CN119993566A

CN119993566A - Rehabilitation nursing remote monitoring transmission system

Info

Publication number: CN119993566A
Application number: CN202510476214.4A
Authority: CN
Inventors: 韩梅; 胡晓红; 崔延超; 王雪; 龚梦青; 胡佳惠
Original assignee: First Affiliated Hospital of Xian Jiaotong University
Current assignee: First Affiliated Hospital of Xian Jiaotong University
Priority date: 2025-04-16
Filing date: 2025-04-16
Publication date: 2025-05-13

Abstract

The invention discloses a rehabilitation nursing remote monitoring transmission system which comprises the following steps of acquiring a plurality of physiological parameters of a user, judging fluctuation conditions of the physiological parameters, automatically sampling data when parameter changes exceed a set threshold value, identifying state mutation by utilizing an analysis method, dynamically adjusting a sampling strategy according to an identification result by the system, optimizing resource utilization efficiency, synchronizing processed data and an evaluation result between terminal equipment and a remote platform by the system through a two-way communication mechanism, realizing continuous monitoring and stage judgment of a rehabilitation process, and assisting nursing intervention. The invention has the capabilities of automatic analysis and feedback regulation and control, and is suitable for remote rehabilitation management scenes.

Description

Rehabilitation nursing remote monitoring transmission system

Technical Field

The invention relates to the technical field of remote rehabilitation management, in particular to a rehabilitation nursing remote monitoring transmission system.

Background

With the increasing demand for telemedicine and intelligent rehabilitation, higher requirements are put on continuous monitoring and personalized management of physiological states in rehabilitation care scenes. Especially in applications such as home rehabilitation, postoperative rehabilitation and chronic intervention, the physiological state of the patient often presents dynamic changes, and how to realize an efficient, accurate and remotely controllable monitoring mechanism becomes a key technical challenge in the remote Kang Fuguan.

In the prior art, a periodic sampling mode is mostly adopted in a remote rehabilitation monitoring system, and health assessment is carried out by combining static threshold judgment. This type of approach suffers from significant drawbacks in several respects:

1. the existing method is mostly based on fixed frequency sampling, and the sampling frequency cannot be dynamically adjusted according to actual fluctuation of the physiological state, so that redundant data are more, and energy efficiency is low.

2. Abnormal recognition hysteresis most methods only judge mutation events based on statistical analysis or fixed rules, and are difficult to realize real-time capturing and early recognition of rehabilitation state changes.

3. The feedback mechanism is lacking, the current system lacks closed loop adjustment capability between the monitoring result and the sampling strategy, and the sampling triggering condition cannot be reversely optimized according to the monitoring analysis result.

4. The rehabilitation state is expressed roughly, the traditional method is used for judging the health state through single-point index threshold values, modeling and grading expression of periodic characteristics of the rehabilitation process are lacked, and individual management requirements are difficult to meet.

5. The remote coordination capability is limited, the data synchronization mechanism between the edge end and the remote management platform of the existing system is simpler, and the bidirectional communication support of model results, early warning information and control parameters is lacking.

Therefore, how to provide a remote monitoring and transmitting system for rehabilitation nursing is a problem to be solved by those skilled in the art.

Disclosure of Invention

The invention adopts an event-driven dynamic sampling mechanism, a Bayesian online variable point detection method, a self-adaptive sampling regulation strategy and a rehabilitation state segmentation recognition model to analyze physiological parameters of a user in real time and evaluate the state of multiple stages, so that the whole process from data acquisition, variable point recognition and sampling control to stage-level output is described in detail, the rapid recognition and response of key health events are realized, and the system has the advantages of high monitoring sensitivity, low data redundancy, strong feedback regulation capability and good remote collaborative management capability.

According to the embodiment of the invention, the rehabilitation and nursing remote monitoring transmission system comprises the following steps:

s1, a data acquisition module acquires multisource physiological parameters of a user, constructs a sliding time window, calculates dynamic variation amplitude of each multisource physiological parameter, and performs data sampling when the dynamic variation amplitude of any multisource physiological parameter exceeds a set sampling trigger threshold value to generate a sampling data sequence;

S2, a data preprocessing module performs preprocessing on the sampled data sequence, wherein the preprocessing comprises filtering noise reduction and time alignment, and standardized time sequence data is generated;

S3, a variable point detection module performs recursive analysis by adopting a Bayes online variable point detection method based on standardized time sequence data to obtain variable point posterior probability of each time point, and recognizes the corresponding time point as a state mutation time point when the variable point posterior probability is larger than a set probability judgment threshold;

s4, a sampling regulation and control module adjusts sampling triggering conditions according to comparison results of the variable point posterior probability at any time point and the set probability judgment threshold value, and realizes closed-loop self-adaptive regulation of sampling control;

S5, a rehabilitation state identification module segments the standardized time sequence data based on the identified state mutation time points, generates a rehabilitation stage interval and outputs a rehabilitation stage grade corresponding to each state mutation time point;

S6, the edge cooperation and transmission module synchronously transmits the standardized time sequence data, the variable point posterior probability and the rehabilitation stage level between the edge end and the remote server through a two-way communication mechanism, wherein the edge end is local terminal equipment for executing data acquisition and analysis tasks, the remote server is a rehabilitation monitoring platform for centralized storage and control, and the communication mechanism comprises uploading of monitoring data and issuing of remote control instructions so as to realize remote monitoring and dynamic management of a rehabilitation process.

