Intelligent management method and device for home care and storage medium

Abstract

The invention discloses a home care intelligent management method and device and a storage medium, and relates to the technical field of intelligent home care. Comprising the steps of 1, selecting nursing personnel according to home nursing requirements of a nursed person, 2, carrying out feature calculation on physiological data, and 3, processing according to a result obtained through analysis. The problems that data depends on manual recording and is easy to lose, and long-term trend analysis is difficult to form in home care are solved; based on the intelligent nursing platform, personalized matching of nursing personnel is realized, and specialty and pertinence are ensured; by combining intelligent equipment and physiological data analysis, health indexes such as blood pressure, heart rate and blood oxygen are monitored in real time, the rehabilitation progress is predicted, and the nursing strategy is dynamically adjusted; intervention measures are optimized through an exponential smoothing method and a Markov model, and the nursing precision is improved; and the intelligent early warning system automatically adjusts monitoring and equipment parameters according to the health fluctuation level.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent home care, and in particular to an intelligent home care management method, device and storage medium. Background Art

[0002] With the aging of the population and the increase in the number of patients with chronic diseases, the demand for home care is growing. Traditional home care mainly relies on family members and caregivers, and there are problems such as insufficient nursing professionalism, low nursing efficiency, and lack of real-time monitoring of health data. Intelligent management methods use the Internet of Things, artificial intelligence and big data technologies to achieve all-weather health monitoring, remote care and intelligent early warning for the elderly or patients with chronic diseases, and improve the quality and safety of care. In addition, the collection and analysis of nursing data relies on manual records, data is easily lost and it is difficult to form effective long-term trend analysis; nursing staff often cannot perceive the fluctuations in the health status of the person being cared for in real time, resulting in delayed intervention measures; at the same time, the intelligence level of nursing equipment is low, and it is unable to automatically adjust the nursing plan according to health data, affecting the nursing effect; therefore, a management method based on intelligent technology is needed to improve the accuracy, real-time and personalization of nursing, so as to improve the quality and efficiency of home care. Summary of the invention

[0003] The present invention provides a home care intelligent management method, device and storage medium, which are used to solve the above-mentioned technical problems.

[0004] The first aspect of the present invention provides a home care intelligent management method, comprising: Step 1: Select a caregiver based on the home care needs of the person being cared for, specifically: Enter the information and needs of the person being cared for; select the nursing staff needs; match the nursing staff, fill in the vector composed of various information according to the needs of the person being cared for, and assign the vector composed of various information to the nursing staff needs vector D1 i1 and nursing staff qualification vector P1 j1 The requirements and qualifications need to be quantified according to the actual situation and the matching degree D (D1 i1 , P1 j1 ); The person being cared for or his / her family selects a suitable caregiver based on the push results; Step 2: Calculate the characteristics of physiological data, specifically: Calculate the fluctuation of physiological data through the coefficient of variation CV to generate stable signals, mild fluctuation signals or severe fluctuation signals; identify the trend of physiological indicators based on sliding windows and first-order difference method to generate deterioration signals or recovery signals; calculate the rehabilitation progress and adjust it according to long-term physiological data changes and nursing needs; Step 3: Perform processing based on the analysis results.

[0005] As a preferred embodiment of the present invention, the calculation obtains the matching degree D (D1 i1 , P1 j1 The specific process of ) is: Generate corresponding demand texts according to the needs of the care recipient and the caregiver, simplify the demand texts to obtain simplified demand texts, and then identify the preset words in the simplified demand texts to match caregivers; assign the preset words in the simplified demand texts of the care recipients as the care recipients’ demand vectors D1 i1 , assign the preset words in the simplified nursing staff demand text to the nursing staff qualification vector P1 j1 ; If there are n1 demand items and m1 nursing staff qualification items; then the nursing staff demand vector: D1 i1 =[D1 11 , D1 21 , …, D1 n1 ], where D1 i1 Represents the i1th requirement; nursing staff qualification vector: P1 j1 =[P1 11 , P1 21 , …, P1 m1 ], where P1 j1 Represents the j1th qualification; assign different weights to each requirement and qualification ω i1 and j1 , the matching degree calculation formula is: Output matching degree D (D1 i1 , P1 j1 ),ω i1 It is the demand item of the person being cared for D1 i1 The weight, υ j1 Is a P1 for nursing staff j1 The weight of .

