Postoperative psychological state evaluation system and method based on MPNFS theory

Abstract

The invention belongs to the technical field of postoperative psychological assessment, and particularly relates to a postoperative psychological state assessment system and method based on an MPNFS theory. According to the method, the accuracy and reliability of evaluation are effectively improved by performing multi-dimensional monitoring and analysis on the postoperative psychological state data of the patient, the MPNFS theory is adopted as an evaluation basis, the change characteristics of the psychological state of the patient can be comprehensively captured, the subjectivity and one-sidedness of a traditional evaluation method are avoided, and the evaluation accuracy and reliability are improved. Meanwhile, the influence of individual differences of patients on the evaluation result is fully considered, and the pertinence and effectiveness of the evaluation result are ensured. In specific implementation, trend features, periodic features and mutability features of the combined feature vectors are extracted, confidence assessment is performed on trend variation, periodic fluctuation and instantaneous variation parameters in combination with a confidence assessment strategy, and confidence weights are distributed according to assessment results, so that personalized psychological state assessment of patients can be realized, and the accuracy of psychological state assessment is improved. And the postoperative psychological state of the patient is predicted.

Description

Technical Field

[0001] The present invention belongs to the technical field of postoperative psychological assessment, and in particular relates to a postoperative psychological state assessment system and method based on the MPNFS theory. Background Art

[0002] With the advancement of medical technology, the assessment of the psychological state of postoperative patients has received increasing attention, so as to provide patients with more accurate personalized treatment plans, promote their comprehensive physical and mental recovery, reduce postoperative complications, and improve their quality of life. Traditional psychological assessment methods mostly rely on subjective judgment and lack systematic and quantitative basis. The MPNFS theory can provide objective and multi-dimensional data analysis, effectively make up for the shortcomings of traditional methods, improve the accuracy and reliability of assessment, and provide strong support for clinical decision-making.

[0003] In the existing technology, although there are some psychological state assessment methods based on MPNFS theory, they generally have problems such as complex data processing and poor real-time performance, which makes it difficult to meet the clinical rapid response needs, resulting in delayed assessment results and affecting the timing of treatment. In addition, the psychological state assessment systems in the existing technology often ignore the impact of individual differences in patients on the assessment results, resulting in limited assessment accuracy. The psychological state changes of different individuals have different patterns, and the corresponding assessment characteristics are also different. The use of a unified assessment standard is obviously unable to fully reflect individual differences. Based on this, this solution provides a postoperative psychological state assessment method based on MPNFS theory to solve the above problems. Summary of the Invention

[0004] The purpose of the present invention is to provide a postoperative psychological state assessment system and method based on the MPNFS theory, which can determine the specific selected assessment features according to the individual differences of patients and achieve the purpose of personalized assessment.

[0005] The technical solutions adopted by the present invention are as follows: A postoperative psychological status assessment method based on MPNFS theory, including: Collect postoperative psychological status assessment data of patients under drug treatment, psychological intervention, clinical nursing, family care and social support. The psychological status assessment data includes respiratory monitoring data, voice recording data and bioelectric signal data; Perform time-frequency synchronization processing on respiratory monitoring data, voice recording data, and bioelectric signal data to output a joint feature vector. Perform time series analysis on the joint feature vector through a sliding window mechanism to output the time series features of the joint feature vector. The time series features include trend features, periodic features, and mutation features. Determine the trend change amount of the joint feature vector based on the trend characteristics, determine the periodic fluctuation amount of the joint feature vector based on the periodic characteristics, and determine the instantaneous change parameter of the joint feature vector based on the mutation characteristics; Conduct confidence assessments on trend change, periodic fluctuation, and instantaneous change parameters respectively, and assign confidence weights to trend change, periodic fluctuation, and instantaneous change based on the confidence assessment results; Based on the trend change amount, periodic fluctuation amount and instantaneous change parameters, as well as the corresponding confidence weights, the preset prediction nodes are comprehensively compared and predicted, and a psychological state score is output. When the psychological state score exceeds the preset assessment threshold, an alarm signal is issued and a psychological state assessment report is generated.

[0006] In a preferred embodiment, when collecting respiratory monitoring data, the patient's respiratory rate, respiratory depth and respiratory rhythm are monitored by wearable devices. The wearable devices include smart bracelets, chest-strap respiratory monitors and respiratory sensors attached to the patient's skin. When collecting voice recording data, the patient's voice content, including voice volume, speaking speed, tone and emotional tendency, is recorded by a microphone. When collecting bioelectric signal data, electrocardiogram, electromyogram and electroencephalogram signals are collected using bioelectric sensors attached to the patient's skin.

[0007] In a preferred embodiment, the step of performing time-frequency synchronization processing on the respiratory monitoring data, the voice recording data, and the bioelectric signal data and outputting a joint feature vector includes: Perform time-frequency decomposition on respiratory monitoring data to extract respiratory fluctuation characteristics in the respiratory monitoring data; Performing spectral processing on the voice recording data to extract intonation variation features in the voice recording data; Perform wavelet transform on bioelectric signal data to extract rhythmic change features in bioelectric signals; Dynamic time alignment technology is used to align the timestamps of respiratory fluctuation features, intonation change features and rhythm change features to generate a time-frequency synchronized joint feature vector.

[0008] In a preferred solution, the step of performing time series analysis on the joint feature vector through a sliding window mechanism and outputting the time series features of the joint feature vector includes: Set the window length and sliding step of the sliding window. The window length is dynamically adjusted according to the data sampling frequency, and the sliding step is set according to the sensitivity of data changes. Calculate the difference between adjacent joint feature vectors in the sliding window, and when the difference exceeds a preset difference threshold, record the corresponding sliding window as a mutation window; when the difference does not exceed the preset threshold, record it as a stable window; Count the number of mutation windows and the mutation amount of the joint feature vector within the mutation window, and store them as mutation type features; Identify the joint eigenvectors within the allowable fluctuation range in each stationary window, and output the fluctuation amplitude and fluctuation period of the joint eigenvectors within the allowable fluctuation range as periodic features for storage; The average rate of change of the joint eigenvector within the stationary window is analyzed by trend fitting and stored as a trend feature.