Optionally, the S1 specifically includes:

S11, a physiological parameter acquisition unit acquires multi-source physiological parameters of a user to form a multi-channel original physiological parameter which is synchronously recorded;

S12, a sliding window processing unit constructs the original physiological parameters corresponding to each channel into the length of The sliding time window is updated according to a fixed step length to form a plurality of sliding time window sections with continuous time;

S13, a change amplitude extraction unit takes the original physiological parameters in each sliding time window as input to calculate the corresponding dynamic change amplitude, wherein the dynamic change amplitude The definition is as follows:

;

Wherein, Represent the firstThe original set of physiological parameters within the sliding time window,Is shown in the firstLast original physiological parameters in the sliding time window;

S14, sampling triggering judging unit, which calculates the obtained dynamic change amplitude With a set sampling trigger thresholdComparing if it meetsTriggering a data sampling event;

S15, the sampling data construction unit extracts and packages all original physiological parameters in the sliding window corresponding to the triggering event into a group of sampling fragments, and the sampling fragments are combined in time sequence to form a sampling data sequence.

Optionally, the step S14 specifically includes:

s141, a threshold configuration unit for setting a sampling trigger threshold The sampling trigger threshold is a positive real number and is used for comparing with the dynamic change amplitude to judge whether sampling is triggered or not;

S142, a single-channel comparison unit for receiving dynamic variation amplitude And corresponding each channelAnd (3) withComparing, if the condition is satisfiedMarking the current sliding time window of the channel as a candidate trigger window;

s143, a multi-channel fusion judging unit gathers candidate trigger windows of all channels, and generates a sampling trigger signal if at least one channel meets comparison conditions;

S144, the trigger control unit extracts the data index corresponding to the current sliding time window based on the sampling trigger signal, and generates a sampling event instruction for driving the sampling data construction unit to execute sampling data sequence generation operation.

Optionally, the step S3 specifically includes:

s31, a variable point probability calculation unit receives the standardized time sequence data, performs recursive analysis based on a Bayesian online variable point detection method, and calculates a variable point posterior probability value of which each time point is a state mutation point, wherein the variable point posterior probability value is defined as:

;

Wherein, Indicating a point in timeIs provided with a normalized time series data subsequence,Indicating a point in timeThe length of the interval between the two mutation points and the last state mutation point,Indicating a point in timeAs a point of a new state mutation,Represented in normalized time series data subsequenceTime point under given conditionsIs the posterior distribution of the state mutation points,For the point in timeA corresponding change point posterior probability value;

s32, a variable point probability sequence construction unit receives variable point posterior probability values of all time points Sequentially arranged in time sequence to generate variable point probability sequenceWhereinA termination time point index indicating time series data;

S33, a variable point judging unit for setting a probability judging threshold value Posterior probability values for each change point in the sequence of change point probabilitiesAnd (3) withComparing, judging the condition as, wherein,To set the probability judgment threshold, the range isConfidence level for defining status mutation point determination, determining a time point when a condition is satisfiedIs a candidate state mutation point;

s34, a state mutation output unit receives the candidate state mutation points, performs label confirmation and outputs a state mutation time point set WhereinRepresents the point in time identified as a state mutation.

Optionally, the step S31 specifically includes:

s311, an observation sequence construction unit receives the standardized time sequence data at each time point Constructing corresponding time sequence data subsequences;

S312, prior modeling unit, defining state variablesIndicating a point in timeThe distance between the mutation point and the previous state mutation point is set to be prior distributionThe method comprises the following steps:

;

Wherein, The prior probability of the occurrence of the state mutation point is satisfied,The value of the interval length is given;

S313, an observation model evaluation unit constructs a conditional observation model Assume a normalized temporal data subsequence within a stateless mutation point intervalObeying gaussian distribution, expressed as:

;

Wherein, Representing the mean value of the current segment, performing an autoregressive update based on the data from the start of the segment to the time immediately prior to the current time point,The method comprises the steps of presetting an observation noise variance;

s314, posterior probability calculation unit based on prior distribution Condition observation modelPosterior distribution at the last moment, wherein,For standardizing time series data subsequencesTime point under given conditionsPosterior distribution generated for variable point recursion, and calculating time points in a recursion modeIs a variational posterior probability value.

Optionally, the step S4 specifically includes:

S41, a variable point probability extraction unit receives the variable point probability sequence And extract the time pointCorresponding variational point posterior probability values;

S42, threshold comparison unit for setting probability judgment thresholdWill time pointCorresponding variational point posterior probability valuesAnd probability judgment thresholdComparing to determine whether it meets;

S43, sampling adjusting signal generating unit judges that the signal satisfiesGenerating a sampling adjustment command signalWhereinIndicating that dynamic adjustment of the sampling trigger threshold is required,Indicating that the sampling trigger threshold value does not need to be dynamically adjusted;

S44, triggering a threshold adjusting unit to receive the sampling adjustment instruction signal Dynamically updating sampling trigger thresholdsThe adjustment strategy is:

;

Wherein, The adjustment coefficient is fixed for a positive real number,Indicating a point in timeIs an updated sampling trigger threshold;

S45, feeding back the closed-loop control unit to trigger the updated sampling threshold value The method is applied to setting of trigger conditions in a data acquisition module and used for controlling dynamic adjustment of a trigger threshold value of subsequent sampling so as to realize closed-loop self-adaptive adjustment of sampling control.