[0006] As a preferred embodiment of the present invention, the specific process of calculating the fluctuation of physiological data by using the coefficient of variation CV to generate the stable signal, the first fluctuation signal and the second fluctuation signal is as follows: Obtain time series and historical records to perform stability analysis on physiological data fluctuations; calculate data stability based on coefficient of variation CV and mean square error σ; The measured values ​​of a physiological indicator at the most recent N time points are x1, x2, ..., x N , mean and standard deviation calculation: , obtain the coefficient of variation CV, and use the coefficient of variation CV as the stability coefficient of the data; if the stability coefficient is less than the preset stability value one, it is determined that the physiological state is stable, and a stable signal is generated; if the stability coefficient is less than the preset stability value two, and the stability coefficient is greater than or equal to the preset stability value one, it is determined that the state fluctuates greatly, and a fluctuation signal one is generated; if the stability coefficient is greater than or equal to the preset stability value two, it is determined that the state fluctuates violently, and a fluctuation signal two is generated.

[0007] As a preferred embodiment of the present invention, the specific process of identifying the change trend of physiological indicators and generating a deterioration signal or a recovery signal based on a sliding window and a first-order difference method is as follows: Retrieve the sliding window trend analysis and calculate it with the first-order difference method: If there is a window of size W, then the mean value in the window is: ; Combined with the first-order difference to calculate the trend change, the formula is obtained: , X t1 is the original observation value of a certain indicator in the time series at time point t; when Δx t1 Greater than 0 indicates an increase in the index; when Δx t1 Less than 0 indicates a decline in the indicator; eliminate the impact of short-term fluctuations and calculate the cumulative trend change, that is: , T is the analysis window, t is the current time point, and t0 represents the starting time point of the analysis window; based on the cumulative trend change T calculated by the sliding window mean and the first-order difference method, the thresholds δ1 and δ2 are set; if T>δ1, it is an upward trend, and an indicator deterioration signal is generated; if T<-δ2, it is a downward trend, and an indicator recovery signal is generated; if -δ2≤T≤δ1, there is no obvious change in the physiological data.

[0008] As a preferred embodiment of the present invention, the specific process of calculating the rehabilitation progress is: The rehabilitation progress is smoothed by exponential smoothing: , S t is the smoothed rehabilitation progress index, x t is the rehabilitation data at the current moment; is the smoothing result of the previous moment; α is the smoothing factor; The Markov process model is then used to predict the transition of the recovery state; if the recovery state R t A discrete random process with a set of states Y1, Y2, ..., Y m ; Learning from historical data, constructing the state transfer matrix Z, Z ij Indicates that from state Y i Transfer to state Y j probability; get the current recovery status P t And the state transfer matrix Z, predict the recovery state at the next moment: P t+1=P t ×Z;P t is the probability distribution of the current recovery status; P t+1 is the state probability distribution at the next moment.

[0009] Combine the exponential smoothing method with the state transition model to determine the complex trend; generate auxiliary reduction signals and auxiliary enhancement signals; As a preferred embodiment of the present invention, the specific process of processing according to the analysis results is as follows: When receiving a stable signal trigger, continue to monitor physiological data without additional equipment adjustments or reminders; When receiving a fluctuation signal, the sampling frequency is increased from the normal mode to the enhanced mode, and adaptively adjusted according to the historical fluctuation data; and the warning information is fed back to the nursing staff through data visualization or notification; When receiving the second fluctuation signal and the auxiliary reinforcement signal, a high-priority alarm is triggered to notify the nursing staff; the key data of abnormal fluctuations is recorded and uploaded to the nursing management platform, and the nursing staff obtains complete information through the nursing management platform; then, according to the medical records, abnormal types, and historical data of the person being cared for, the best treatment criteria are retrieved from the intelligent nursing database and pushed to the nursing staff; the treatment criteria include whether emergency intervention is needed, equipment auxiliary adjustment methods, and observation and follow-up strategies; When the sensor detects that the person being cared for is on the smart nursing bed, the bed surface angle is automatically adjusted to the optimal buffer position, and the bed surface angle adjustment mechanism is driven by a stepper motor to achieve smooth adjustment; When the sensor detects that the person being cared for is in the smart wheelchair, the pressure sensor detects the pressure distribution between the patient's body and the seat, and the posture sensor monitors the tilt angle and dynamic balance of the patient's upper body. The control command is sent to the drive system of the smart seat according to the optimal angle, and the stepper motor drives the rear support mechanism of the seat to smoothly adjust the tilt angle; and the inflatable support airbag or electric back support of the smart backrest is adjusted to increase the support force of the corresponding area; When receiving the trigger of the indicator deterioration signal, the indicator is confirmed. If the indicator is confirmed to have deteriorated, a high-priority warning is immediately triggered, and a red warning signal is displayed, indicating the patient's serious indicator data; internal message push or mobile application push notification; notify the doctor to conduct examination and treatment; When receiving the indicator recovery signal and the auxiliary reduction signal, the recovery trend of the indicator will be monitored in real time, the changes in the health indicators will be recorded, and then it will be determined whether the health indicators continue to recover stably. If the recovery trend continues, it is in a recovery state; if the smart bed is used to support the patient, the inclination of the bed will be gradually reduced according to the recovery trend.