[0009] In a preferred embodiment, the steps of determining the trend change amount of the joint feature vector based on the trend feature, determining the periodic fluctuation amount of the joint feature vector based on the periodic feature, and determining the instantaneous change parameter of the joint feature vector based on the mutation feature include: Collect the joint feature vector of the current evaluation node and the demand forecast interval between the current node and the forecast node, input the joint feature vector, demand forecast interval and average change rate of the current node into the trend evaluation function, and output the trend change of the joint feature vector; Extract the periodic fluctuation amplitude of the joint eigenvector under each fluctuation period, perform weighted average processing, and output the periodic fluctuation amount of the joint eigenvector; Extract the mutation amount and number of mutations in each mutation window, determine the distribution segment of the mutation window according to the time interval between the mutation windows, perform weighted fusion processing on the mutation amount and number of mutations in each distribution segment, and output them as instantaneous change parameters; Among them, the closer the distribution segment is to the current evaluation node, the greater the corresponding weighted weight.

[0010] In a preferred embodiment, the step of performing confidence assessment on the trend change amount, the periodic fluctuation amount and the instantaneous change parameter respectively includes: Randomly extract multiple groups of sample data from existing historical data, perform simulation calculations on each group of sample data, and output simulated trend change, simulated period fluctuation and simulated instantaneous change parameters; Calculate the deviation rate between the simulated trend change amount and the trend change amount, and record it as a first condition parameter. Then compare the first condition parameter with a preset first threshold. When the first condition parameter is higher than the first threshold, directly determine that the trend change amount output is wrong, increase sample data and re-simulate the calculation. When the first condition parameter is lower than the first threshold, calculate the excess ratio of the first condition parameter and record it as a first confidence assessment indicator. Calculate the matching degree between the simulated periodic fluctuation amount and the periodic fluctuation amount, and record it as the second condition parameter. Then compare the second condition parameter with a preset second threshold value. When the second condition parameter is lower than the second threshold value, directly determine that the periodic fluctuation amount output is wrong, increase sample data and re-simulate the calculation. When the second condition parameter is higher than the second threshold value, calculate the excess ratio of the second condition parameter and record it as the second confidence evaluation indicator. Calculate the similarity between the simulated instantaneous change parameter and the instantaneous change parameter, and record it as the third condition parameter. Then compare the third condition parameter with the preset third threshold. When the third condition parameter is lower than the third threshold, directly increase the sample data and re-simulate the calculation. If the third condition parameter is still lower than the third threshold after adding the sample, it is determined that the instantaneous change parameter intervention is invalid, and no confidence assessment is performed; if the third condition parameter is higher than the third threshold, calculate its excess ratio and record it as the third confidence assessment indicator.

[0011] In a preferred embodiment, the step of allocating confidence weights to the trend change amount, the periodic fluctuation amount, and the instantaneous change amount according to the confidence evaluation result includes: Obtaining a first confidence evaluation index, a second confidence evaluation index, and a third confidence evaluation index; Inputting the first confidence index, the second confidence index, and the third confidence index into a preset weight distribution function, and using the output of the weight distribution function as the confidence weights of the trend change amount, the periodic fluctuation amount, and the instantaneous change amount; When the third confidence evaluation index is not output, the value of the third confidence evaluation index is assumed to be zero.

[0012] In a preferred embodiment, the step of performing a comprehensive comparison and prediction of preset prediction nodes based on the trend change amount, the periodic fluctuation amount, the instantaneous change parameter, and the corresponding confidence weights, and outputting a psychological state score includes: Get the prediction node timestamp and the historical timestamp parallel to the prediction node timestamp; Collect the confidence weights of the first confidence evaluation indicator, the second confidence evaluation indicator, and the third confidence evaluation indicator at the historical timestamp, and arrange the confidence weights of the first confidence indicator, the second confidence indicator, and the third confidence indicator in descending order, and record the confidence weight with the highest ranking as the primary confidence weight; When the main confidence weight corresponds to the trend change amount, the predicted patient's mental state score is calculated based on the time interval between the prediction node timestamp and the historical timestamp; When the primary confidence weight corresponds to the periodic fluctuation, the predicted fluctuation ratio at the prediction node timestamp will be determined based on the periodic fluctuation at the historical timestamp, and the historical psychological state score at the historical timestamp will be proportionally adjusted based on the predicted fluctuation to obtain the psychological state score at the prediction node timestamp. When the main confidence weight corresponds to the instantaneous change, the historical psychological state score at the historical timestamp is recorded as the psychological state score at the prediction node timestamp.

[0013] The present invention also provides a postoperative psychological state assessment system based on the MPNFS theory, using the above-mentioned postoperative psychological state assessment method based on the MPNFS theory, comprising: The data collection module is used to collect the patient's postoperative psychological status assessment data under drug treatment, psychological intervention, clinical nursing, family care and social support. The psychological status assessment data includes respiratory monitoring data, voice recording data and bioelectric signal data; The time series analysis module is used to perform time-frequency synchronization processing on respiratory monitoring data, voice recording data, and bioelectric signal data, output a joint feature vector, and perform time series analysis on the joint feature vector through a sliding window mechanism to output the time series features of the joint feature vector. The time series features include trend features, periodic features, and mutation features. A feature quantification module is used to determine the trend change amount of the joint feature vector based on the trend feature, determine the periodic fluctuation amount of the joint feature vector based on the periodic feature, and determine the instantaneous change parameter of the joint feature vector based on the mutation feature; The weight calculation module is used to perform confidence assessment on the trend change amount, periodic fluctuation amount and instantaneous change parameters respectively, and assign confidence weights to the trend change amount, periodic fluctuation amount and instantaneous change amount according to the confidence assessment results; The state prediction module is used to perform comprehensive comparison and prediction of preset prediction nodes based on trend change, periodic fluctuation and instantaneous change parameters, as well as corresponding confidence weights, output a psychological state score, and issue an alarm signal when the psychological state score exceeds the preset assessment threshold, and generate a psychological state assessment report.

