Depression quantitative evaluation method and system based on long-time-history electrocardio data

Abstract

The invention provides a depression quantitative evaluation method and system based on long-time-history electrocardio data, and belongs to the technical field of auxiliary diagnos.The depression quantitative evaluation method specifically comprises the steps that credible signals are determined according to the duration time of a credible period of a user to be screened; determining the depression degree evaluation value of the user to be screened according to the time domain analysis parameter and the frequency domain analysis parameter of the credible signal, and taking the set time as a test period; obtaining the number of credible signals of which the depression degree evaluation values do not meet the requirements of the to-be-screened user in a plurality of test periods in a set time period and depression degree evaluation values of different credible signals; and combining the number of the credible signals of which the depression degree evaluation values of the to-be-screened user in the set time do not meet the requirement and the depression degree evaluation values of different credible signals to obtain the depression degree comprehensive evaluation value of the to-be-screened user, so that quantitative evaluation of the depression degree of the user is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of auxiliary diagnosis, and in particular relates to a method and system for quantitatively evaluating depression based on long-term electrocardiogram data. Background Art

[0002] Depression is the most common mental illness nowadays, with continuous and long-term low mood as the main clinical feature. Clinically, it can be seen that low mood and unhappiness in reality, long-term low mood, from the initial depression to the final grief, inferiority, pain, pessimism, world-weariness, feeling that every day of life is desperately torturing oneself, negative, escapist, and even suicidal tendencies and behaviors.

[0003] In order to achieve a quantitative assessment of depression, the prior art discloses a quantitative assessment of depression based on electrocardiogram analysis. Specifically, in the authorized invention patent CN114027842B "A system, method and device for objective screening of depression", an electrocardiogram signal is obtained from an electrocardiogram signal acquisition device, and the heart rate parameters and HRV characteristic parameters of the electrocardiogram signal are input into a depression screening model to obtain the depression risk index of the subject. However, there are the following technical problems:

[0004] Existing technical solutions often use a single ECG signal at a certain moment. However, for a specific user, the ECG signal at a certain moment may have emotional fluctuations, so using a single ECG signal at a certain moment often cannot accurately reflect the actual depression risk of the subject.

[0005] In response to the above technical problems, the present invention provides a method and system for quantitatively evaluating depression based on long-term electrocardiogram data. Summary of the invention

[0006] The purpose of the present invention is to provide a hospital diagnosis and treatment information system that is domestically produced and independently controllable.

[0007] In order to solve the above technical problems, the present invention provides a method for quantitatively evaluating depression based on long-term electrocardiogram data, which is characterized by specifically comprising:

[0008] S11 obtains electrocardiogram data of the user to be screened within a set time, and divides the electrocardiogram data into reliable periods and abnormal periods according to waveform characteristics;

[0009] S12 determines the credible signal according to the duration of the credible period of the user to be screened, and evaluates the credibility of the electrocardiogram data according to the time length and number of credible signals of the user to be screened within the set time, the credible period, and the number of abnormal periods, and when the credibility of the electrocardiogram data meets the requirements, proceeds to the next step;

[0010] S13 determines the depression evaluation value of the to-be-screened user according to the time domain analysis parameters and the frequency domain analysis parameters of the credible signal, and when the number of credible signals whose depression evaluation values ​​of the to-be-screened user do not meet the requirements is greater than a preset number, proceeds to the next step;

[0011] S14 takes the set time as the test period, obtains the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements in multiple test periods within the set time period and the depression assessment values ​​of different credible signals, and calculates a comprehensive depression assessment value of the user to be screened based on the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements within the set time and the depression assessment values ​​of different credible signals.

[0012] On the other hand, the present application provides a depression quantitative assessment system based on long-term electrocardiogram data, specifically comprising:

[0013] Period division module, credibility assessment module, credibility signal determination module, depression assessment module;

[0014] The cycle division module is responsible for obtaining the electrocardiogram data of the user to be screened within a set time, and dividing the electrocardiogram data into reliable cycles and abnormal cycles according to waveform characteristics;

[0015] The credibility evaluation module is responsible for determining the credible signal according to the duration of the credible period of the user to be screened, and evaluating the credibility of the electrocardiogram data according to the time length and number of the credible signals of the user to be screened within the set time, the credible period and the number of abnormal periods;

[0016] The credible signal determination module is responsible for determining the depression evaluation value of the to-be-screened user according to the time domain analysis parameters and the frequency domain analysis parameters of the credible signal, and determining whether the number of credible signals whose depression evaluation values ​​of the to-be-screened user do not meet the requirements is greater than a preset number;

[0017] The comprehensive depression assessment value is responsible for taking the set time as the test period, obtaining the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements in multiple test periods within the set time period, as well as the depression assessment values ​​of different credible signals, and combining the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements within the set time, as well as the depression assessment values ​​of different credible signals to derive the comprehensive depression assessment value of the user to be screened.

