Psychological pressure detection method based on intelligent bed heart impact signal

Abstract

The invention discloses a psychological stress detection method based on an intelligent bed ballistocardiogram signal, which comprises the following steps: information collection: collecting basic characteristics, sleep data and psychological stress evaluation data of a sample object and a detected object, the sleep data comprising a ballistocardiogram signal; data processing: processing the ballistocardiogram signal to obtain model input characteristics; a psychological stress detection model is constructed, the psychological stress detection model comprises a feature extraction layer, a feature selection layer and a prediction layer, feature extraction is conducted on model input features through the feature extraction layer, the extracted features are selected through the feature selection layer, and prediction is conducted through the selected features; and psychological stress assessment: inputting the data of the tested object into the psychological stress detection model to obtain a psychological stress assessment result of the tested object. According to the psychological pressure detection method based on the intelligent bed ballistocardiogram signal, the ballistocardiogram signal is obtained in a non-contact mode, prediction is conducted through the psychological pressure prediction model, the prediction result is close to that of traditional contact type psychological pressure testing equipment, and then a psychological pressure monitoring means in a universal environment is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of psychological stress assessment methods, and in particular to a psychological stress detection method based on a smart bed cardiac impact signal. Background Art

[0002] Excessive mental stress will not only damage physical health, but will even directly lead to fatal diseases, but few people pay attention to the problems caused by stress in life. On the one hand, people's neglect of stress comes from the concept that they can eliminate the impact of stress through willpower; on the other hand, due to the lack of mental stress monitoring equipment available in a universal environment, people cannot realize that the body is objectively under stress. Ballistocardiogram (BCG) is a reliable physiological monitoring method with the characteristics of low cost, low professionalism, no interference and adaptability to long-term monitoring, which can meet the needs of daily mental stress monitoring.

[0003] According to the invention patent with publication number: CN113812949A and publication date: 2021-12-21, a psychological stress analyzer is disclosed, including a physiological parameter acquisition module, a signal filtering and amplification module, an analog-to-digital conversion module, a microprocessor, a power supply module and a display module; the physiological parameter acquisition module includes a heart rate sensor, an electroencephalogram sensor, a skin temperature sensor, a chest breathing sensor and a motion sensor, and the output end of each sensor is electrically connected to the input end of the signal filtering and amplification module; the output end of the signal filtering and amplification module is electrically connected to the input end of the analog-to-digital conversion module; the output end of the analog-to-digital conversion module is electrically connected to the input end of the microprocessor; the display module is electrically connected to the output end of the microprocessor; the power supply module is connected to the physiological parameter acquisition module, the signal filtering and amplification module, the analog-to-digital conversion module, the microprocessor and the display module to provide them with working voltage. Its main technical effects are: it can collect a variety of physiological parameters and better reflect the psychological stress state of the subject.

[0004] In the prior art, there is no detection method for estimating psychological stress using the ballistocardiogram. Therefore, a psychological stress detection method based on the ballistocardiogram signal of a smart bed is proposed, aiming to propose a method for estimating psychological stress using the ballistocardiogram. Summary of the invention

[0005] The purpose of the present invention is to provide a method for detecting psychological stress based on a ballistocardiogram signal of an intelligent bed, and to propose a method for estimating psychological stress using a ballistocardiogram.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a method for detecting psychological stress based on a smart bed's cardiac impact signal, comprising the following steps: Information collection, collecting the basic characteristics, sleep data, and psychological stress assessment data of the sample object and the object to be measured, where the sleep data includes ballistocardiogram signals; Data processing, processing the ballistocardiogram signals to obtain model input features; Constructing a psychological stress detection model, where the psychological stress detection model includes a feature extraction layer, a feature selection layer, and a prediction layer. Feature extraction is performed on the model input features through the feature extraction layer, feature selection is performed on the extracted features through the feature selection layer, and prediction is performed using the selected features; Psychological stress assessment, inputting the data of the object to be measured into the psychological stress detection model to obtain the psychological stress assessment result of the object to be measured.