[0014] And, an electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned postoperative psychological state assessment method based on MPNFS theory.

[0015] The technical effects achieved by the present invention are: The present invention effectively improves the accuracy and reliability of the assessment by conducting multi-dimensional monitoring and analysis of the patient's postoperative psychological state data. By adopting the MPNFS theory as the basis for assessment, it can comprehensively capture the changing characteristics of the patient's psychological state, avoiding the subjectivity and one-sidedness of traditional assessment methods. At the same time, it also fully considers the impact of individual differences of patients on the assessment results, ensuring the pertinence and effectiveness of the assessment results. In the specific implementation, the trend characteristics, periodic characteristics and mutation characteristics of the joint feature vector are extracted. Combined with the confidence assessment strategy, the trend change amount, periodic fluctuation amount and instantaneous change parameter are evaluated with confidence, and confidence weights are allocated according to the assessment results. In this way, personalized psychological state assessment of patients can be achieved, the patient's postoperative psychological state can be predicted, and the timeliness of medical intervention can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of the method of the present invention; Figure 2 It is a schematic diagram of the system modules of the present invention; Figure 3 It is a schematic structural diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it constitute a separate or selective embodiment that is mutually exclusive of other embodiments.

[0020] See also Figure 1 As shown, the present invention provides a postoperative psychological state assessment method based on the MPNFS theory, comprising: S1. Collect postoperative psychological status assessment data of patients under drug treatment, psychological intervention, clinical nursing, family care, and social support. The psychological status assessment data includes respiratory monitoring data, voice recording data, and bioelectric signal data. In step S1, when evaluating the patient's postoperative psychological state, it is first necessary to collect the patient's postoperative psychological state evaluation data under the influence of multiple factors such as drug treatment, psychological intervention, clinical nursing, family care and social support. The psychological state evaluation data covers multiple dimensions such as respiratory monitoring data, voice recording data and bioelectric signal data to ensure the comprehensiveness and accuracy of the evaluation. Among them, when collecting respiratory monitoring data, the patient's respiratory rate, respiratory depth and respiratory rhythm are monitored by wearable devices. The wearable devices include smart bracelets, chest-strap respiratory monitors and respiratory sensors attached to the patient's skin. When collecting voice recording data, the patient's voice content, including voice volume, speaking speed, tone and emotional tendency, is recorded by a microphone. When collecting bioelectric signal data, electrocardiogram, electromyogram and electroencephalogram signals are collected using bioelectric sensors attached to the patient's skin. Specifically, when collecting respiratory monitoring data, voice recording data and bioelectric signal data, in order to ensure the accuracy and completeness of the data, the collected raw data needs to be preprocessed. The preprocessing steps include denoising the respiratory monitoring data to eliminate the interference of environmental noise and instrument errors on the respiratory data, filtering the voice recording data to improve the clarity and recognition of the voice signal, and baseline correction and artifact removal of the bioelectric signal data to reduce interference caused by electrode movement or muscle activity, so as to provide accurate data support for subsequent analysis and ensure the accuracy and reliability of psychological state assessment.

[0021] S2. Perform time-frequency synchronization processing on the respiratory monitoring data, voice recording data, and bioelectric signal data to output a joint feature vector. Perform time series analysis on the joint feature vector through a sliding window mechanism to output the time series features of the joint feature vector. The time series features include trend features, periodic features, and mutation features. In step S2, after the respiratory monitoring data, voice recording data and bioelectric signal data are output, time-frequency synchronization processing is first performed to ensure the consistency of the respiratory monitoring data, voice recording data and bioelectric signal data on the time axis. Then, the synchronized respiratory monitoring data, voice recording data and bioelectric signal data are fused into a joint feature vector to comprehensively reflect the changes in the patient's psychological state. Then, a sliding window mechanism is used to perform time series analysis on the joint feature vector to extract the time series features of the joint feature vector. The time series features include trend features, periodic features and mutation features, which provide key indicators for subsequent psychological state assessment. The trend features reflect the long-term change trend of the patient's psychological state, the periodic features reveal the law of periodic fluctuations of the psychological state over time, and the mutation features capture the abnormal situation of sudden changes in the psychological state in a short period of time. Among them, the steps of performing time-frequency synchronization processing on the respiratory monitoring data, voice recording data and bioelectric signal data and outputting the joint feature vector include: Perform time-frequency decomposition on respiratory monitoring data to extract respiratory fluctuation characteristics in the respiratory monitoring data; Performing spectral processing on the voice recording data to extract intonation variation features in the voice recording data; Perform wavelet transform on bioelectric signal data to extract rhythmic change features in bioelectric signals; Dynamic time alignment technology is used to align the timestamps of respiratory fluctuation features, intonation change features, and rhythm change features to generate a time-frequency synchronized joint feature vector. Specifically, when outputting the joint feature vector, the respiratory monitoring data will be subjected to time-frequency decomposition processing, and the respiratory fluctuation characteristics will be extracted by decomposing different frequency components to reflect the potential connection between the patient's breathing pattern and psychological state. The voice recording data will be spectrally processed, and the pitch, speed, and emotional tendency of the voice will be extracted using methods such as Fourier transform or Mel spectrum analysis, which will help understand the patient's emotional state. The bioelectric signal data will be processed by wavelet transform to extract the rhythm change characteristics in the electrocardiogram, electromyogram and electroencephalogram signals, so as to reveal the patient's autonomic nervous system activity and brain activity state. Finally, the dynamic time alignment technology is used to align the respiratory fluctuation characteristics, intonation change characteristics and rhythm change characteristics according to the timestamp, so as to generate a time-frequency synchronized joint feature vector, which provides an accurate data basis for subsequent psychological state assessment.

