to obtain a processing matrix Z; in, is the value after equalizing the difference between the i-th column and the x-th column of the matrix Q; is the value of the jth row and xth column of the matrix Q; is the value of the jth row and ith column of the matrix Q; i=1, 2, 3...n, x=1, 2, 3...n, j=1, 2, 3...p; Calculate the characteristic coefficients of the processing matrix Z; Where D is the characteristic coefficient of the processing matrix Z; is the identity matrix, To pass The transformed coefficients solved; is a vector containing n values; To vector The data is sorted from large to small; C is the conversion vector, is the i-th value of the conversion vector C, i=1, 2, 3...n; judging whether the characteristic coefficient of the processing matrix Z is within a preset range, and when it is determined that the characteristic coefficient of the processing matrix Z is not within the preset range, performing equalization processing on the processing matrix Z; when it is determined that the characteristic coefficient of the processing matrix Z is within the preset range, taking the processing matrix Z as a qualified target matrix; The patient's body position changes are matched with the data in the target matrix, and the corresponding values ​​in the judgment vector Y corresponding to the matched data are used to determine whether the patient has fallen, and an early warning is issued for the occurrence of a patient fall.

5. A monitoring and positioning system for a 5G smart LED wristband in a hospital according to claim 4, characterized in that: The data collection module comprises: A data acquisition unit, used to acquire the patient's walking gait information and patient position information dynamically sensed by the motion sensor and GPS positioning module in real time; The data integration unit is used to integrate the gait information and patient position information obtained when the patient is walking, and determine the real-time gait position data of the patient.

6. A monitoring and positioning system for a 5G smart LED wristband for a hospital according to claim 5, characterized in that: The data analysis module includes: A data storage unit, used to store pre-set patient gait position standard data for monitoring and positioning the patient; A data indexing unit, used to index pre-set patient gait position standard data; Acquire patient gait position representation data, and index pre-set patient gait position standard data based on the patient gait position representation data; A data retrieval unit, used to retrieve pre-set standard data of patient gait position; Obtaining the indexed pre-set patient gait position standard data, and retrieving the patient gait position standard data; A comparison and analysis unit, used for comparing and analyzing the patient's gait position representation data; Acquire patient gait position representation data and patient gait position standard data, compare and analyze the patient gait position representation data based on the patient gait position standard data, and determine the patient monitoring and positioning analysis results based on the patient gait position; In the case where the patient's gait position representation data is within the patient's gait position standard data range, the patient monitoring and positioning analysis result is that the patient monitoring and positioning are normal; In the case where the patient's gait position characterization data is not within the patient's gait position standard data range, the patient monitoring and positioning analysis result is that the patient monitoring and positioning is abnormal, and the patient needs to be monitored and positioned early warning.

7. A monitoring and positioning system for a 5G smart LED wristband in a hospital according to claim 6, characterized in that: The early warning control module includes: The early warning management unit is used to monitor, locate and warn patients, and to manage patients intelligently. It can remotely transmit early warning information to the monitoring terminal, monitor patient position changes and locate patients in real time to facilitate finding patients.