Optionally, the step S44 specifically includes:

s441, a threshold receiving unit for receiving a sampling trigger threshold Sampling adjustment command signal, wherein,Indicating whether adjustment is required;

S442, threshold calculating unit for sampling and adjusting command signal When the sampling trigger threshold value is triggered, the sampling trigger threshold value is updated according to a linear decrementing strategy, and an updating formula is as follows:

;

Wherein, Fixing an adjustment coefficient for a positive real number;

s443, a threshold holding unit for sampling and adjusting the command signal When the current sampling trigger threshold is maintained unchanged, namely:

;

S444, a threshold output unit for outputting the adjusted sampling trigger threshold The method is used for setting the trigger conditions in the subsequent data acquisition module and is used as the initial threshold value of the dynamic sampling control of the next period.

Optionally, the step S5 specifically includes:

S51, a mutation point receiving unit receives the state mutation time point set WhereinRepresenting a point in time identified as a status mutation;

S52, stage segmentation unit, according to the state abrupt change time point set Segmenting the standardized time sequence data according to time sequence to construct a plurality of non-overlapping time intervalsWherein each phase interval is defined as:

;

Wherein, ,,Represent the firstThe time intervals corresponding to the individual rehabilitation phases,Representing the total number of recovery phase time intervals;

s53, a stage feature extraction unit for each rehabilitation stage interval The normalized time sequence data in the system calculates statistical characteristic indexes including mean value, variance and trend slope to form a stage characteristic vector;

S54, stage level judging unit based on stage feature vectorIntroducing a stage level mapping functionMapping each stage interval to a rehabilitation stage grade:

;

Wherein, Is the firstStage recovery stage grade.

The beneficial effects of the invention are as follows:

(1) According to the invention, by combining an event-driven dynamic sampling mechanism, a Bayesian online variable point detection method and a self-adaptive sampling control strategy, the efficient monitoring of the multisource physiological parameters of the user in the remote rehabilitation and nursing process is realized, the key physiological state mutation can be timely identified, the sampling strategy is dynamically regulated, the redundant data quantity and the system energy consumption are obviously reduced, and the sensing capability and the response speed to the abnormal state in the rehabilitation process are improved.

(2) According to the invention, by constructing a closed-loop control mechanism taking the variable point posterior probability as a feedback basis, the sampling trigger condition can be automatically updated according to the real-time judgment result of the state mutation, the self-adaptive adjustment of the sampling frequency is realized, the resource utilization efficiency and the system monitoring sensitivity are effectively improved, and the method is particularly suitable for the edge end part environment with limited resources.

(3) According to the invention, the rehabilitation data is subjected to staged segmentation processing based on the mutation point identification result, segmentation characteristics are extracted, and multi-stage identification and evaluation of the rehabilitation process are realized through the rehabilitation stage grade mapping model, so that the system has the automatic discrimination and grade classification capability of the rehabilitation stage, and the individualized rehabilitation management and remote intervention opportunity optimization are supported.

(4) According to the invention, through a bidirectional communication mechanism between the edge equipment and the remote management platform, synchronous transmission of standardized time sequence data, the variable point posterior probability and the rehabilitation stage grade result is realized, the remote monitoring, model control and data interaction capability is provided, and the collaborative intelligent level and remote operability of the rehabilitation nursing system are improved.

Drawings

The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:

fig. 1 is a general architecture diagram of a remote monitoring and transmitting system for rehabilitation nursing according to the present invention.

Detailed Description

The invention will now be described in further detail with reference to the accompanying drawings. The drawings are simplified schematic representations which merely illustrate the basic structure of the invention and therefore show only the structures which are relevant to the invention.

Referring to fig. 1, the rehabilitation and nursing remote monitoring transmission system comprises the following steps:

According to the embodiment, the sampling data sequence is subjected to filtering treatment by adopting a band-pass filtering algorithm, high-frequency noise and low-frequency drift in physiological parameter signals are effectively removed, then time alignment of multi-channel data is completed by adopting an interpolation method and a unified time stamp strategy, synchronism of various physiological parameters on the same time axis is ensured, finally, the treated data is subjected to standardized conversion, parameters of different scales and numerical ranges are comparable, standardized time sequence data in a unified format are generated, stable, synchronous and computable input data sources are provided for subsequent variable point detection and recovery stage identification by the treatment process, and accuracy of model analysis and overall robustness of the system are remarkably improved.

S6, an edge cooperation and transmission module synchronously transmits the standardized time sequence data, the variable point posterior probability and the rehabilitation stage level between an edge end and a remote server by constructing a two-way communication mechanism, wherein the edge end is local terminal equipment for executing data acquisition and analysis tasks, the remote server is a rehabilitation monitoring platform for centralized storage and control, and the communication mechanism comprises uploading of monitoring data and issuing of remote control instructions so as to realize remote monitoring and dynamic management of a rehabilitation process.