[0022] Secondly, the steps of performing time series analysis on the joint feature vector through a sliding window mechanism and outputting the time series features of the joint feature vector include: Set the window length and sliding step of the sliding window. The window length is dynamically adjusted according to the data sampling frequency, and the sliding step is set according to the sensitivity of data changes. Calculate the difference between adjacent joint feature vectors in the sliding window, and when the difference exceeds a preset difference threshold, record the corresponding sliding window as a mutation window; when the difference does not exceed the preset threshold, record it as a stable window; Count the number of mutation windows and the mutation amount of the joint feature vector within the mutation window, and store them as mutation type features; Identify the joint eigenvectors within the allowable fluctuation range in each stationary window, and output the fluctuation amplitude and fluctuation period of the joint eigenvectors within the allowable fluctuation range as periodic features for storage; The average rate of change of the joint eigenvector within the stationary window is analyzed by trend fitting and stored as a trend feature; In the above, when determining the time series characteristics of the joint eigenvector, it is first necessary to pre-set the parameters of the sliding window, including the window length and the sliding step, to ensure that the window length matches the data sampling frequency. The sliding step is adjusted according to the sensitivity of the data change to ensure the accuracy of data capture. In the sliding window, the difference between adjacent joint eigenvectors will be calculated. The difference reflects the intensity of the change of the psychological state in the time series. When the difference exceeds the preset difference threshold, it means that the patient's psychological state has a significant mutation. At this time, the corresponding sliding window will be marked as a mutation window. When the difference does not exceed the preset threshold, it is marked as a stable window, indicating that the psychological state is relatively stable during this period. For the mutation window, the number of mutation windows and the mutation amount of the joint eigenvector in each mutation window will be further counted and stored as a mutation feature. The mutation feature generally reflects the patient's The strong emotional reaction or psychological fluctuation of the patient in a specific situation, for the stable window, the joint eigenvector within the allowable fluctuation range will be identified, and the joint eigenvector within the allowable fluctuation range reflects the fluctuation of the patient's psychological state in a stable state, and the corresponding fluctuation amplitude and fluctuation period are recorded as periodic characteristics. The periodic characteristics generally reflect the patient's emotional ups and downs and regular changes in daily life. For example, the patient's mood is more active in the morning and evening, and the fluctuation period is consistent with the biological rhythm. Finally, the joint eigenvector within the stable window will be subjected to trend fitting analysis, and its average rate of change will be calculated and stored as a trend feature. The trend feature reflects the long-term trend of the patient's psychological state, which helps to evaluate the overall direction of the psychological state. The trend feature generally reflects the slow changes in the patient's psychological state over a long period of time, such as the gradual decline or recovery of mood.

[0023] S3. Determine the trend change amount of the joint feature vector based on the trend feature, determine the periodic fluctuation amount of the joint feature vector based on the periodic feature, and determine the instantaneous change parameter of the joint feature vector based on the mutation feature; In the above, after the trend characteristics, periodic characteristics and mutation characteristics are determined, the trend change amount of the joint feature vector is determined based on the extracted trend characteristics, so as to assist in understanding the long-term development trend of the patient's psychological state. The periodic fluctuation amount of the joint feature vector is determined based on the periodic characteristics to assist in identifying the periodic change law of the patient's psychological state. The instantaneous change parameter of the joint feature vector is determined based on the mutation characteristics to achieve the purpose of capturing the sudden change of the patient's psychological state. Among them, the steps of determining the trend change amount of the joint feature vector based on the trend characteristics, determining the periodic fluctuation amount of the joint feature vector based on the periodic characteristics, and determining the instantaneous change parameter of the joint feature vector based on the mutation characteristics include: Collect the joint feature vector of the current evaluation node and the demand forecast interval between the current node and the forecast node, input the joint feature vector, demand forecast interval and average change rate of the current node into the trend evaluation function, and output the trend change of the joint feature vector; Extract the periodic fluctuation amplitude of the joint eigenvector under each fluctuation period, perform weighted average processing, and output the periodic fluctuation amount of the joint eigenvector; Extract the mutation amount and number of mutations in each mutation window, determine the distribution segment of the mutation window according to the time interval between the mutation windows, perform weighted fusion processing on the mutation amount and number of mutations in each distribution segment, and output them as instantaneous change parameters; Among them, the closer the distribution segment is to the current evaluation node, the greater the corresponding weight; Specifically, when determining the trend change of the joint feature vector based on the trend characteristics, the joint feature vector under the current evaluation node will be collected first. At the same time, considering that the development of the patient's psychological state has a certain inertia, it is also necessary to obtain the demand prediction interval between the current node and the prediction node. The demand prediction interval reflects the time span of the psychological state from the current state to a certain predicted state in the future. Then, the joint feature vector under the current node, the demand prediction interval and the average change rate obtained by the trend fitting analysis are input into the trend evaluation function together to predict and output the trend change of the joint feature vector. The expression of the trend evaluation function is: trend change = average change rate × demand prediction interval + current joint feature vector. In the calculation of the periodic fluctuation amount, the periodic fluctuation amplitude of the joint feature vector under each fluctuation period will be extracted, and the periodic fluctuation amplitude will be calculated. The weight of the weighted average can be determined comprehensively based on factors such as the stability and frequency of the fluctuation period, and can be set specifically according to actual conditions to output the periodic fluctuation amount of the joint eigenvector. In the process of determining the instantaneous change parameter, the mutation amount and number of mutations in each mutation window are extracted. The mutation amount reflects the intensity of the change in the psychological state in a short period of time, and the number of mutations reflects the frequency of the psychological state mutation. At the same time, the distribution segment of the mutation window is determined according to the time interval between the mutation windows (the distribution segment can be determined by the interval partitioning method or the clustering algorithm). The closer the distribution segment is to the current evaluation node, the greater the impact on the instantaneous change parameter, and therefore the corresponding weighted weight is also greater. Afterwards, the mutation amount and number of mutations in each distribution segment are weightedly fused and output as the instantaneous change parameter.