[0010] A second aspect of the present invention provides a home care intelligent management device, the home care intelligent management device comprising a memory and a processor, the memory storing computer-readable instructions, the computer-readable instructions being executed by the processor, causing the processor to execute the steps of the home care intelligent management method in the above-mentioned embodiments; A third aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned intelligent home care management method.

[0011] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects: 1. The present invention has a highly personalized matching capability, and can accurately select suitable caregivers according to the health status, life needs and psychological state of the person being cared for; through the intelligent nursing platform, the system can automatically analyze the matching degree between the demand vector and the nursing staff's qualification vector, improve the professionalism and pertinence of nursing services, ensure that nursing staff have the corresponding skills and experience, and improve the quality and satisfaction of nursing.

[0012] 2. The present invention realizes real-time health monitoring and intelligent management through intelligent devices and physiological data analysis; the present invention can perform feature extraction, stability analysis and abnormal trend detection on physiological data such as blood pressure, heart rate, blood oxygen, etc., to ensure timely identification of health fluctuations; combined with the exponential smoothing method and the Markov state transition model, the present invention predicts rehabilitation progress and automatically adjusts nursing strategies, such as increasing or decreasing the intervention intensity of nursing equipment, thereby optimizing rehabilitation effects and improving nursing efficiency.

[0013] 3. The present invention improves nursing safety and response speed through intelligent early warning and automated nursing intervention. For different levels of health fluctuations, corresponding measures can be taken, such as enhanced monitoring, triggering early warning or adjusting smart device parameters to ensure that nursing staff can intervene in time. When the recovery trend is positive, equipment intervention is reduced, autonomous recovery is encouraged, and more scientific, intelligent and efficient nursing management is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0015] Figure 1 It is a diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0016] Embodiments of the present invention provide a method, device and storage medium for intelligent management of home care. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0017] For ease of understanding, the specific process of the embodiment of the present invention is described below. See also Figure 1 In an embodiment of the present invention, a first aspect of the present invention provides an embodiment of a home care intelligent management method, including: Step 1: Select a caregiver based on the home care needs of the person being cared for; specifically: The information and needs of the care recipient are entered through the intelligent care platform: first, the care recipient or his / her supervisor fills in the basic information of the care recipient, including name, age, gender, health status, etc.; obtains the care needs of the care recipient based on his / her health status, including filling out a detailed health questionnaire or directly conducting remote consultation with a care expert; records the patient's daily life needs, such as whether he / she needs help with eating, washing, dressing, going to the toilet, etc.; analyzes whether the patient needs psychological support, whether he / she has emotional problems, loneliness, depression, etc., and determines whether psychological counseling or companionship is needed.

[0018] Selection of nursing staff needs: Choose the appropriate level of care based on the health needs of the person being cared for, whether professional medical care, rehabilitation care, daily life care or hospice care is needed; then choose the specific skill requirements of the nursing staff based on the specific needs of the person being cared for; whether the nursing staff is required to have experience in Alzheimer's disease care, postoperative rehabilitation care, or whether they need to have professional psychological counseling skills; and choose the nursing staff's working hours, whether it is full-time care, part-time care, or regular care several times a week; if the person being cared for has language communication barriers, choose a nursing staff who can communicate effectively.