According to the embodiment, the data acquisition and analysis module is arranged at the edge, after the standardized processing of the physiological parameters of the user, the state mutation identification and the recovery stage grade judgment are completed at the local terminal in real time, the standardized time sequence data, the variable point posterior probability and the recovery stage grade are synchronously uploaded to the remote recovery monitoring platform by utilizing the constructed bidirectional communication mechanism, and meanwhile, the control parameters and the intervention instructions issued by the platform are received, so that the bidirectional synchronization of the data and the control instructions is realized. The mechanism ensures that the edge equipment is consistent with the remote platform in real time in terms of data state and control logic, thereby not only improving the continuity and responsiveness of remote monitoring, but also realizing the floor application of the efficient rehabilitation management and dynamic adjustment strategy on the basis of edge calculation.

In this embodiment, the S1 specifically includes:

The sliding time window is realized by constructing a continuous time period according to a fixed length for an original physiological parameter sequence and gradually sliding forwards with a set step length. When the method is implemented, the system firstly sets the time length covered by each window, and divides the original multi-channel data according to the time sequence, so that each window contains a plurality of physiological parameter samples in adjacent time periods, and the system moves the window starting position according to the preset step length to generate a plurality of overlapped or non-overlapped window sections, thereby realizing continuous slicing processing of the data stream and facilitating the follow-up analysis of the change trend and the trigger judgment.

S13, a change amplitude extraction unit takes the original physiological parameters in each sliding time window as input to calculate the corresponding dynamic change amplitude, wherein the dynamic change amplitudeThe definition is as follows:

;

According to the embodiment, the physiological parameter acquisition unit is used for acquiring the multi-source physiological signals of the user, each signal channel is dynamically processed based on a sliding time window mechanism, and the change amplitude of the physiological parameters is extracted in each window to serve as a characteristic index of state fluctuation. The system compares the variation amplitude with a preset threshold value, triggers data sampling when a sudden change trend is detected, packages the original data in the period into a complete sampling segment sequence, and constructs standard input for subsequent analysis. The method not only realizes accurate capture of key physiological changes, but also avoids redundant sampling through a window mechanism, improves data processing efficiency and system response sensitivity, and provides basic data guarantee for subsequent change point detection and rehabilitation evaluation.

In this embodiment, the step S14 specifically includes:

According to the embodiment, the intelligent identification of key physiological state fluctuation is realized by setting the sampling trigger threshold and carrying out fusion judgment by combining the dynamic variation amplitude of the multichannel physiological parameters. The system firstly compares the dynamic change amplitude of the current sliding time window with a preset threshold value on each channel to mark candidate trigger windows meeting the conditions, then performs cross-channel summarization in a multi-channel fusion judging unit, and triggers a sampling control unit to generate a sampling event instruction as long as any channel has obvious fluctuation, so that the subsequent data extraction and processing processes are driven. The mechanism effectively improves response sensitivity to abnormal states, avoids resource waste caused by fixed frequency sampling, and enhances real-time performance and adaptability of the system in an actual rehabilitation monitoring scene.

In this embodiment, the step S3 specifically includes:

;

the formula normalizes the time series data subsequence Posterior probabilityVariable point posterior probability valueThe probability value of the new change point is calculated in a recursive manner at each time point by comprehensively considering the relation between the current time sequence data and the historical time sequence data, so that the real-time judgment of the abrupt change moment is realized. The method has the core principle that the prior information is fused with the current time sequence data by using a Bayesian inference framework, the judgment basis of the change point is dynamically updated, the online processing capability is realized, the possibility of state change can be continuously evaluated in the continuous updating process of the data, and a reliable mutation judgment basis is provided for the subsequent stage identification and sampling regulation.

S32, a variable point probability sequence construction unit receives variable point posterior probability values of all time pointsSequentially arranged in time sequence to generate variable point probability sequenceWhereinA termination time point index indicating time series data;

S33, a variable point judging unit for setting a probability judging threshold value Posterior probability values for each change point in the sequence of change point probabilitiesAnd (3) withComparing, judging the condition as, wherein,To set the probability judgment threshold, the range is dimensionlessConfidence level for defining status mutation point determination, determining a time point when a condition is satisfiedIs a candidate state mutation point;

The variable point identification method based on threshold judgment is implemented by the system, after receiving posterior probability values of the corresponding time points, comparing the probability values with a preset judgment threshold one by one, and when the probability value of a certain time point exceeds the threshold, considering that the time point has possibility of state mutation and marking the time point as a candidate variable point. According to the method, the sensitivity of variable point identification is controlled by setting the threshold value, so that the system can effectively filter out misjudgment caused by short-term fluctuation while maintaining the real-time performance, and the accuracy of state mutation identification is improved.

S34, a state mutation output unit receives the candidate state mutation points, performs label confirmation and outputs a state mutation time point setWhereinRepresents the point in time identified as a state mutation.

The embodiment is based on a Bayes online variable point detection method to carry out recursive analysis on standardized time sequence data, and utilizes the state at the previous moment and the current observation value to jointly calculate posterior probability of each time point as a state mutation point, constructs a variable point probability sequence arranged in time sequence, judges the variable point probability of each time point by setting a probability judgment threshold value, identifies potential state mutation points and finally outputs a state mutation time point set. The method can timely detect the abrupt change moment of the physiological state of the user under the condition of limited data volume, improves the abnormal recognition precision and instantaneity in the rehabilitation process, and provides high-reliability abrupt change information support for subsequent rehabilitation state segmentation and sampling strategy self-adaptive adjustment.

In this embodiment, the step S31 specifically includes:

;

The formula adopts a geometric distribution form, the principle is based on the assumption that the mutation event obeys the occurrence of fixed probability on a time axis, the time sparsity and uncertainty of the occurrence of the change point can be effectively expressed, a mathematical foundation is provided for state transition modeling in subsequent Bayesian recurrence, and the modeling capability of the system on sudden state change is enhanced.