[0024] S4. Conduct confidence assessments on the trend change amount, periodic fluctuation amount, and instantaneous change parameters respectively, and assign confidence weights to the trend change amount, periodic fluctuation amount, and instantaneous change amount based on the confidence assessment results; In step S4, after the trend change amount, the periodic fluctuation amount, and the instantaneous change parameter are determined, the corresponding confidence levels are calculated respectively, and the confidence weights of the trend change amount, the periodic fluctuation amount, and the instantaneous change parameter in the subsequent evaluation process are determined according to the confidence evaluation results. The steps of performing confidence evaluation on the trend change amount, the periodic fluctuation amount, and the instantaneous change parameter respectively include: Randomly extract multiple groups of sample data from existing historical data, perform simulation calculations on each group of sample data, and output simulated trend change, simulated period fluctuation and simulated instantaneous change parameters; Calculate the deviation rate between the simulated trend change amount and the trend change amount, and record it as a first condition parameter. Then compare the first condition parameter with a preset first threshold. When the first condition parameter is higher than the first threshold, directly determine that the trend change amount output is wrong, increase sample data and re-simulate the calculation. When the first condition parameter is lower than the first threshold, calculate the excess ratio of the first condition parameter and record it as a first confidence assessment indicator. Calculate the matching degree between the simulated periodic fluctuation amount and the periodic fluctuation amount, and record it as the second condition parameter. Then compare the second condition parameter with a preset second threshold value. When the second condition parameter is lower than the second threshold value, directly determine that the periodic fluctuation amount output is wrong, increase sample data and re-simulate the calculation. When the second condition parameter is higher than the second threshold value, calculate the excess ratio of the second condition parameter and record it as the second confidence evaluation indicator. Calculate the similarity between the simulated instantaneous change parameter and the instantaneous change parameter and record it as the third condition parameter. Then compare the third condition parameter with the preset third threshold. If the third condition parameter is lower than the third threshold, directly increase sample data and re-simulate the calculation. If the third condition parameter is still lower than the third threshold after adding samples, it is determined that the instantaneous change parameter intervention is invalid and no confidence assessment is performed. If the third condition parameter is higher than the third threshold, calculate its excess ratio and record it as the third confidence assessment indicator. Specifically, when the confidence of trend change quantity, periodic fluctuation quantity and instantaneous change parameter is evaluated respectively, multiple groups of sample data are first randomly extracted from the existing historical database. The sample data include psychological status evaluation records of different patients. Then, each group of sample data is simulated and calculated. The simulation calculation process is consistent with the actual evaluation process, including preprocessing of respiratory monitoring data, voice recording data and bioelectric signal data, time-frequency synchronization processing and time series analysis, so as to output the simulated trend change quantity, simulated periodic fluctuation quantity and simulated instantaneous change parameter. Then, the deviation rate between the simulated trend change quantity and the trend change quantity obtained through the actual evaluation process is calculated. The deviation rate reflects the degree of difference between the simulation result and the actual result, and this deviation rate is recorded as the first conditional parameter. Then, the first conditional parameter is compared with the preset first threshold. If the first conditional parameter is higher than the first threshold, it means that the difference between the simulation result and the actual result is too large. At this time, it is directly determined that there is an error in the output of the trend change quantity, and the sample data is added and re-simulated to improve the accuracy of the evaluation. If the first conditional parameter is higher than the first threshold, it means that the difference between the simulation result and the actual result is too large. At this time, it is directly determined that there is an error in the output of the trend change quantity, and the sample data is added and re-simulated to improve the accuracy of the evaluation. If the conditional parameter is lower than the first threshold, it indicates that the simulation result is close to the actual result. In this case, the excess ratio of the first conditional parameter is calculated and recorded as the first confidence assessment index. Similarly, the matching degree between the simulated periodic fluctuation amount and the periodic fluctuation amount, and the similarity between the simulated instantaneous change parameter and the instantaneous change parameter are calculated, and the matching degree and similarity are recorded as the second conditional parameter and the third conditional parameter, respectively. The second conditional parameter and the third conditional parameter are then compared with the preset second threshold and the third threshold, respectively. Based on the comparison results, corresponding processing is performed, including determining output errors, adding sample data and re-simulating the calculation, and calculating the excess ratio, etc., to determine the second confidence assessment index and the third confidence assessment index. For the instantaneous change parameter, if the third conditional parameter is still lower than the third threshold after adding samples, the intervention of the instantaneous change parameter is determined to be ineffective. Specifically, in certain specific cases, the patient's mental state changes are too complex or abnormal, making it impossible to accurately assess using the current assessment method. In this case, the confidence assessment will no longer be performed to avoid misleading subsequent assessment results.

[0025] In addition, the steps of allocating confidence weights to the trend change amount, the periodic fluctuation amount, and the instantaneous change amount according to the confidence evaluation result include: Obtaining a first confidence evaluation index, a second confidence evaluation index, and a third confidence evaluation index; Inputting the first confidence index, the second confidence index, and the third confidence index into a preset weight distribution function, and using the output of the weight distribution function as the confidence weights of the trend change amount, the periodic fluctuation amount, and the instantaneous change amount; When the third confidence evaluation index is not output, the value of the third confidence evaluation index is assumed to be zero.

[0026] Specifically, after the first confidence evaluation index, the second confidence evaluation index, and the third confidence evaluation index are determined, the corresponding confidence weights will be assigned to the trend change, periodic fluctuation, and instantaneous change through a preset weight distribution function to ensure that each parameter has a reasonable proportion in the comprehensive evaluation. The expression of the weight distribution function is: , where Indicates the The corresponding confidence weight of the confidence evaluation indicator, Indicates the Confidence evaluation indicators, Represents the sum of the first confidence index, the second confidence index, and the third confidence index. In addition, it should be clear that when the third confidence evaluation index is zero, the weight distribution function will automatically adjust the weight ratio of the remaining two indicators to ensure the balance of the evaluation results.