[0019] Generate corresponding demand texts according to the needs of the care recipient and the caregiver, simplify the demand texts to obtain simplified demand texts, and then identify the preset words in the simplified demand texts to match caregivers; assign the preset words in the simplified demand texts of the care recipients as the care recipients’ demand vectors D1 i1 , assign the preset words in the simplified nursing staff demand text to the nursing staff qualification vector P1 j1 ; If there are n1 demand items and m1 nursing staff qualification items; then the demand vector of the person being cared for is: D1 i1 =[D1 11 , D1 21 , …, D1 n1 ], where D1 i1 Represents the i1th requirement (e.g., whether fall monitoring is needed, whether psychological counseling is needed, etc.); Nursing staff qualification vector: P1 j1 =[P1 11 , P1 21 , …, P1 m1 ], where P1 j1 Represents the j1st qualification.

[0020] The matching degree is calculated by calculating the care recipient's demand vector D1 i1 and nursing staff qualification vector P1 j1 In practical applications, some requirements and qualifications are more important than others, so different weights are given to each requirement and qualification. i1 and j1 , the weighted matching degree calculation formula is: Output matching degree D (D1 i1 , P1 j1 ), where ω i1 It is the demand item of the person being cared for D1 i1 The weight, υ j1 Is a P1 for nursing staff j1 The weight is determined by historical data analysis and other methods; according to the calculated matching degree D (D1 i1 , P1 j1 ), all qualified caregivers are sorted; caregivers with high matching degree are recommended to the person being cared for; if there are multiple qualified caregivers, further screening is carried out based on other factors, and finally the most suitable caregiver is selected.

[0021] The person being cared for or his / her family can select a suitable caregiver based on the push results. If there are any questions, they can communicate with the caregiver online or have a video interview through the platform. After confirming the caregiver, the characteristics are calculated based on the specific situation of the person being cared for. Step 2: Obtain physiological data for feature calculation, specifically: Obtain time series and historical records to perform stability analysis on physiological data fluctuations; calculate data stability based on coefficient of variation CV and mean square error σ; The measured values ​​of a physiological indicator (such as heart rate) at the last N time points are x1, x2, ..., x N , and its mean and standard deviation are calculated as: , obtain the coefficient of variation CV, and use the coefficient of variation CV as the stability coefficient of the data; if the stability coefficient is less than the preset stability value 1 (such as 5%), the physiological state is considered to be stable and a stable signal is generated; if the stability coefficient is less than the preset stability value 2 (such as 5%), and the stability coefficient is greater than or equal to the preset stability value 1 (such as 5% to 15%), the state fluctuates greatly, and a fluctuation signal 1 is generated (the preset stability value 1 is less than the preset stability value 2); if the stability coefficient is greater than or equal to the preset stability value 2 (such as above 15%), it is considered that the state fluctuates violently, and a fluctuation signal 2 is generated.

[0022] Perform long-term abnormal trend analysis on physiological data; retrieve sliding window trend analysis and calculate with first-order difference method: if there is a window with size W, then the mean value in the window is: ; Combined with the first-order difference to calculate the trend change, the formula is obtained: , X t1 is the original observation value of a certain indicator in the time series at time point t; t1 Greater than 0 indicates an increase in the index (e.g., increased blood pressure); when Δx t1 Less than 0 indicates a decrease in the indicator (for example, a decrease in heart rate); in order to eliminate the impact of short-term fluctuations, the cumulative trend change is calculated, that is: , where T is the analysis window (such as the past 1 hour or 24 hours), t is the current time point, and t0 represents the starting time point of the analysis window; based on the cumulative trend change T calculated by the sliding window mean and the first-order difference method, the thresholds δ1 and δ2 are set to judge the trend change; if T>δ1, it is an upward trend (indicator deterioration), and an indicator deterioration signal is generated; if T<-δ2, it is a downward trend (indicator recovery or abnormal decline), and an indicator recovery signal is generated; if -δ2≤T≤δ1, it means that there is no obvious change in the physiological data.