S313, an observation model evaluation unit constructs a conditional observation modelAssume a normalized temporal data subsequence within a stateless mutation point intervalObeying gaussian distribution, expressed as:

;

The formula is based on a normal distribution assumption, physiological parameter fluctuation of each non-mutation stage is regarded as random fluctuation near a stable mean value, and the deviation degree of a current observed value relative to historical data can be calculated by using the model, so that a stable probability basis is provided for variable point detection. According to the formula, a reference baseline of physiological state stability is established through the mean value and the variance of data in the fitting stage, so that the data can be timely reflected in posterior probability when the state is obviously deviated, and the sensitivity and the accuracy of variable point identification are improved.

S314, posterior probability calculation unit based on prior distributionCondition observation modelPosterior distribution at the last moment, wherein,For standardizing time series data subsequencesTime point under given conditionsPosterior distribution generated for variable point recursion, and calculating time points in a recursion modeIs a variational posterior probability value.

In this embodiment, based on the normalized time series data sequence, a corresponding observation sub-sequence is first constructed at each time point for subsequent bayesian analysis. On the basis, defining the interval between the current moment and the last variable point represented by a state variable, setting the prior distribution of the state variable as geometric distribution for expressing the time probability structure of the occurrence of mutation, further constructing an observation model, assuming that the data in the continuous non-variable point section obeys Gaussian distribution, performing autoregressive estimation on the average value based on section history data, and constructing a complete likelihood function by combining preset noise variance. And then, inputting the prior distribution, the observation model and the posterior probability at the previous moment into a Bayesian recurrence formula, and calculating the posterior probability of each time point as a variable point in real time. The method realizes the online identification of the state mutation in the rehabilitation process, has higher timeliness and sensitivity, can keep stable judgment capability in the edge environment with limited data quantity, and provides reliable support for rehabilitation early warning and monitoring strategy optimization.

In this embodiment, the S4 specifically includes:

S42, a threshold comparison unit for setting a dimensionless probability judgment threshold valueWill time pointCorresponding variational point posterior probability valuesAnd probability judgment thresholdComparing to determine whether it meets;

And if the condition is met, generating a control signal for adjusting the sampling strategy. In actual operation, the system extracts the current variable point probability value in each monitoring period, compares the current variable point probability value with the threshold value in real time, and generates a control instruction for starting the updating flow of the subsequent sampling trigger threshold value if the state mutation trend is identified to be obvious. The control signal is expressed in a Boolean form, has definite judgment basis and operation condition, ensures that sampling adjustment is effectively triggered only when the monitoring state is suddenly changed, and forms an adjustment mechanism based on state probability driving.

S44, triggering a threshold adjusting unit to receive the sampling adjustment instruction signalDynamically updating sampling trigger thresholdsThe adjustment strategy is:

;

The formula realizes dynamic sampling strategy regulation and control based on state mutation probability, and the principle is that whether the current system state has mutation trend is judged, and then sampling triggering conditions are subjected to descending adjustment, so that the triggering threshold value is automatically reduced when the system detects possible physiological abnormality, the sampling frequency is improved, and the response capability of the system to sudden events is enhanced. The formula embodies a linear feedback mechanism, carries out threshold updating with limited amplitude according to a preset step factor, ensures that the system keeps dynamic balance between sensitivity and resource consumption, and realizes the self-adaptive sampling effect of closed-loop control.

S45, feeding back the closed-loop control unit to trigger the updated sampling threshold valueThe method is applied to setting of trigger conditions in a data acquisition module and used for controlling dynamic adjustment of a trigger threshold value of subsequent sampling so as to realize closed-loop self-adaptive adjustment of sampling control.

According to the embodiment, based on a comparison result of the variable point posterior probability value and the set threshold, a sampling adjustment signal is dynamically generated, the triggering condition of data sampling is adjusted according to the sampling adjustment signal, specifically, the variable point posterior probability of a current time point is extracted and is compared with a preset probability judgment threshold, if the state abrupt change trend is judged, a sampling adjustment instruction signal is generated, a driving system reduces the sampling triggering threshold in a fixed step size mode, and then the updated threshold is reapplied to a data acquisition module through a feedback mechanism, so that closed-loop self-adaptive adjustment of sampling control is realized. The mechanism remarkably improves the dynamic response capability of the sampling strategy, so that the system can automatically improve the sampling frequency when the state is suddenly changed, reduces the resource consumption in a stable state, and achieves the beneficial effects of effectively controlling the energy consumption and the data redundancy while guaranteeing the monitoring sensitivity.

In this embodiment, the step S44 specifically includes:

;

Wherein, Fixing an adjustment coefficient for a positive real number;

;

Based on the above, the present embodiment dynamically updates the current sampling trigger threshold by using a linear decreasing strategy when the sampling adjustment command signal is received as1 by setting the adjustment signal control mechanism, the update amplitude is determined by a fixed step-size coefficient, and if the adjustment signal is 0, the threshold is kept unchanged. The updated sampling trigger threshold is used as feedback input to be applied to data sampling judgment of the next period, and a closed-loop control link with self-regulating parameters is formed. The method realizes the self-adaptive response of the sampling strategy to the monitoring sensitivity change, can automatically enhance the data sampling capability when the key state is suddenly changed, and suppresses invalid sampling in a stable stage, thereby remarkably improving the resource utilization rate and the dynamic adaptability of the system.