[0027] S5. Perform a comprehensive comparison and prediction of the preset prediction nodes based on the trend change amount, periodic fluctuation amount, and instantaneous change parameters, as well as the corresponding confidence weights, and output a psychological state score. When the psychological state score exceeds a preset assessment threshold, an alarm signal is issued, and a psychological state assessment report is generated; In step S5, when calculating the psychological state score, the trend change amount, the periodic fluctuation amount, the instantaneous change parameter, and the corresponding confidence weight are comprehensively considered. When the output psychological state score exceeds a preset assessment threshold, an alarm mechanism is immediately triggered to prompt relevant personnel to pay attention to the patient's psychological state changes. The steps of comprehensively comparing and predicting the preset prediction nodes based on the trend change amount, the periodic fluctuation amount, the instantaneous change parameter, and the corresponding confidence weights and outputting the psychological state score include: Get the prediction node timestamp and the historical timestamp parallel to the prediction node timestamp; Collect the confidence weights of the first confidence evaluation indicator, the second confidence evaluation indicator, and the third confidence evaluation indicator at the historical timestamp, and arrange the confidence weights of the first confidence indicator, the second confidence indicator, and the third confidence indicator in descending order, and record the confidence weight with the highest ranking as the primary confidence weight; When the main confidence weight corresponds to the trend change amount, the predicted patient's mental state score is calculated based on the time interval between the prediction node timestamp and the historical timestamp; When the primary confidence weight corresponds to the periodic fluctuation, the predicted fluctuation ratio at the prediction node timestamp will be determined based on the periodic fluctuation at the historical timestamp, and the historical psychological state score at the historical timestamp will be proportionally adjusted based on the predicted fluctuation to obtain the psychological state score at the prediction node timestamp. When the main confidence weight corresponds to the instantaneous change, the historical psychological state score at the historical timestamp is recorded as the psychological state score at the prediction node timestamp.

[0028] Specifically, when outputting the psychological state score, it is first necessary to determine the timestamp of the preset prediction node. This timestamp is the time node of the psychological state of the patient at a certain moment in the future. Then, the historical timestamp parallel to the prediction node timestamp will be found, and then the confidence weights of the first confidence assessment indicator, the second confidence assessment indicator and the third confidence assessment indicator under the historical timestamp will be collected, and the confidence weights will be arranged in descending order to determine the main confidence weight. The main confidence weight will determine the calculation method of the psychological state score. If the main confidence weight corresponds to the trend change amount, the patient's psychological state score will be predicted based on the time interval between the prediction node timestamp and the historical timestamp, as well as the value of the trend change amount. Specifically, the trend change rate between the two is calculated, and weighted adjustment is performed in combination with the historical psychological state score to finally obtain the psychological state score under the prediction node timestamp. If the main confidence weight corresponds to the trend change amount, the patient's psychological state score will be predicted based on the time interval between the prediction node timestamp and the historical timestamp, as well as the value of the trend change amount. Specifically, the trend change rate between the two is calculated, and weighted adjustment is performed in combination with the historical psychological state score to finally obtain the psychological state score under the prediction node timestamp. If the weight corresponds to the periodic fluctuation amount, the predicted fluctuation ratio at the prediction node timestamp will be determined based on the periodic fluctuation amount at the historical timestamp, and then the historical psychological state score at the historical timestamp will be proportionally adjusted based on this predicted fluctuation amount to obtain the psychological state score at the prediction node timestamp. If the main confidence weight corresponds to the instantaneous change amount, since the instantaneous change amount reflects the sudden change in the patient's psychological state, the historical psychological state score at the historical timestamp will be directly recorded as the psychological state score at the prediction node timestamp, as a rapid response to the sudden change in the current psychological state. When the output psychological state score exceeds the pre-set evaluation threshold, the alarm mechanism will be triggered immediately, and relevant personnel will be prompted to pay attention to the patient's psychological state changes through various means such as sound, light and electricity, and a corresponding psychological state assessment report will be generated to provide corresponding reference for medical staff.

[0029] See also Figure 2 A postoperative psychological state assessment system based on the MPNFS theory, using the above-mentioned postoperative psychological state assessment method based on the MPNFS theory, includes: The data collection module is used to collect the patient's postoperative psychological status assessment data under drug treatment, psychological intervention, clinical nursing, family care and social support. The psychological status assessment data includes respiratory monitoring data, voice recording data and bioelectric signal data; The time series analysis module is used to perform time-frequency synchronization processing on respiratory monitoring data, voice recording data, and bioelectric signal data, output a joint feature vector, and perform time series analysis on the joint feature vector through a sliding window mechanism to output the time series features of the joint feature vector. The time series features include trend features, periodic features, and mutation features. A feature quantification module is used to determine the trend change amount of the joint feature vector based on the trend feature, determine the periodic fluctuation amount of the joint feature vector based on the periodic feature, and determine the instantaneous change parameter of the joint feature vector based on the mutation feature; The weight calculation module is used to perform confidence assessment on the trend change amount, periodic fluctuation amount and instantaneous change parameters respectively, and assign confidence weights to the trend change amount, periodic fluctuation amount and instantaneous change amount according to the confidence assessment results; The state prediction module is used to perform comprehensive comparison and prediction of preset prediction nodes based on trend change, periodic fluctuation and instantaneous change parameters, as well as corresponding confidence weights, output a psychological state score, and issue an alarm signal when the psychological state score exceeds the preset assessment threshold, and generate a psychological state assessment report.