[0018] The beneficial effects of the present invention are:

[0019] By screening the credible cycles and abnormal cycles, the influence of the electrocardiogram data of the user in motion on the accuracy of the depression assessment of the screened user is avoided, and the accuracy of the depression assessment results is further guaranteed.

[0020] The credibility of the electrocardiogram data is evaluated by the time length and number of credible signals, credible cycles and the number of abnormal cycles of the user to be screened within the set time, thereby avoiding the influence of credibility due to small amount of credible signal data or small amount of credible cycle data, further improving the accuracy of credibility assessment, avoiding unnecessary depression assessment, and improving processing efficiency.

[0021] By comprehensively considering the depression assessment results of multiple test periods within a time period, the technical problem of inaccurate assessment results caused by single consideration of the depression assessment results of a certain day is avoided, and the accuracy and stability of the assessment results are further improved.

[0022] Other features and advantages will be described in the following description, and partly become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0025] Figure 1 This is a flowchart of a method for quantitatively evaluating depression based on long-term electrocardiogram data according to Example 1.

[0026] Figure 2 This is a framework diagram of a depression quantitative assessment system based on long-term electrocardiogram data in Example 2.

[0027] Figure 3 This is a flowchart of a specific method for evaluating the reliability of electrocardiogram data in Example 2.

[0028] Figure 4 This is a flowchart of the Self-Training data labeling process in Example 2.

[0029] Figure 5 This is a flow chart of the Self-Training data labeling principle in Example 2. DETAILED DESCRIPTION

[0030] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0031] The terms "a", "an", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "comprising" and "having" are used to express an open-ended inclusive meaning and mean that additional elements / components / etc. may be present in addition to the listed elements / components / etc.

[0032] Example 1

[0033] In order to solve the above technical problems, Figure 1 As shown, the present invention provides a method for quantitatively evaluating depression based on long-term electrocardiogram data, which is characterized by specifically comprising:

[0034] S11 obtains electrocardiogram data of the user to be screened within a set time, and divides the electrocardiogram data into reliable periods and abnormal periods according to waveform characteristics;

[0035] In this embodiment, by screening the credible cycles and abnormal cycles, the influence of the electrocardiogram data of the user in motion on the accuracy of the depression assessment of the screened user is avoided, and the accuracy of the depression assessment result is further guaranteed.

[0036] S12 determines the credible signal according to the duration of the credible period of the user to be screened, and evaluates the credibility of the electrocardiogram data according to the time length and number of credible signals of the user to be screened within the set time, the credible period, and the number of abnormal periods, and when the credibility of the electrocardiogram data meets the requirements, proceeds to the next step;

[0037] In this embodiment, the credibility of the electrocardiogram data is evaluated based on the time length and number of credible signals, credible cycles and the number of abnormal cycles of the user to be screened within the set time, thereby avoiding the impact of the credibility due to the small amount of credible signal data or the small amount of credible cycle data, further improving the accuracy of the credibility assessment, and also avoiding unnecessary depression assessment, thereby improving processing efficiency.

[0038] S13 determines the depression evaluation value of the to-be-screened user according to the time domain analysis parameters and the frequency domain analysis parameters of the credible signal, and when the number of credible signals whose depression evaluation values ​​of the to-be-screened user do not meet the requirements is greater than a preset number, proceeds to the next step;

[0039] S14 takes the set time as the test period, obtains the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements in multiple test periods within the set time period and the depression assessment values ​​of different credible signals, and calculates a comprehensive depression assessment value of the user to be screened based on the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements within the set time and the depression assessment values ​​of different credible signals.

[0040] In this embodiment, by comprehensively considering the depression assessment results of multiple test periods within a time period, the technical problem of inaccurate assessment results caused by considering only the depression assessment results of a certain day is avoided, and the accuracy and stability of the assessment results are further improved.