[0023] Adjust the rehabilitation progress calculation according to long-term physiological data changes and nursing needs; Exponential smoothing is used to smooth the rehabilitation progress, eliminate sudden fluctuations, and highlight long-term trends. It is mainly used for tracking and analyzing chronic disease recovery and long-term rehabilitation plans. The formula for exponential smoothing is: , S t is the smoothed rehabilitation progress index, x tThe current recovery data (such as blood oxygen recovery level, body temperature recovery, etc.); is the smoothing result of the previous moment; α is the smoothing factor, which usually ranges from 0.1 to 0.3 and controls the influence of new and old data on the smoothing result.

[0024] The Markov process model is then used to predict the transition of rehabilitation status; the future rehabilitation progress is predicted by discrete modeling of the rehabilitation process; if the rehabilitation status R t A discrete random process whose possible states are Y1, Y2, ..., Y m , each state represents a different recovery progress; for example: Y1: severe state; Y2: moderate recovery state; Y3: fully recovered state; by learning from historical data, construct the state transfer matrix Z, where Z ij Indicates that from state Y i Transfer to state Y j probability; get the current recovery status P t And the state transfer matrix Z, predict the recovery state at the next moment: P t+1 =P t ×Z; where P t is the probability distribution of the current recovery state, indicating the probability of each recovery state; P t+1 is the state probability distribution at the next moment.

[0025] The exponential smoothing method and the state transition model are combined to determine the complex trend; if the smoothed rehabilitation index S t If the blood oxygen level rises continuously, it indicates that the recovery is progressing well; an auxiliary reduction signal is generated; if the blood oxygen recovery level S t Continuously rise, generate auxiliary reduction signal; if the smoothed rehabilitation index S t If the recovery state R t A significant decline has occurred, meaning that the patient may have entered a more severe stage of recovery, generating an auxiliary reinforcement signal.

[0026] Step 3: Process according to the analysis results, specifically: When receiving a stable signal trigger, the physiological data continues to be monitored without additional equipment adjustment or reminders; when receiving a fluctuation signal, the sampling frequency is increased from normal mode to enhanced mode, that is: enhanced mode = K1 × normal mode, where k1>1 is the frequency enhancement factor, which is adaptively adjusted according to historical fluctuation data; if the fluctuation continues to increase, the acquisition frequency is adjusted in stages to form a dynamic adjustment strategy; including: mild fluctuation (enhanced mode = 1.5 × normal mode), moderate fluctuation (enhanced mode = 2.0 × normal mode) and high fluctuation (enhanced mode = 3.0 × normal mode); and adaptive Kalman filtering (AKF) or wavelet denoising algorithm is used to reduce noise interference and improve monitoring accuracy; if the signal amplitude increases, the sensor range is automatically adjusted to prevent distortion caused by over-range; if the fluctuation signal continues to exceed the preset time threshold (such as 10s), a mild warning is triggered; the warning information is fed back to the nursing staff through data visualization or notification (SMS / email).

[0027] When receiving the second fluctuation signal and the auxiliary reinforcement signal, a high-priority alarm is triggered, and the nursing staff is notified through a variety of notification methods, including mobile phone APP push, SMS, buzzer, etc.; and the key data of the abnormal fluctuation (time, amplitude, duration, related physiological parameters) is recorded and uploaded to the nursing management platform. The nursing staff obtains complete information through the nursing management platform; then, based on the medical history, abnormal type, and historical data of the person being cared for, the best processing criteria are retrieved from the intelligent nursing database and pushed to the nursing staff; the processing criteria include whether emergency intervention is needed (such as drugs, manual operation), equipment-assisted adjustment methods, and observation and follow-up strategies.

[0028] The best treatment criteria are retrieved from the intelligent care database, and the fluctuation type is determined by the KNN classifier or decision tree model based on machine learning (set K (such as 3 or 5), and select the Euclidean distance to calculate the similarity between samples: Get the straight-line distance d(A, B) in n-dimensional space, where A i and B i are the values ​​of data points A and B on the i-th feature respectively; find the K nearest known samples for the new data point, and use majority voting to determine the fluctuation type), including mild discomfort fluctuations (instability caused by body position adjustment), moderate abnormal fluctuations (such as muscle spasms, tremors) and severe risk fluctuations (such as epileptic seizures, loss of autonomous control). Then use knowledge graph matching technology: [Elevated heart rate] - belongs to - > [Mild fluctuation] [Center of gravity shift>10cm]——belongs to——>[Severe fluctuation] [Mild Fluctuation]——Applicable——>[Increase the support strength of the nursing bed] [Severe Fluctuation] - Applicable - > [Notify Nursing Staff] Quickly locate the treatment plan that meets the current fluctuation situation to obtain the best treatment criteria.