[0007] Preferably, the data processing includes: Band-pass filtering, performing noise reduction processing on the ballistocardiogram signals through band-pass filtering to obtain cleaner heartbeat signals; Wavelet decomposition and coefficient extraction, using multi-Bessel wavelets to process the noise-reduced ballistocardiogram signals, decomposing the noise-reduced ballistocardiogram signals into detail signals and approximation signals of different frequencies, and extracting detail coefficients and approximation coefficients; Reconstructing the signal, using inverse wavelet transform to restore the detail coefficients to time-domain signals, thereby retaining the heartbeat signals; Peak detection, extracting the peaks of the ballistocardiogram signals; After extracting the peaks, calculate the RR intervals; After obtaining the RR intervals, perform heart rate variability analysis to obtain model input features.

[0008] The heart rate variability analysis includes time-domain analysis and frequency-domain analysis; The time-domain analysis is expressed as follows: ; where, is the number of RR intervals, is the th RR interval of the th heartbeat, ; where, is the number of RR intervals, is the th RR interval of the th heartbeat, ; where, is the number of RR intervals, is the th RR interval of the It is the average value of the RR interval.

[0009] Preferably, the frequency domain analysis is expressed as follows: Convert the RR interval signal into frequency components, evaluate the activities of the sympathetic and parasympathetic nerves under different frequency bandwidths, calculate the ratio of low-frequency power to high-frequency power, and measure the balance of sympathetic and parasympathetic nerve activities: ; Wherein, is the low-frequency power, is the high-frequency power.

[0010] Preferably, the peak detection specifically includes: Let the ballistocardiogram signal be , then in the interval , if and , then is a local maximum. During the peak detection process, first screen the local extrema to eliminate the pseudo-peaks caused by noise and retain the main peak related to the heartbeat, that is, the R-wave peak.

[0011] Preferably, the feature extraction layer includes a convolutional neural network and a long short-term memory network, and the feature selection layer includes principal component analysis and Shapley additive explanation analysis; The principal component analysis is specifically expressed as: ; Wherein, is the input data matrix, is the eigenvector matrix, is the data matrix after dimensionality reduction; Quantify the contribution of each feature to the prediction result through Shapley additive explanation analysis, expressed as: ; Wherein, is the Shapley value of feature , is the feature subset, is the model output when only including the feature subset .

[0012] Preferably, the prediction layer includes a support vector machine and an ensemble algorithm, and the decision function of the support vector machine is: ; Wherein, is the weight vector, is the input data, is the bias term.

[0013] Preferably, the integrated algorithm includes random forest and extreme gradient boosting tree; The random forest is specifically represented as: ; where is the prediction result of the th decision tree, is the number of decision trees; The loss function of the extreme gradient boosting tree is: ; where is the prediction error, is the regularization term, is the th decision tree.

[0014] Preferably, the sleep data is derived from the ballistocardiogram signal collected by the smart bed, and the psychological stress assessment data includes stress resistance, stress index, fatigue index, activity of the autonomic nervous system, balance of the autonomic nervous system, average heart rate, cardiac stability, and abnormal heart rate.

[0015] Preferably, the basic features include gender, age, height, weight, etc.

[0016] In the above technical solution, a psychological stress detection method based on the ballistocardiogram signal of a smart bed provided by the present invention has the following beneficial effects: In the present invention, psychological stress analysis is performed by collecting ballistocardiogram signals, and a psychological stress prediction model is constructed for psychological stress prediction. Moreover, the ballistocardiogram signals provided in the embodiments of the present invention are non-contact ballistocardiogram signals obtained by a smart bed. The prediction is performed through the psychological stress prediction model, and the prediction result is close to that of a traditional contact-type psychological stress testing device, thereby providing a means for monitoring psychological stress in a general environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0018] Figure 1 is a schematic diagram of the method provided by an embodiment of the present invention; Figure 2 is a schematic diagram of the overall process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] As Figure 1 - Figure 2 shown, a method for detecting psychological stress based on the impact signal of an intelligent bed core includes the following steps: Information collection, collecting the basic characteristics, sleep data, and psychological stress assessment data of the sample object and the object to be measured. The sleep data includes ballistocardiogram signals; As an embodiment provided by the present invention, the basic characteristics include information such as gender, age, height, and weight; The basic characteristics can be collected through methods such as mobile phone APP questionnaires.