[0041] Example 2

[0042] On the other hand, Figure 2 As shown, the present application provides a depression quantitative assessment system based on long-term electrocardiogram data, which specifically includes:

[0043] Period division module, credibility assessment module, credibility signal determination module, depression assessment module;

[0044] The cycle division module is responsible for obtaining the electrocardiogram data of the user to be screened within a set time, and dividing the electrocardiogram data into reliable cycles and abnormal cycles according to waveform characteristics;

[0045] The trusted cycle and abnormal cycle referred to in the present invention are both a complete heartbeat cycle of a normal user.

[0046] Specifically, when the credibility of the electrocardiogram data meets the requirements, the depression assessment value of the user to be screened is determined; when the number of credible signals whose depression assessment value of the user to be screened does not meet the requirements is greater than a preset number, the comprehensive depression assessment value of the user to be screened is determined.

[0047] It should be noted that the value range of the set time is at least more than 24 hours.

[0048] It can be understood that the electrocardiogram data is divided into reliable cycles and abnormal cycles according to the waveform characteristics, specifically including:

[0049] The cardiac cycles of the electrocardiogram data are divided according to the waveform characteristics of the electrocardiogram data, and whether the user to be screened is in a state of intense activity during the cardiac cycle is determined according to the waveform characteristics of different cardiac cycles of the electrocardiogram data. If so, the cardiac cycle is determined to be an abnormal cycle, and if not, the cardiac cycle is determined to be a credible cycle.

[0050] The credibility evaluation module is responsible for determining the credible signal according to the duration of the credible period of the user to be screened, and evaluating the credibility of the electrocardiogram data according to the time length and number of the credible signals of the user to be screened within the set time, the credible period and the number of abnormal periods;

[0051] Furthermore, when there are continuous trusted cycles greater than a set number threshold or the duration of continuous trusted cycles is greater than a preset time threshold in the electrocardiogram data of the user to be screened, the electrocardiogram data corresponding to the continuous trusted cycles greater than the set number threshold or the continuous trusted cycles greater than the preset time threshold are taken as trusted signals.

[0052] Specifically, Figure 2 As shown, the specific method for evaluating the credibility of the electrocardiogram data is:

[0053] S21 obtains the number of trustworthy cycles of the user to be screened within the set time, and determines whether the credibility of the electrocardiogram data meets the requirement according to the number of trustworthy cycles. If so, proceeds to the next step; if not, determines that the credibility of the electrocardiogram data does not meet the requirement;

[0054] S22 obtains the time length and number of the credible signals of the to-be-screened user within the set time, and determines whether the credibility of the electrocardiogram data meets the requirements according to the time length and number of the credible signals of the to-be-screened user within the set time, if yes, proceeds to the next step, if not, determines that the credibility of the electrocardiogram data does not meet the requirements;

[0055] S23: evaluating the credibility of the trusted cycle of the user to be screened according to the number of trusted cycles and the number of abnormal cycles of the user to be screened within the set time; determining a stable trusted signal according to the duration of the trusted signal of the user to be screened within the set time, and evaluating the credibility of the trusted signal of the user to be screened according to the time length and number of the stable trusted signal, the time length and signal of the trusted signal;

[0056] S24 evaluates the credibility of the electrocardiogram data of the user to be screened according to the credibility of the credible signal of the user to be screened and the credibility of the credible cycle.

[0057] Furthermore, when the duration of the credible signal is greater than a set time reference value, the credible signal is determined to be a stable credible signal, wherein the time length of the stable credible signal is determined based on the sum of the durations of all credible signals of the user to be screened within the set time.

[0058] The credible signal determination module is responsible for determining the depression evaluation value of the to-be-screened user according to the time domain analysis parameters and the frequency domain analysis parameters of the credible signal, and determining whether the number of credible signals whose depression evaluation values ​​of the to-be-screened user do not meet the requirements is greater than a preset number;

[0059] It should be noted that the method for determining the depression evaluation value of the user to be screened is:

[0060] Data preprocessing: preprocessing the credible signal through a baseline removal filter and a low-pass filter, and performing peak transformation on the preprocessed data;

[0061] 1) Data preprocessing: First, the reliable signal is preprocessed by removing the baseline filter and the low-pass filter.

[0062] The calculation formula for removing the baseline filter is:

[0063]

[0064] Among them, x is the credible signal data, f(i) is the data after baseline removal filtering, and n is the filtering window.