In this embodiment, the step S5 specifically includes:

;

The formula divides the complete rehabilitation process into a plurality of non-overlapping time intervals by dividing continuous time sequence data according to adjacent state abrupt change time points. The method takes the mutation points as phase boundaries, and ensures that the data in each phase has relatively consistent dynamic characteristics. The principle is that the state mutation is used as a natural demarcation point of the rehabilitation state change, so that the phase division conforming to the physiological change logic is constructed, the subsequent feature extraction and grade judgment are more accurate and reliable, and the phase division error possibly caused by manually setting a fixed time window is avoided.

The adopted phase division method takes the identified state mutation time points as boundaries, and sequentially carries out slicing operation on the standardized time sequence data according to a time sequence, when the method is specifically implemented, all the state mutation time points are firstly arranged according to a time ascending sequence, then a time range between the current mutation point and the previous mutation point is used as a recovery phase interval, the first phase is continued from the monitoring starting time to the monitoring ending time, the clear starting and ending boundaries of each phase interval are ensured, the original time sequence data are completely covered, and the structured phase organization of the data is realized.

S53, a stage feature extraction unit for each rehabilitation stage intervalThe normalized time sequence data in the system calculates statistical characteristic indexes including mean value, variance and trend slope to form a stage characteristic vector;

The statistical characteristic indexes are used for calculating the average value of all data points in each rehabilitation stage interval by traversing the standardized time sequence data in the interval, dividing the total level of the stage by the number of the data points after adding all the standardized time sequence data values in the stage interval, calculating the variance, calculating the square of the difference between each data point and the average value and averaging the square on the basis of obtaining the average value, calculating the trend slope, carrying out first-order linear regression fitting on the data points by adopting a least square method, and extracting the slope value as the index of the data change trend in the stage. The above operation is independently executed in each segment interval, and the three output characteristic values jointly form a stage characteristic vector which is used as the input of the subsequent rehabilitation stage grade judgment.

;

Wherein, Is the firstStage recovery stage grade.

Inputting a feature vector of each stage interval into a predefined grading rule set, wherein the rule set is definitely divided based on the value ranges of the mean value, the variance and the trend slope in the feature vector, and each group of rules corresponds to a unique rehabilitation stage grade label;

The system is matched item by item according to rule conditions, determines the grade to which the current stage belongs, and outputs a corresponding grade result of the rehabilitation stage, which is used for marking the rehabilitation state of the stage, and realizing standardized hierarchical expression of the rehabilitation process.

According to the embodiment, the state mutation time point set identified by the change point detection module is received, the standardized time sequence data is divided into a plurality of rehabilitation stage intervals in time sequence, statistical features such as mean values, variances, trend slopes and the like are extracted in each stage interval, stage feature vectors are constructed, each feature vector is analyzed through a preset stage grade mapping function, corresponding rehabilitation stage grade labels are output, automatic segmentation and grade classification of a rehabilitation process are achieved, accurate judgment of individual rehabilitation progress is facilitated, stage decision basis is provided for remote nursing intervention, and the intelligentization and individuation level of rehabilitation management are improved.

Example 1:

In order to verify the feasibility of the invention in implementation, the invention is applied to a rehabilitation medical center of a trimethyl hospital, and an actual clinical test point is developed for three months. The center mainly receives a great number of postoperative recovery, cerebral apoplexy recovery and long-term chronic patients on daily basis, and the rehabilitation evaluation flow of the center mainly depends on periodic manual inquiry and periodic detection equipment, such as a gait analyzer, a surface myoelectric instrument and the like, but due to the long detection period and the dependence of professional operation in the process, the problems that physiological state change cannot be continuously tracked, the fluctuation of the rehabilitation state of the patients is difficult to capture in time and the like exist, and accurate and dynamic rehabilitation intervention is difficult to realize.

In the test point, 150 rehabilitation inpatients are provided with the edge acquisition terminal equipment of the system for acquiring multisource physiological parameters including heart rate, galvanic skin response, lower limb myoelectricity, respiratory rate and the like. The system automatically starts real-time data acquisition after each patient wears the equipment, dynamically analyzes the variation amplitude of the physiological parameters through a sliding time window, automatically triggers data sampling when any index variation exceeds a set threshold, and avoids resource waste caused by redundant sampling.

And after filtering, denoising and time alignment treatment are carried out on the sampled data, the data is transmitted into an embedded processing module, and the state mutation probability is calculated in real time based on a Bayes online variable point detection algorithm. And the system adaptively adjusts sampling triggering conditions according to the calculated variable point posterior probability value. If the system detects that the mutation probability is higher than the judgment threshold, the sampling trigger threshold is dynamically reduced, so that the sampling frequency of a key period is improved, and the automatic balance of resource allocation and monitoring sensitivity is realized.

In the rehabilitation process, the system automatically segments the standardized data through the variable point judgment result, extracts the characteristics of mean value, variance, change trend and the like in each segment, outputs the rehabilitation stage grade of the patient through the established rehabilitation stage mapping model, and automatically synchronizes to the remote rehabilitation management platform. The rehabilitation doctor can remotely evaluate the current progress of the patient based on the grade change trend and the stage length, and adjust the rehabilitation scheme accordingly, such as increasing the rehabilitation training frequency, changing the intervention means, and the like. Meanwhile, the remote platform can also return the regulation and control parameters to the edge terminal, so that real remote monitoring and bidirectional regulation closed loop is realized.