[0021] As an embodiment provided by the present invention, the psychological stress assessment data includes stress resistance, stress index, fatigue index, activity of the autonomic nervous system, balance of the autonomic nervous system, average heart rate, cardiac stability, abnormal heart rate, etc. The sleep data and the psychological stress assessment instrument data are collected synchronously, with a sampling frequency of 500 Hz and a sampling time of 5 minutes for both.

[0022] As a specific embodiment provided by the present invention, the sleep data of a certain number of sample objects is collected. As an embodiment provided by the present invention, the number of sample objects collected is 150 - 250 people. The sleep data is ballistocardiogram signals, and the psychological stress assessment data includes stress resistance, stress index, fatigue index, activity of the autonomic nervous system, and cardiac stability.

[0023] The sleep data of the object to be measured is collected through an intelligent bed. It should be noted that collecting the ballistocardiogram signals of the user through an intelligent bed is a conventional technique in the prior art, and the implementation principle and technical details will not be elaborated here.

[0024] Data processing, processing the ballistocardiogram signals to obtain model input features; The data processing specifically includes: Band-pass filtering, performing noise reduction processing on the ballistocardiogram signals through band-pass filtering to obtain cleaner heartbeat signals; Collect the sleep data of about 150 - 250 people and the data of the psychological stress physiological assessment instrument using a Butterworth filter with a frequency range of 0.5 - 20 Hz. The data of the psychological stress physiological assessment instrument is used as the target feature during model construction.

[0025] As an embodiment provided by the present invention, the data collected by the psychological stress physiological assessment instrument is as follows: {'id': 001, 'anti-stress ability': 100,'stress index': 92, 'fatigue index': 99, 'activity of the autonomic nervous system (ANS activity)': 99, 'cardiac stability': 98}.

[0026] The transfer function of the band-pass filter is as follows: ; wherein, is the complex frequency variable, is the cut-off frequency, is the order of the filter.

[0027] Wavelet decomposition and coefficient extraction: Use the Daubechies Wavelet to process the ballistocardiogram signal after band-pass filtering, decompose the ballistocardiogram signal into detail signals and approximation signals of different frequencies, and extract the detail coefficients and approximation coefficients; The core idea of wavelet decomposition is to divide the signal into detail signals and approximation signals of different frequencies through layer-by-layer decomposition; As an embodiment provided by the present invention, 3-5 layers of wavelet decomposition are selected. Specifically, the coefficients of the j-th layer of decomposition are expressed as: ; ; wherein, is the approximation coefficient, is the detail coefficient, and are the coefficients of the low-pass and high-pass filters respectively, is the input signal.

[0028] Signal reconstruction: Adopt the inverse wavelet transform to restore the detail coefficients to the time-domain signal, thereby retaining the heartbeat signal, which is expressed as: ; wherein, is the approximation coefficient, is the detail coefficient, and are the coefficients of the low-pass and high-pass filters respectively, is the input signal.

[0029] Peak detection: Extract the peak value of the ballistocardiogram signal; The peak detection specifically includes: Let the ballistocardiogram signal be , then in the interval , if and , then This is a local maximum. During the peak detection process, first, false peaks caused by noise are eliminated by screening local extrema, and the main peak related to the heartbeat, i.e., the R-wave peak, is retained.

[0030] After the peak is extracted, the RR interval is calculated; After obtaining the R-wave peak, the RR interval can be calculated. The RR interval is the interval between adjacent heartbeats, and the calculation formula for the RR interval is: ; After obtaining the RR interval, heart rate variability analysis is used to obtain the model input features.