[0065] The low-pass filter is mainly used to filter out interference signals such as power frequency interference and machine interference. The filter adopts FIR equal ripple design, where the passband cutoff frequency is 35Hz, the stopband cutoff frequency is 45Hz, the passband attenuation coefficient is 0.2dB, the stopband attenuation coefficient is 55dB, and the filter order is 59th.

[0066] Then, the peak value transformation is performed on the filtered data. The peak value transformation mainly highlights the R wave of the ECG waveform and suppresses other waveforms. The specific formula of the peak value transformation is:

[0067]

[0068] Among them, x is the ECG data after filtering, y(i) is the data after peak transformation, and w is the peak transformation window (the window is set to an empirical value, which is set to 20 milliseconds here).

[0069] Feature identification and extraction of feature parameters: The R wave position of the credible signal is found by the threshold method for the peak transformed data. After the R wave position is found, the front and rear windows are set at the corresponding positions in the preprocessed data, and the start and end positions of the P, Q, R, S, and T waves and the corresponding peak positions are found by the threshold method;

[0070] Specifically, the corresponding parameters are extracted from the result of feature recognition:

[0071] QT_list: QT interval parameter array, which refers to the time from the beginning of cardiac depolarization to the end of repolarization, that is, the time from the beginning of the QRS wave to the end of the T wave;

[0072] RR_list: RR interval parameter array, which refers to the time between two R waves;

[0073] HR_list: Heart rate trend chart.

[0074] Heart rate variability parameter calculation: heart rate variability parameters are calculated according to the start and end positions of the P, Q, R, S, and T waves of the credible signal and the corresponding peak positions to obtain the time domain analysis parameters and frequency domain analysis parameters of the credible signal;

[0075] i) Time domain analysis parameters:

[0076] SDNN: standard deviation of normal RR interval;

[0077] RMSSD: root mean square difference between adjacent normal RR intervals;

[0078] NN50: the number of heartbeats with a difference of more than 50 milliseconds between adjacent normal RR intervals in the whole recording;

[0079] PNN50: NN50 divided by the percentage of the total number of normal RR intervals.

[0080] ii) Frequency domain analysis parameters:

[0081] VLF: power in the frequency range of 0.003Hz to 0.04Hz;

[0082] LF: power in the frequency range of 0.04Hz to 0.15Hz;

[0083] HF: power in the frequency range of 0.15 Hz to 0.4 Hz;

[0084] LF_HF: ratio of LF to HF;

[0085] W: Total power, the variable power of all normal RR intervals with a frequency less than 0.4 Hz.

[0086] iii) Nonlinear analysis parameters:

[0087] Lorenz_plot: Poincare scatter plot, also known as Lorenz scatter plot, is a scatter plot formed by plotting all normal cardiac cycles in sequence, with the previous one of two adjacent cardiac cycles as the horizontal coordinate and the next cardiac cycle as the vertical coordinate as a point;

[0088] SD1: short axis of the scatter plot, representing the magnitude of the difference between adjacent NN intervals;

[0089] SD2: is the long axis of the scatter plot, representing the magnitude of changes in heart rate over a long period of time;

[0090] MP_plot: Scattered point difference plot, a scatter plot plotted with the difference between two adjacent cardiac cycles, with the first difference as the horizontal axis and the second difference as the vertical axis, plotting all cardiac cycles in sequence. According to the positive and negative values ​​of the difference, the scattered points are distributed in the four quadrants of the rectangular vertical axis, reflecting the discreteness of the RR difference.

[0091] Depression evaluation value output: According to the time domain analysis parameters and frequency domain analysis parameters of the credible signal, an evaluation model based on a gradient enhancement algorithm is used to determine the depression evaluation value of the user to be screened.

[0092] Furthermore, before outputting the depression assessment value, the model needs to be trained, the data set needs to be integrated, and the depression assessment value needs to be determined after the integrated data set is trained on the model.

[0093] Specifically, the integrated data set includes:

[0094] The data set mainly consists of ECG data feature parameters and heart rate variability parameters, which are segmented using the idea of ​​total-part. The ECG data feature parameters include frequency domain analysis parameters and frequency domain analysis parameters. The specific method is as follows:

[0095] Total: Process the analyzed data as a whole. Specific parameters include QT average, RR interval average, average heart rate, QT_list, RR_list, HR_list, SDNN, RMSSD, NN50, PNN50, VLF, LF, HF, LF_HF, W, Lorenz_plot, SD1, SD2, and MP_plot.