In a three month run, the hospital collected 150 patients' recovery data and analyzed the system in comparison to the traditional recovery regimen. The following is part of the core data:

TABLE 1 comparison of the effects of the inventive system with the traditional rehabilitation evaluation method

The data table shows that the system greatly improves the effective data capturing rate to 91.8% while reducing the sampling frequency by 42%, obviously improves the energy efficiency and analysis effectiveness of the system, improves the accuracy from 73.2% of the traditional mode to 94.5% in the aspects of recognition and grade output in the rehabilitation stage, and can finish automatic grade output within 5 minutes, thereby greatly shortening the time required by manual evaluation.

For example, in the test procedure, abnormal fluctuation of heart rate and skin electricity occurs in the second stage rehabilitation process of a patient after cerebral apoplexy operation, the system identifies state mutation, the sampling frequency is increased through a dynamic regulation mechanism, rehabilitation fluctuation exists in the stage within 15 minutes, the grade model is used for judging that the patient is degenerated from 'moderate rehabilitation' to 'initial recovery' level, a doctor immediately adjusts a rehabilitation training scheme after receiving the feedback, and the traditional endurance training is adjusted to a combined scheme of slightly repeated movement and nerve stimulation, so that the step back caused by rehabilitation overload of the patient is avoided, and the subsequent recovery period is shortened.

In summary, the rehabilitation nursing remote monitoring transmission system provided by the invention constructs a closed-loop intelligent rehabilitation management flow from perception to evaluation to intervention feedback through real-time perception, variable point analysis, self-adaptive regulation and stage grade identification of the physiological state, solves the problems of long period, slow response, thick grading and the like of the traditional rehabilitation monitoring, remarkably improves the visualization degree, intelligent evaluation capability and intervention response efficiency of the rehabilitation process, and provides effective support for promoting the floor application of intelligent rehabilitation in clinical care.

The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.

Claims

1. A remote monitoring and transmission system for rehabilitation nursing, characterized by comprising:

S1, data acquisition module, collects multi-source physiological parameters of the user, constructs a sliding time window and calculates the dynamic change amplitude of each multi-source physiological parameter, and performs data sampling when the dynamic change amplitude of any multi-source physiological parameter exceeds the set sampling trigger threshold to generate a sampling data sequence;

S2, data preprocessing module, preprocesses the sampled data sequence, including filtering, noise reduction and time alignment, to generate standardized time series data;

S3, change point detection module, based on standardized time series data, uses Bayesian online change point detection method to perform recursive analysis to obtain the posterior probability of the change point at each time point, and when the posterior probability of the change point is greater than the set probability judgment threshold, the corresponding time point is identified as the state mutation time point;

S4, sampling control module, adjusts the sampling trigger condition according to the comparison result between the posterior probability of the change point at any time point and the set probability judgment threshold, and realizes closed-loop adaptive adjustment of sampling control;

S5, a rehabilitation status recognition module, which segments the standardized time series data based on the identified state mutation time points, generates rehabilitation stage intervals, and outputs the rehabilitation stage level corresponding to each state mutation time point;

S6, edge collaboration and transmission module, synchronously transmits standardized time series data, change point posterior probability and rehabilitation stage level between the edge and the remote server through a two-way communication mechanism, where the edge is a local terminal device that performs data collection and analysis tasks, and the remote server is a rehabilitation monitoring platform for centralized storage and control.

2. The rehabilitation nursing remote monitoring transmission system according to claim 1, characterized in that S1 specifically comprises:

S11, a physiological parameter acquisition unit, which acquires multi-source physiological parameters of the user to form multi-channel original physiological parameters for synchronous recording;

S12, sliding window processing unit, constructs the original physiological parameters corresponding to each channel with a length of The sliding time window is updated according to a fixed step size to form multiple sliding time window segments that are continuous in time;

S13, a change amplitude extraction unit, taking the original physiological parameters in each sliding time window as input, and calculating the corresponding dynamic change amplitude, the dynamic change amplitude Defined as:

;

in, Indicates The original physiological parameter set within a sliding time window, Indicated in The last original physiological parameter in a sliding time window;

S14, the sampling trigger judgment unit calculates the dynamic change amplitude The sampling trigger threshold is set Compare, if satisfied , a data sampling event is triggered;

S15, a sampling data construction unit extracts and encapsulates all original physiological parameters in the sliding window corresponding to the trigger event into a group of sampling segments, and combines them in chronological order to form a sampling data sequence.

3. The rehabilitation nursing remote monitoring transmission system according to claim 2, characterized in that the step S14 specifically comprises:

S141, threshold configuration unit, setting sampling trigger threshold , the sampling trigger threshold is a positive real number, which is used to compare with the dynamic change amplitude to determine whether to trigger sampling;

S142, single channel comparison unit, receiving dynamic change amplitude , and the corresponding and Compare and if the conditions are met , then mark the current sliding time window of the channel as a candidate trigger window;

S143, a multi-channel fusion determination unit summarizes the candidate trigger windows of all channels, and generates a sampling trigger signal if at least one channel satisfies the comparison condition;

S144, the trigger control unit extracts the data index corresponding to the current sliding time window based on the sampling trigger signal, and generates a sampling event instruction for driving the sampling data construction unit to perform a sampling data sequence generation operation.