[0029] The device assists in adjusting the position. When the sensor detects that the person being cared for is on the smart nursing bed, the angle of the bed surface is automatically adjusted to the optimal buffer position (the sensor obtains the current body movement data of the person being cared for (including acceleration, pressure distribution, electromyographic signals, etc.) for analysis, based on the body pressure distribution (formula: Get the pressure distribution balance P, F i is the force detected by the i-th pressure sensor; F p is the average force measured by all sensors; n is the total number of pressure sensors distributed on the nursing bed) to calculate the contact pressure distribution between the patient's body and the bed surface. If the pressure is concentrated in a certain area, the bed surface is tilted to disperse the force until the bed surface is evenly distributed. Then based on inertia analysis (by formula: Get the center of gravity offset C, where (X c , Y c ) is the patient's current center of gravity coordinate, which is calculated by the body pressure distribution sensor + IMU inertial sensor, (X s , Y s ) is the center coordinate of the chair (the center of gravity position in the standard sitting posture)) to calculate the center of gravity offset of the current patient. If the center of gravity is biased to one side, the angle of the bed surface is adjusted to balance the stability; then the balance stability is calculated (by combining the center of gravity offset, pressure distribution balance and posture change rate: Get, C max is the maximum allowable center of gravity offset, V is the attitude change rate, V max The maximum allowable posture change rate, ω1, ω2 and ω3 are all weight coefficients, satisfying ω1+ω2+ω3=1), until the corresponding balance stability of the adjusted bed angle belongs to the preset range; the tilt angle θ is set by the angle calculation model, through the formula: Where P is the body pressure distribution weight; C is the center of gravity offset; H is the historical best angle; θ max是 Maximum tilt angle, C max is the maximum allowable center of gravity offset, H max is the maximum height; the decision tree predicts the most suitable angle at the moment), w1, w2 and w3 are all weight coefficients; the bed angle adjustment mechanism is driven by a stepper motor to achieve smooth adjustment to avoid secondary impact on the patient; it should also be noted that during the adjustment process, the patient's status is monitored in real time. If there is an abnormal situation (such as abnormal tension of the patient, deterioration of pressure distribution), the adjustment is terminated and an alarm is issued.

[0030] When the sensor detects that the person being cared for is in the smart wheelchair, the pressure sensor detects the pressure distribution of the patient's body in contact with the seat, and the posture sensor monitors the inclination angle and dynamic balance of the patient's upper body. If the center of gravity offset C exceeds the preset safety range (such as more than 5 cm), there is a risk of tipping over; if the pressure distribution balance P is calculated, if the pressure on one side is greater than G times that on the other side, the seat support force needs to be adjusted; if the posture stability index S is calculated, if S is lower than the preset threshold (such as 0.7), it is determined that the patient has an obvious imbalance trend; similarly, the input variables body pressure distribution P, center of gravity offset C, imbalance risk index S, and previous adjustment record H are input; similarly, the optimal inclination angle θ2=f(P, C, S, H) is calculated, and the optimal angle is determined by regression analysis; and according to the optimal angle, the control command is sent to the drive system of the smart seat, and the stepper motor drives the rear support mechanism of the seat to smoothly adjust the inclination angle; and the inflatable support airbag or electric back support of the smart backrest is adjusted to increase the support force of the corresponding area to prevent tipping over.

[0031] When receiving the trigger signal of indicator deterioration, the indicator is confirmed. If the indicator is confirmed to have deteriorated, a high-priority warning is triggered immediately, displaying a red warning signal and indicating the serious indicator data of the patient's condition; notification is given through internal message push or mobile application push; the doctor is notified through email, SMS, phone call, etc. for examination and treatment.