[0096] Division: divide the data according to different time windows and then integrate them. The time windows are divided into 5 minutes and 1 hour. The specific parameters include:

[0097] i) 5-minute window: QT mean, RR interval mean, mean heart rate, SDNN, RMSSD, NN50, PNN50, VLF, LF, HF, LF_HF, W;

[0098] ii) 1 hour window: QT average, RR interval average, average heart rate, QT_list, RR_list, HR_list, SDNN, RMSSD, NN50, PNN50, VLF, LF, HF, LF_HF, W;

[0099] Among them, Lorenz_plot and MP_plot are image data, and the others are parameter data.

[0100] Furthermore, the specific steps for constructing the evaluation model of the gradient boosting algorithm are:

[0101] The depression assessment value adopts the gradient boosting algorithm (GBDT), which improves the performance of the existing model by iteratively adding weak learners (such as decision trees), and uses the negative gradient of the loss function to guide the next learning direction.

[0102] GBDT is an ensemble learning method based on gradient boosting. It uses multiple regression trees as base learners and improves the predictive ability of the model by iteratively fitting the negative gradient of the loss function (also called residual or pseudo residual).

[0103] The basic process of GBDT regression algorithm is as follows:

[0104] 1) Initialize a constant as the first base learner, usually the mean of the training data labels or the value that minimizes the loss function.

[0105]

[0106] Among them, L is the loss function, y i is the true label and γ is a constant.

[0107] The loss function L is the multiple linear models learned Converted to the probability of the corresponding classification, each Corresponding to a classification category, the specific calculation formula is:

[0108]

[0109] For each round of iteration, the negative gradient (residual) of each sample is calculated as the new response variable. The number of iterations is m = 1, 2, 3, ..., M. The specific process is as follows:

[0110] i) For sample i=1, 2, 3, ..., N, calculate the negative gradient:

[0111]

[0112] ii) Use the regression tree to fit the new response variable and obtain a regression tree, dividing its leaf nodes into several regions γ jm , j = 1, 2, 3, ..., J m ;

[0113] iii) For j=1, 2, 3, ..., J mCalculate the best fit value, usually the residual mean of all samples in the region or the value that minimizes the loss function:

[0114]

[0115] iv) Update the current model, multiply the regression tree obtained in this round by a step size (learning rate) and add it to the previous model:

[0116]

[0117] v) Repeat steps i-iv until the preset number of iterations is reached or the model performance no longer improves.

[0118] 3) Output the final model expression:

[0119]

[0120] The comprehensive depression assessment value is responsible for taking the set time as the test period, obtaining the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements in multiple test periods within the set time period, as well as the depression assessment values ​​of different credible signals, and combining the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements within the set time, as well as the depression assessment values ​​of different credible signals to derive the comprehensive depression assessment value of the user to be screened.

[0121] It should also be noted that after the evaluation model is successfully built, the parameters of the evaluation model need to be self-updated so that the accuracy of the final evaluation model can be improved.

[0122] As subsequent data sets gradually accumulate, in order to achieve a more optimized evaluation model, the existing model needs to be continuously updated and iterated. This method uses the semi-supervised learning method Self-Training.

[0123] Self-Training is to first train a weak classifier with a small amount of labeled data, and then use this weak classifier to label the unlabeled samples. When certain conditions are met (the predicted probability is greater than the threshold), the predicted result is used as the true label of the sample. Then, the model is trained again with the current labeled data and the unlabeled samples are labeled and iterated repeatedly until the stopping condition is reached.

[0124] like Figure 4As described above, the first step is to train a weak classifier with a small amount of labeled data; the second step is to use this weak classifier to predict unlabeled data samples, and use the predicted partial labels as the true labels of the samples according to a certain strategy; the third step is to use all current labeled samples (including samples processed in the second step) to train the classifier, and if the stopping condition is met, enter the fourth step, otherwise continue to enter the second step to predict the unlabeled samples; the fourth step is to use the classifier trained in the third step to evaluate on the test set.