[0031] Heart rate variability analysis includes time-domain analysis and frequency-domain analysis; Specifically, the time-domain analysis is expressed as follows: ; Among them, is the number of RR intervals, is the th RR interval of the heartbeat, is the average value of the RR intervals; ; Among them, is the number of RR intervals, is the th RR interval of the heartbeat, is the average value of the RR intervals; ; Among them, is the number of RR intervals, is the th RR interval of the heartbeat, is the average value of the RR intervals; The frequency-domain analysis is expressed as follows: The RR interval signal is transformed into frequency components, the activities of the sympathetic and parasympathetic nerves under different frequency bandwidths are evaluated, and the ratio of low-frequency power to high-frequency power is calculated to measure the balance of sympathetic and parasympathetic nerve activities: ; Among them, is the low-frequency power, is the high-frequency power.

[0032] The correlation between the heart rate variability calculated from the intelligent bed ballistocardiogram signal and the heart rate variability of the psychological stress analyzer is between 0.84 and 0.95, showing a high degree of consistency.

[0033] Construct a psychological stress detection model. The psychological stress detection model includes a feature extraction layer, a feature selection layer, and a prediction layer. The feature extraction layer extracts features from the input features of the model, the feature selection layer selects the extracted features, and the selected features are used for prediction. Taking the data of the psychological stress physiological assessment instrument as the target features and the features calculated from the ballistocardiogram signals as the feature variables, the model is constructed and trained accordingly.

[0034] Specifically, the feature extraction layer includes a convolutional neural network and a long short-term memory network, the feature selection layer includes principal component analysis and Shapley additive explanation analysis, and the prediction layer includes a support vector machine (Support Vector Machine, SVM) and an ensemble algorithm; The convolutional neural network is used for convolution operation, expressed as: ; Among them, is the input signal, is the convolution kernel, is the output feature map.

[0035] The convolutional neural network is suitable for processing data with obvious spatio-temporal characteristics. When processing heart rate variability biosignals, the convolutional layer of the convolutional neural network can capture the local features of the signal through the local receptive field, and reduce redundant information through the pooling layer after extracting these features, thereby improving the computational efficiency of the model.

[0036] Through the convolution operation, the convolutional neural network can automatically learn the local features of the signal, and these features can contain the key patterns in the heart rate variability biosignals, such as the periodic changes and waveform features of the heartbeat signal. Then the long short-term memory network is used to capture the temporal information. The psychological stress state has dynamic characteristics, and the long short-term memory network is an ideal choice for processing temporal data. The long short-term memory network can retain information over a long time span through its special "memory cell" structure, avoiding the gradient vanishing problem of traditional recurrent neural networks.

[0037] The long short-term memory network is used to capture the temporal information. The core equations of the long short-term memory network include the forget gate, the input gate, and the output gate.

[0038] Among them, the forget gate controls which information needs to be forgotten, the input gate determines which information needs to be updated at the current moment, and the output gate determines the current output.

[0039] The forget gate is expressed as: ; Among them, is the activation function, usually the sigmoid function, which compresses the output value between 0 and 1. is the weight matrix of the forget gate, which is used to process the hidden state at the previous moment and the input at the current moment for linear transformation. represents the hidden state at the previous moment. represents the input at the current moment. is the bias term of the forget gate.

[0040] The input gate is expressed as: ; where is the weight matrix of the input gate. Similar to the forget gate, it is used for linear transformation. is the bias term of the input gate.

[0041] ; where tanh is the hyperbolic tangent activation function, which compresses the output value between -1 and 1. is the weight matrix of the candidate cell state layer. is the bias term of the candidate cell state layer.

[0042] The output gate is expressed as: ; where is the weight matrix of the output gate. is the bias term of the output gate.