[0030] In the above, the data acquisition module is responsible for collecting the patient's psychological state assessment data from multiple dimensions. The psychological state assessment data comes from multiple aspects such as drug treatment, psychological intervention, clinical nursing, family care and social support, specifically including respiratory monitoring data, voice recording data and bioelectric signal data. The time series analysis module performs time-frequency synchronization processing on the collected psychological state assessment data, extracts the joint feature vector, and analyzes its time series characteristics through the sliding window mechanism. The time series characteristics are specifically classified into trend characteristics, periodic characteristics and mutation characteristics. The feature quantification module determines the trend change amount of the joint feature vector according to the trend characteristics, and calculates the trend change amount of the joint feature vector according to the periodic characteristics. The periodic fluctuation amount is determined based on the characteristics of mutation, and the instantaneous change parameters are determined based on the mutation characteristics. The weight measurement module conducts confidence assessment on the trend change amount, periodic fluctuation amount and instantaneous change parameters, and allocates confidence weights according to the assessment results. The state prediction module conducts a comprehensive comparison and prediction of the psychological state of the preset prediction nodes based on the quantified trend change amount, periodic fluctuation amount and instantaneous change parameters, and finally outputs the psychological state score. If the psychological state score exceeds the preset assessment threshold, the alarm mechanism will be triggered to remind relevant personnel to pay attention to the patient's psychological state changes, and generate a corresponding psychological state assessment report to provide a scientific basis for the subsequent intervention of medical staff.

[0031] See also Figure 3 , an electronic device, the electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned postoperative psychological state assessment method based on MPNFS theory.

[0032] The processor of the above-mentioned electronic device can be a central processing unit (CPU), a graphics processing unit (GPU) or a digital signal processor (DSP), etc., or it can be a processor cluster composed of multiple processors. The memory can be a random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory (Flash Memory), a hard disk or a solid-state drive, etc. The memory is used to store computer programs that can be executed by the processor. When the electronic device is running, the processor executes the computer program stored in the memory to implement the above-mentioned postoperative psychological state assessment method based on MPNFS theory. The electronic device may also include input devices such as a keyboard and a touch screen for receiving user instructions; output devices such as a display screen and a printer for displaying assessment results; and an operator such as an arithmetic logic unit (ALU) for performing data processing.

[0033] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0034] The foregoing is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained herein shall, unless otherwise specified or limited, be implemented in accordance with conventional means in the art.

Claims

1. A postoperative psychological status assessment method based on the MPNFS theory, characterized by: include: Collect postoperative psychological status assessment data of patients under drug treatment, psychological intervention, clinical nursing, family care and social support. The psychological status assessment data includes respiratory monitoring data, voice recording data and bioelectric signal data; Perform time-frequency synchronization processing on respiratory monitoring data, voice recording data, and bioelectric signal data to output a joint feature vector. Perform time series analysis on the joint feature vector through a sliding window mechanism to output the time series features of the joint feature vector. The time series features include trend features, periodic features, and mutation features. Determine the trend change amount of the joint feature vector based on the trend characteristics, determine the periodic fluctuation amount of the joint feature vector based on the periodic characteristics, and determine the instantaneous change parameter of the joint feature vector based on the mutation characteristics; Conduct confidence assessments on trend change, periodic fluctuation, and instantaneous change parameters respectively, and assign confidence weights to trend change, periodic fluctuation, and instantaneous change based on the confidence assessment results; Based on the trend change amount, periodic fluctuation amount and instantaneous change parameters, as well as the corresponding confidence weights, the preset prediction nodes are comprehensively compared and predicted, and a psychological state score is output. When the psychological state score exceeds the preset assessment threshold, an alarm signal is issued and a psychological state assessment report is generated.

2. A postoperative psychological state assessment method based on MPNFS theory according to claim 1, characterized in that: When collecting respiratory monitoring data, the patient's respiratory rate, respiratory depth and respiratory rhythm are monitored through wearable devices. The wearable devices include smart bracelets, chest-strap respiratory monitors and respiratory sensors attached to the patient's skin. When collecting voice recording data, the patient's voice content, including voice volume, speaking speed, tone and emotional tendency, is recorded through a microphone. When collecting bioelectric signal data, electrocardiogram, electromyogram and electroencephalogram signals are collected using bioelectric sensors attached to the patient's skin.

3. The method for postoperative psychological status assessment based on MPNFS theory according to claim 1, characterized in that: The step of performing time-frequency synchronization processing on the respiratory monitoring data, the voice recording data, and the bioelectric signal data and outputting a joint feature vector includes: Perform time-frequency decomposition on respiratory monitoring data to extract respiratory fluctuation characteristics in the respiratory monitoring data; Performing spectral processing on the voice recording data to extract intonation variation features in the voice recording data; Perform wavelet transform on bioelectric signal data to extract rhythmic change features in bioelectric signals; Dynamic time alignment technology is used to align the timestamps of respiratory fluctuation features, intonation change features and rhythm change features to generate a time-frequency synchronized joint feature vector.

4. The method for postoperative psychological status assessment based on MPNFS theory according to claim 1, characterized in that: The step of performing time series analysis on the joint feature vector through a sliding window mechanism and outputting the time series features of the joint feature vector includes: Set the window length and sliding step of the sliding window. The window length is dynamically adjusted according to the data sampling frequency, and the sliding step is set according to the sensitivity of data changes. Calculate the difference between adjacent joint feature vectors in the sliding window, and when the difference exceeds a preset difference threshold, record the corresponding sliding window as a mutation window; when the difference does not exceed the preset threshold, record it as a stable window; Count the number of mutation windows and the mutation amount of the joint feature vector within the mutation window, and store them as mutation type features; Identify the joint eigenvectors within the allowable fluctuation range in each stationary window, and output the fluctuation amplitude and fluctuation period of the joint eigenvectors within the allowable fluctuation range as periodic features for storage; The average rate of change of the joint eigenvector within the stationary window is analyzed by trend fitting and stored as a trend feature.