[0032] When receiving the indicator recovery signal and the auxiliary reduction signal, the recovery trend of the indicator will be monitored in real time, the changes in the health indicator will be recorded, and then the above calculation process will be used to determine whether the health indicator continues to recover stably. If the recovery trend continues, it is in a recovery state; if the smart bed is used to support the patient, the inclination of the bed will be gradually reduced according to the recovery trend; for example, from the initial 30° to 20°, 10°, and finally restored to a horizontal state (0°). It is adjusted in real time according to the patient's health status (for example, each health status corresponds to an inclination. If it is not appropriate, the bed inclination and support strength are dynamically adjusted based on data feedback). Whenever the health indicator recovers, the bed will be further adjusted until it is fully restored to a horizontal state to avoid excessive support interfering with the patient's autonomous activities while ensuring the safety of the patient.

[0033] A second aspect of the present invention provides a home care intelligent management device, the home care intelligent management device comprising a memory and a processor, the memory storing computer-readable instructions, the computer-readable instructions being executed by the processor, causing the processor to execute the steps of the home care intelligent management method in the above-mentioned embodiments; The third aspect of the present invention provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the intelligent home care management method.

[0034] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0035] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0036] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A home care intelligent management method, characterized in that: include: Step 1: Select a caregiver based on the home care needs of the person being cared for, specifically: Enter the information and needs of the person being cared for; select the nursing staff needs; match the nursing staff, fill in the vector composed of various information according to the needs of the person being cared for, and assign the vector composed of various information to the nursing staff needs vector D1 i1 and nursing staff qualification vector P1 j1 The requirements and qualifications need to be quantified according to the actual situation and the matching degree D (D1 i1 , P1 j1 ); The person being cared for or his / her family selects a caregiver based on the push results; Step 2: Calculate the characteristics of physiological data, specifically: Calculate the fluctuation of physiological data through the coefficient of variation CV to generate stable signals, mild fluctuation signals or severe fluctuation signals; identify the trend of physiological indicators based on sliding windows and first-order difference method to generate deterioration signals or recovery signals; calculate the rehabilitation progress and adjust it according to long-term physiological data changes and nursing needs; Step 3: Perform processing based on the analysis results.

2. A home care intelligent management method according to claim 1, characterized in that: The specific process of calculating the matching degree is as follows: Generate corresponding demand texts according to the needs of the care recipient and the caregiver, simplify the demand texts to obtain simplified demand texts, and then identify the preset words in the simplified demand texts to match caregivers; assign the preset words in the simplified demand texts of the care recipients as the care recipients’ demand vectors D1 i1 , assign the preset words in the simplified nursing staff demand text to the nursing staff qualification vector P1 j1 ; If there are n1 demand items and m1 nursing staff qualification items; then the nursing staff demand vector: D1 i1 =[D1 11 , D1 21 , …, D1 n1 ], where D1 i1 Represents the i1th requirement; nursing staff qualification vector: P1 j1 =[P1 11 , P1 21 , …, P1 m1 ], where P1 j1 Represents the j1th qualification; assign different weights to each requirement and qualification ω i1 and j1 , the matching degree calculation formula is: Output matching degree D (D1 i1 , P1 j1 ),ω i1 It is the demand item of the person being cared for D1 i1 The weight, υ j1 Is a P1 for nursing staff j1 The weight of .

3. A home care intelligent management method according to claim 1, characterized in that: The specific process of calculating the fluctuation of physiological data by using the coefficient of variation CV to generate stable signal, fluctuation signal 1 and fluctuation signal 2 is as follows: Obtain time series and historical records to perform stability analysis on physiological data fluctuations; calculate data stability based on coefficient of variation CV and mean square error σ; The measured values ​​of a physiological indicator at N time points are x1, x2, ..., x N ,pass Obtaining a coefficient of variation CV, and using the coefficient of variation CV as a stability coefficient of the data; if the stability coefficient is less than a preset stability value of one, it is determined that the physiological state is stable, and a stability signal is generated; If the stability coefficient is less than the preset stability value 2, and the stability coefficient is greater than or equal to the preset stability value 1, it is determined that the state fluctuates greatly and a fluctuation signal 1 is generated; if the stability coefficient is greater than or equal to the preset stability value 2, it is determined that the state fluctuates violently and a fluctuation signal 2 is generated.