[0043] Through these gating mechanisms, the long short-term memory network can effectively capture the dynamic changes of the heart rate variability signal over time, so as to better simulate the gradual changes of mental stress.

[0044] After extracting features through the convolutional neural network and the long short-term memory network, they are input into the feature selection layer; Specifically, the feature selection layer includes principal component analysis and Shapley additive explanation analysis; Principal component analysis is specifically expressed as: ; where is the input data matrix, is the feature vector matrix, is the data matrix after dimensionality reduction; Principal component analysis can reduce the feature dimension while retaining most of the data information. Through linear transformation, principal component analysis projects the original features onto a few principal components, thereby reducing the dimension of the data, accelerating model training, and reducing the risk of overfitting at the same time.

[0045] Quantify the contribution of each feature to the prediction result through Shapley Additive exPlanations (SHAP) analysis, expressed as: ; where is the Shapley value of feature , is the feature subset, is the model output when only including the feature subset .

[0046] Through Shapley Additive exPlanations analysis, it is possible to determine which features are the most important in psychological stress prediction, thereby further optimizing the performance of the model. Through principal component analysis and Shapley Additive exPlanations analysis, redundant information can be effectively removed, providing higher efficiency for the classification model.

[0047] The prediction layer includes a support vector machine and an ensemble algorithm.

[0048] Specifically, the decision function of the support vector machine is: ; where is the weight vector, is the input data, is the bias term.

[0049] As an embodiment provided by the present invention, the ensemble algorithm includes a random forest and Extreme Gradient Boosting (XGBoost).

[0050] The random forest is specifically expressed as: ; where is the prediction result of the th decision tree, is the number of decision trees; The random forest is an ensemble algorithm based on decision trees, which effectively reduces overfitting by randomly selecting features and samples for training.

[0051] The loss function of Extreme Gradient Boosting is: ; where is the prediction error, is the regularization term, is the th decision tree. Extreme Gradient Boosting can not only improve the classification accuracy but also prevent overfitting through regularization.

[0052] The support vector machine classifies by finding a hyperplane that maximizes the class margin. In high-dimensional spaces, the support vector machine exhibits strong classification ability, and its advantage lies in its ability to maintain high generalization ability when dealing with high-dimensional data. Then, an ensemble algorithm is used to improve the performance of the overall model by combining multiple weak classifiers.

[0053] After building the model, evaluate the prediction accuracy of the model; Use a confusion matrix to evaluate the prediction accuracy of the model. The relevant formula is as follows: A confusion matrix for classification : ; Accuracy: ; TP (True Positives) is the value on the diagonal (the number of correctly predicted samples); Total Samples is the sum of all values in the confusion matrix.

[0054] Precision: ; FP (False Positives) is the number of samples predicted as class i but actually not belonging to class i.

[0055] Recall: ; FN (False Negatives): The number of samples that actually belong to class i but are not predicted as class i.

[0056] F1 Score: ; Use the validation dataset of the psychological stress physiological assessment instrument to estimate the accuracy.

[0057] The estimated metrics include stress index, stress resistance, fatigue index, ANS activity, and cardiac stability.

[0058] The prediction accuracy of the psychological stress detection model is between 75.22% and 93.04%, with a relatively high accuracy. It can realize the evaluation of psychological stress based on the BCG signal during the human sleep period.

[0059] For psychological stress assessment, input the data of the measured object into the psychological stress detection model to obtain the psychological stress assessment result of the measured object.

[0060] As an embodiment provided by the present invention, the psychological stress assessment result is transmitted through data to a mobile phone APP for display.

[0061] In the present invention, the ballistocardiogram signal is collected for psychological stress analysis, and a psychological stress prediction model is constructed for psychological stress prediction. Moreover, the ballistocardiogram signal provided by the embodiment of the present invention is a non-contact ballistocardiogram signal obtained by an intelligent bed. The prediction result is close to that of a traditional contact-type psychological stress testing device through the psychological stress prediction model, thereby providing a monitoring means for psychological stress in a general environment.