5. The method for postoperative psychological status assessment based on MPNFS theory according to claim 4, characterized in that: The steps of determining the trend change amount of the joint feature vector based on the trend feature, determining the periodic fluctuation amount of the joint feature vector based on the periodic feature, and determining the instantaneous change parameter of the joint feature vector based on the mutation feature include: Collect the joint feature vector of the current evaluation node and the demand forecast interval between the current node and the forecast node, input the joint feature vector, demand forecast interval and average change rate of the current node into the trend evaluation function, and output the trend change of the joint feature vector; Extract the periodic fluctuation amplitude of the joint eigenvector under each fluctuation period, perform weighted average processing, and output the periodic fluctuation amount of the joint eigenvector; Extract the mutation amount and number of mutations in each mutation window, determine the distribution segment of the mutation window according to the time interval between the mutation windows, perform weighted fusion processing on the mutation amount and number of mutations in each distribution segment, and output them as instantaneous change parameters; Among them, the closer the distribution segment is to the current evaluation node, the greater the corresponding weighted weight.

6. The method for postoperative psychological status assessment based on MPNFS theory according to claim 1, characterized in that: The steps of respectively performing confidence assessment on the trend change amount, the periodic fluctuation amount and the instantaneous change parameter include: Randomly extract multiple groups of sample data from existing historical data, perform simulation calculations on each group of sample data, and output simulated trend change, simulated period fluctuation and simulated instantaneous change parameters; Calculate the deviation rate between the simulated trend change amount and the trend change amount, and record it as a first condition parameter. Then compare the first condition parameter with a preset first threshold. When the first condition parameter is higher than the first threshold, directly determine that the trend change amount output is wrong, increase sample data and re-simulate the calculation. When the first condition parameter is lower than the first threshold, calculate the excess ratio of the first condition parameter and record it as a first confidence assessment indicator. Calculate the matching degree between the simulated periodic fluctuation amount and the periodic fluctuation amount, and record it as the second condition parameter. Then compare the second condition parameter with a preset second threshold value. When the second condition parameter is lower than the second threshold value, directly determine that the periodic fluctuation amount output is wrong, increase sample data and re-simulate the calculation. When the second condition parameter is higher than the second threshold value, calculate the excess ratio of the second condition parameter and record it as the second confidence evaluation indicator. Calculate the similarity between the simulated instantaneous change parameter and the instantaneous change parameter, and record it as the third condition parameter. Then compare the third condition parameter with the preset third threshold. When the third condition parameter is lower than the third threshold, directly increase the sample data and re-simulate the calculation. If the third condition parameter is still lower than the third threshold after adding the sample, it is determined that the instantaneous change parameter intervention is invalid, and no confidence assessment is performed; if the third condition parameter is higher than the third threshold, calculate its excess ratio and record it as the third confidence assessment indicator.

7. The method for postoperative psychological status assessment based on MPNFS theory according to claim 6, characterized in that: The step of allocating confidence weights to the trend change amount, the periodic fluctuation amount, and the instantaneous change amount according to the confidence evaluation result includes: Obtaining a first confidence evaluation index, a second confidence evaluation index, and a third confidence evaluation index; Inputting the first confidence index, the second confidence index, and the third confidence index into a preset weight distribution function, and using the output of the weight distribution function as the confidence weights of the trend change amount, the periodic fluctuation amount, and the instantaneous change amount; When the third confidence evaluation index is not output, the value of the third confidence evaluation index is assumed to be zero.

8. The method for postoperative psychological status assessment based on MPNFS theory according to claim 1, characterized in that: The step of performing comprehensive comparison and prediction on preset prediction nodes based on trend change amount, periodic fluctuation amount and instantaneous change parameter, as well as corresponding confidence weights, and outputting a psychological state score includes: Get the prediction node timestamp and the historical timestamp parallel to the prediction node timestamp; Collect the confidence weights of the first confidence evaluation indicator, the second confidence evaluation indicator, and the third confidence evaluation indicator at the historical timestamp, and arrange the confidence weights of the first confidence indicator, the second confidence indicator, and the third confidence indicator in descending order, and record the confidence weight with the highest ranking as the primary confidence weight; When the main confidence weight corresponds to the trend change amount, the predicted patient's mental state score is calculated based on the time interval between the prediction node timestamp and the historical timestamp; When the primary confidence weight corresponds to the periodic fluctuation, the predicted fluctuation ratio at the prediction node timestamp will be determined based on the periodic fluctuation at the historical timestamp, and the historical psychological state score at the historical timestamp will be proportionally adjusted based on the predicted fluctuation to obtain the psychological state score at the prediction node timestamp. When the main confidence weight corresponds to the instantaneous change, the historical psychological state score at the historical timestamp is recorded as the psychological state score at the prediction node timestamp.

9. A postoperative psychological status assessment system based on MPNFS theory, characterized by: The method for postoperative psychological state assessment based on the MPNFS theory according to any one of claims 1 to 8 comprises: The data collection module is used to collect the patient's postoperative psychological status assessment data under drug treatment, psychological intervention, clinical nursing, family care and social support. The psychological status assessment data includes respiratory monitoring data, voice recording data and bioelectric signal data; The time series analysis module is used to perform time-frequency synchronization processing on respiratory monitoring data, voice recording data, and bioelectric signal data, output a joint feature vector, and perform time series analysis on the joint feature vector through a sliding window mechanism to output the time series features of the joint feature vector. The time series features include trend features, periodic features, and mutation features. A feature quantification module is used to determine the trend change amount of the joint feature vector based on the trend feature, determine the periodic fluctuation amount of the joint feature vector based on the periodic feature, and determine the instantaneous change parameter of the joint feature vector based on the mutation feature; The weight calculation module is used to perform confidence assessment on the trend change amount, periodic fluctuation amount and instantaneous change parameters respectively, and assign confidence weights to the trend change amount, periodic fluctuation amount and instantaneous change amount according to the confidence assessment results; The state prediction module is used to perform comprehensive comparison and prediction of preset prediction nodes based on trend change, periodic fluctuation and instantaneous change parameters, as well as corresponding confidence weights, output a psychological state score, and issue an alarm signal when the psychological state score exceeds the preset assessment threshold, and generate a psychological state assessment report.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the postoperative mental state assessment method based on MPNFS theory as described in any one of claims 1 to 8.

Images