4. The rehabilitation nursing remote monitoring transmission system according to claim 1, characterized in that said S3 specifically comprises:

S31, a change point probability calculation unit receives standardized time series data, performs recursive analysis based on the Bayesian online change point detection method, and calculates the change point posterior probability value of each time point as a state mutation point. The change point posterior probability value is defined as:

;

in, Indicates time point A standardized time series data subsequence of Indicates time point The length of the interval between the previous state mutation point, Indicates time point is the new state mutation point, Indicates the normalized time series data subsequence Under given conditions, the time point is the posterior distribution of the state mutation point, For time point The corresponding change point posterior probability value;

S32, a change point probability sequence construction unit receives the change point posterior probability value at each time point , arranged in chronological order, generating a change point probability sequence ,in Indicates the end time point index of the time series data;

S33, change point discrimination unit, setting probability judgment threshold , the posterior probability value of each change point in the change point probability sequence and Compare and judge the condition as follows ,in, To set the probability judgment threshold, the range is , used to define the confidence level of the state mutation point determination. When the conditions are met, the time point is determined. is a candidate state mutation point;

S34, state mutation output unit, receives candidate state mutation points, performs tag confirmation and outputs a state mutation time point set ,in Indicates the time point identified as a state mutation.

5. The rehabilitation nursing remote monitoring transmission system according to claim 4, characterized in that said S31 specifically comprises:

S311, observation sequence construction unit, receiving standardized time series data, at each time point Construct the corresponding time series data subsequence ;

S312, prior modeling unit, define state variables Indicates time point The length of the interval between the previous state mutation point and the prior distribution of the for:

;

in, is the prior probability of the state mutation point occurring, satisfying , , is the value of the interval length;

S313, observation model evaluation unit, building conditional observation model , assuming that the standardized time series data subsequence within the stateless mutation point interval It obeys Gaussian distribution and is expressed as:

;

in, Represents the mean of the current segment, which is updated by autoregression based on the data from the segment start point to the moment before the current time point. is the preset observation noise variance;

S314, posterior probability calculation unit, based on prior distribution , conditional observation model , and the posterior distribution of the previous moment ,in, To standardize the time series data subsequence Under given conditions, the time point The posterior distribution generated by the change point recursion is calculated by recursion at the time point is the posterior probability value of the change point.

6. The rehabilitation nursing remote monitoring transmission system according to claim 4, characterized in that said S4 specifically comprises:

S41, change point probability extraction unit, receiving change point probability sequence , and extract the time point The corresponding change point posterior probability value ;

S42, threshold comparison unit, setting probability judgment threshold , the time point The corresponding change point posterior probability value Probability judgment threshold Compare and determine whether ;

S43, sampling adjustment signal generating unit, when judging that the When the sampling adjustment command signal is generated ,in Indicates that the sampling trigger threshold needs to be adjusted dynamically. Indicates that there is no need to dynamically adjust the sampling trigger threshold;

S44, triggering the threshold adjustment unit to receive the sampling adjustment instruction signal , dynamically update the sampling trigger threshold , the adjustment strategy is:

;

in, is a positive real fixed adjustment coefficient, Indicates time point The sampling trigger threshold is the updated sampling trigger threshold;

S45, feedback closed-loop control unit, updates the sampling trigger threshold Applied to the setting of trigger conditions in the data acquisition module, it is used to control the dynamic adjustment of subsequent sampling trigger thresholds to achieve closed-loop adaptive regulation of sampling control.

7. The rehabilitation nursing remote monitoring transmission system according to claim 6, characterized in that said S44 specifically comprises:

S441, threshold receiving unit, receiving sampling trigger threshold And sampling adjustment command signal ,in, Indicates whether adjustment is needed;

S442, the threshold value calculation unit, when sampling and adjusting the instruction signal When , the threshold adjustment operation is triggered, and the sampling trigger threshold is updated according to the linear decreasing strategy. The update formula is:

;

in, is a positive real fixed adjustment coefficient;

S443, the threshold holding unit, when sampling and adjusting the instruction signal When , the current sampling trigger threshold remains unchanged, that is:

;

S444, threshold output unit, outputs the adjusted sampling trigger threshold , used to set the trigger conditions in the subsequent data acquisition module and as the starting threshold for the next cycle of dynamic sampling control.

8. The rehabilitation nursing remote monitoring transmission system according to claim 4, characterized in that said S5 specifically comprises:

S51, mutation point receiving unit, receiving state mutation time point set ,in Indicates the time point identified as a state mutation;

S52, stage segmentation unit, according to the state mutation time point set Segment the standardized time series data in chronological order to construct multiple non-overlapping time intervals , where each stage interval is defined as:

;

in, , , Indicates The time interval corresponding to each rehabilitation stage is represents the total number of time intervals during the rehabilitation phase;

S53, a stage feature extraction unit, for each time interval corresponding to the rehabilitation stage The standardized time series data in the calculation of statistical characteristic indicators, including mean, variance, trend slope, constitute the stage characteristic vector ;

S54, stage level determination unit, based on the stage feature vector Introducing stage-level mapping functions , mapping each stage interval to a rehabilitation stage level :

;

in, For the Stages of rehabilitation levels.