[0062] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0063] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0064] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0065] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1Steps of the functions specified in one or more boxes.

[0066] In the present invention, specific embodiments are used to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

[0067] The embodiments of the present application also provide a specific implementation manner of an electronic device that can implement all the steps in the method in the above embodiments. The electronic device specifically includes the following: A processor, a memory, a communication interface, and a bus; Wherein, the processor, the memory, and the communication interface complete mutual communication through the bus; The processor is used to call the computer program in the memory. When the processor executes the computer program, it implements all the steps in the method in the above embodiments. For example, when the processor executes the computer program, it implements the following steps: Information collection, collecting the basic characteristics, sleep data, and psychological stress assessment data of the sample object and the object to be measured. The sleep data includes ballistocardiogram signals; Data processing, processing the ballistocardiogram signals to obtain model input features; Constructing a psychological stress detection model. The psychological stress detection model includes a feature extraction layer, a feature selection layer, and a prediction layer. The model input features are extracted through the feature extraction layer, the extracted features are selected through the feature selection layer, and predictions are made using the selected features; Psychological stress assessment, inputting the data of the object to be measured into the psychological stress detection model to obtain the psychological stress assessment result of the object to be measured.

[0068] The embodiments of the present application also provide a computer-readable storage medium that can implement all the steps in the method in the above embodiments. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, it implements all the steps in the method in the above embodiments. For example, when the processor executes the computer program, it implements the following steps: Information collection, collecting the basic characteristics, sleep data, and psychological stress assessment data of the sample object and the object to be measured. The sleep data includes ballistocardiogram signals; Data processing, processing the ballistocardiogram signals to obtain model input features; Construct a psychological stress detection model, where the psychological stress detection model includes a feature extraction layer, a feature selection layer, and a prediction layer. The feature extraction layer extracts features from the input features of the model, the feature selection layer selects the extracted features, and prediction is performed using the selected features; Psychological stress assessment: Input the data of the object to be measured into the psychological stress detection model to obtain the psychological stress assessment result of the object to be measured.

[0069] Each embodiment in this specification is described in a progressive manner. For the parts that are the same or similar among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the hardware + program type of embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content. Although the method operation steps as described in the embodiments of this specification are provided, more or fewer operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiments is only one way among the execution orders of numerous steps and does not represent the only execution order. When the actual device or terminal product executes, it can be executed in the order shown in the embodiments or the drawings or in parallel (for example, in an environment of parallel processors or multi-threaded processing, or even in a distributed data processing environment). The term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, product or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, product or device. Without further limitations, it does not exclude the existence of additional identical or equivalent elements in the process, method, product or device including the said elements. For the convenience of description, the above device is described by dividing it into various modules according to functions. Of course, when implementing the embodiments of this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the modules implementing the same function can be realized by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be in electrical, mechanical or other forms. The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks

[0070] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. In the description of this specification, the description of reference terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of this specification.

[0071] In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples. The above is only the embodiments of the embodiments of this specification and is not used to limit the embodiments of this specification. For those skilled in the art, various changes and modifications can be made to the embodiments of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of this specification shall be included within the scope of the claims of the embodiments of this specification.

Claims

1. A psychological stress detection method based on the heartbeat signal of an intelligent bed, characterized in that: The following steps are involved: Information collection, collecting basic characteristics, sleep data and psychological stress assessment data of the sample object and the measured object, wherein the sleep data includes a ballistocardiogram signal; Data processing: processing the ballistocardiogram signal to obtain model input features; Constructing a psychological stress detection model, the psychological stress detection model includes a feature extraction layer, a feature selection layer and a prediction layer, extracting model input features through the feature extraction layer, selecting the extracted features through the feature selection layer, and using the selected features for prediction; Psychological stress assessment: input the data of the subject to be tested into the psychological stress detection model to obtain the psychological stress assessment results of the subject to be tested.