4. A home care intelligent management method according to claim 3, characterized in that: The specific process of identifying the change trend of physiological indicators and generating a deterioration signal or a recovery signal based on the sliding window and the first-order difference method is as follows: Retrieve the sliding window trend analysis and calculate it with the first-order difference method: If there is a window of size W, then the mean value in the window is: ; Combined with the first-order difference to calculate the trend change, the formula is obtained: , X t1 is the original observation value of a certain indicator in the time series at time point t; when Δx t1 Greater than 0 indicates an increase in the index; when Δx t1 Less than 0 indicates a decline in the indicator; eliminate the impact of short-term fluctuations and calculate the cumulative trend change, that is: , T is the analysis window, t is the current time point, and t0 represents the starting time point of the analysis window; Based on the cumulative trend change T calculated by the sliding window mean and the first-order difference method, set the thresholds δ1 and δ2; if T>δ1, it is an upward trend and an indicator deterioration signal is generated; If T<-δ2, it is a downward trend and an indicator recovery signal is generated; if -δ2≤T≤δ1, there is no obvious change in the physiological data.

5. A home care intelligent management method according to claim 4, characterized in that: The specific process of calculating the rehabilitation progress is as follows: The rehabilitation progress is smoothed by exponential smoothing: , S t is the smoothed rehabilitation progress index, x t is the rehabilitation data at the current moment; is the smoothing result of the previous moment; α is the smoothing factor; The Markov process model is then used to predict the transition of the recovery state; if the recovery state R t A discrete random process with a set of states Y1, Y2, ..., Y m ; Learning from historical data, constructing the state transfer matrix Z, Z ij Indicates that from state Y i Transfer to state Y j probability; get the current recovery status P t And the state transfer matrix Z, predict the recovery state at the next moment: P t+1 =P t ×Z;P t is the probability distribution of the current recovery status; P t+1 is the state probability distribution at the next moment; Combine the exponential smoothing method with the state transition model to determine the complex trend; generate auxiliary reduction signals and auxiliary enhancement signals.

6. A home care intelligent management method according to claim 5, characterized in that: The specific process of processing the results obtained according to the analysis is as follows: When receiving a stable signal trigger, continue to monitor physiological data; When receiving a fluctuation signal, the sampling frequency is increased from the normal mode to the enhanced mode, and adaptively adjusted according to the historical fluctuation data; And feedback the warning information to nursing staff through data visualization or notification; When receiving the second fluctuation signal and the auxiliary reinforcement signal, a high priority alarm is triggered to notify the nursing staff; the key data of abnormal fluctuations is recorded and uploaded to the nursing management platform, and the nursing staff obtains complete information through the nursing management platform; then, according to the medical records, abnormal types, and historical data of the person being cared for, the best processing criteria are retrieved from the intelligent nursing database and pushed to the nursing staff; Management criteria include whether urgent intervention is needed, device-assisted adjustment methods, and observation and follow-up strategies; When the sensor detects that the person being cared for is on the smart nursing bed, the bed surface angle is automatically adjusted to the optimal buffer position, and the bed surface angle adjustment mechanism is driven by a stepper motor to achieve smooth adjustment; When the sensor detects that the person being cared for is in the smart wheelchair, the pressure sensor detects the pressure distribution of the patient's body in contact with the seat, and the posture sensor monitors the tilt angle and dynamic balance of the patient's upper body. The control command is sent to the drive system of the smart seat according to the optimal angle, and the stepper motor drives the rear support mechanism of the seat to smoothly adjust the tilt angle; and adjusts the inflatable support airbag or electric back support of the smart backrest to increase the support force of the corresponding area; When receiving the trigger of the indicator deterioration signal, the indicator is confirmed. If the indicator is confirmed to have deteriorated, a high-priority warning is immediately triggered, and a red warning signal is displayed, indicating the patient's serious indicator data; notifications are sent through internal message push or mobile application push; Notify the doctor for examination and treatment; When receiving the indicator recovery signal and the auxiliary reduction signal, the recovery trend of the indicator will be monitored in real time, the changes in the health indicators will be recorded, and then it will be determined whether the health indicators continue to recover stably. If the recovery trend continues, it is in a recovery state; if the smart bed is used to support the patient, the inclination of the bed will be gradually reduced according to the recovery trend.

7. An intelligent home care management device, characterized in that: The home care intelligent management device includes: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the home care intelligent management device executes the home care intelligent management method as described in any one of claims 1-6.

8. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the intelligent home care management method as described in any one of claims 1-6 is implemented.

Images