2. The method for detecting psychological stress based on the intelligent bed cardiac shock signal according to claim 1, characterized in that: The data processing includes: Bandpass filtering: noise reduction is performed on the ballistocardiogram signal through bandpass filtering to obtain a cleaner heartbeat signal; Wavelet decomposition and coefficient extraction: using Dobesi wavelet to process the denoised ballistocardiogram signal, decomposing the denoised ballistocardiogram signal into detail signals and approximation signals of different frequencies, and extracting detail coefficients and approximation coefficients; Reconstruct the signal, using inverse wavelet transform to restore the detail coefficients to time domain signals, thereby retaining the heartbeat signal; Peak detection, extracting the peak value of the ballistocardiogram signal; After extracting the peak value, calculate the RR interval; After obtaining the RR interval, heart rate variability analysis was used to obtain the model input features.

3. The psychological stress detection method based on the smart bed heart shock signal according to claim 2 is characterized in that: The heart rate variability analysis includes time domain analysis and frequency domain analysis; The time domain analysis includes , and , which is expressed as follows: ； in, is the number of RR intervals, It is The RR interval of a heartbeat, is the average value of the RR interval; ； in, is the number of RR intervals, It is The RR interval of a heartbeat, is the average value of the RR interval; ； in, is the number of RR intervals, It is The RR interval of a heartbeat, is the average value of the RR interval.

4. The method for detecting psychological stress based on the intelligent bed cardiac shock signal according to claim 3, characterized in that: The frequency domain analysis is expressed as follows: Convert the RR interval signal into frequency components, evaluate the activity of sympathetic and parasympathetic nerves at different frequency bandwidths, calculate the ratio of low-frequency power to high-frequency power, and measure the balance of sympathetic and parasympathetic nerve activity: ； in, is the low frequency power, is high frequency power.

5. The method for detecting psychological stress based on the intelligent bed cardiac shock signal according to claim 2, characterized in that: The peak detection specifically includes: Assume that the ballistocardiogram signal is , then in the interval Up, if and ,but In the peak detection process, the pseudo peaks caused by noise are firstly eliminated by screening the local extreme values, and the main peak related to the heartbeat, that is, the R wave peak, is retained.

6. The method for detecting psychological stress based on the intelligent bed cardiac shock signal according to claim 1, characterized in that: The feature extraction layer includes a convolutional neural network and a long short-term memory network, and the feature selection layer includes principal component analysis and Shapley additive interpretation analysis; The principal component analysis is specifically expressed as: ； in, is the input data matrix, is the eigenvector matrix, is the data matrix after dimensionality reduction; The contribution of each feature to the prediction result is quantified by Shapley additive interpretation analysis, expressed as: ； in, It is a feature The Shapley value of is a feature subset, Contains only a subset of features The model output at .

7. The method for detecting psychological stress based on the intelligent bed cardiac shock signal according to claim 1, characterized in that: The prediction layer includes a support vector machine and an integrated algorithm, and the decision function of the support vector machine algorithm is: ； in, is the weight vector, is the input data, is the bias term.

8. The method for detecting psychological stress based on the intelligent bed's cardiac shock signal according to claim 7, characterized in that: The ensemble algorithms include random forests and extreme gradient boosted trees; The random forest is specifically expressed as: ； in, It is The prediction results of a decision tree, is the number of decision trees; The loss function of the extreme gradient boosting tree is: ； in, is the prediction error, is the regularization term, It is A decision tree.

9. The method for detecting psychological stress based on the intelligent bed cardiac shock signal according to claim 1, characterized in that: The sleep data is derived from the ballistocardiogram signal collected by the smart bed, and the psychological stress assessment data includes stress resistance, stress index, fatigue index, activity of the autonomic nervous system, balance of the autonomic nervous system, average heart rate, heart stability, and abnormal heart rate.

10. The method for detecting psychological stress based on the intelligent bed's cardiac shock signal according to claim 1, characterized in that: The basic characteristics include gender, age, height, weight, etc.

Images