[0125] The specific algorithm principle process is as follows Figure 5 As shown, the details are as follows:

[0126] 1) First, use some labeled samples to train a teacher model θ t , minimize the cross entropy loss of label data;

[0127]

[0128] 2) Use the teacher model to generate pseudo labels for unlabeled samples;

[0129]

[0130] 3) Use the above labeled samples and pseudo-label samples to train the student model θ s ;

[0131]

[0132] 4) Use the student model as the new teacher model and repeat step 2) for 10 times or until convergence is reached.

[0133] Furthermore, the method for determining the comprehensive evaluation value of depression of the user to be screened is:

[0134] S41: obtaining the number of credible signals whose depression evaluation values ​​of the to-be-screened user do not meet the requirements in multiple test cycles within a set time period and the depression evaluation values ​​of different credible signals, and determining the initial depression evaluation value of the test cycle according to the number of credible signals whose depression evaluation values ​​in each test cycle do not meet the requirements and the depression evaluation values ​​of different credible signals;

[0135] S42 determines whether the number of test cycles in which the initial depression assessment value of the to-be-screened user does not meet the requirements within the set time period is greater than the preset number of cycles. If so, proceed to step S44; if not, proceed to step S43;

[0136] S43 determines the initial depression assessment value of the user to be screened at the set time according to the number of credible signals whose depression assessment value of the user to be screened at the set time does not meet the requirements and the depression assessment values ​​of different credible signals, and judges whether the initial depression assessment value of the user to be screened at the set time meets the requirements, and if so, determines the comprehensive depression assessment value of the user to be screened according to the initial depression assessment value of the user to be screened at the set time;

[0137] S44 determines the comprehensive depression assessment value of the user to be screened based on the number of test cycles in which the initial depression assessment value of the user to be screened within the set time period does not meet the requirements, the initial depression assessment values ​​of different test cycles, and the initial depression assessment value of the user to be screened at the set time.

[0138] In the embodiments of the present invention, the term "plurality" refers to two or more than two, unless otherwise clearly defined. The terms "installation", "connection", "fixation" and the like should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection. For those skilled in the art, the specific meanings of the above terms in the embodiments of the present invention can be understood according to the specific circumstances.

[0139] In the description of the embodiments of the present invention, it needs to be understood that the directions or positional relationships indicated by the terms "upper" and "lower" are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or unit referred to must have a specific direction, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the embodiments of the present invention.

[0140] In the description of this specification, the description of the terms "one embodiment", "a preferred embodiment", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0141] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for quantitatively evaluating depression based on long-term electrocardiogram data, characterized in that: Specifically include: Obtaining electrocardiogram data of the user to be screened within a set time, and dividing the electrocardiogram data into reliable periods and abnormal periods according to waveform characteristics; Determine the credible signal according to the duration of the credible period of the user to be screened, and evaluate the credibility of the electrocardiogram data according to the time length and number of credible signals of the user to be screened within the set time, the credible period and the number of abnormal periods, and proceed to the next step when the credibility of the electrocardiogram data meets the requirements; Determining the depression evaluation value of the user to be screened according to the time domain analysis parameters and the frequency domain analysis parameters of the credible signal, and when the number of credible signals whose depression evaluation value of the user to be screened does not meet the requirements is greater than a preset number, proceeding to the next step; Taking the set time as the inspection period, the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements in multiple inspection periods within the set time period and the depression assessment values ​​of different credible signals are obtained, and a comprehensive depression assessment value of the user to be screened is calculated based on the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements within the set time and the depression assessment values ​​of different credible signals.

2. A depression quantitative assessment system based on long-term electrocardiogram data, characterized in that: Specifically include: Period division module, credibility assessment module, credibility signal determination module, depression assessment module; The cycle division module is responsible for obtaining the electrocardiogram data of the user to be screened within a set time, and dividing the electrocardiogram data into reliable cycles and abnormal cycles according to waveform characteristics; The credibility evaluation module is responsible for determining the credible signal according to the duration of the credible period of the user to be screened, and evaluating the credibility of the electrocardiogram data according to the time length and number of the credible signals of the user to be screened within the set time, the credible period and the number of abnormal periods; The credible signal determination module is responsible for determining the depression assessment value of the to-be-screened user according to the time domain analysis parameters and the frequency domain analysis parameters of the credible signal, and determining whether the number of credible signals whose depression assessment values ​​of the to-be-screened user do not meet the requirements is greater than a preset number; The comprehensive depression assessment value is responsible for taking the set time as the test period, obtaining the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements in multiple test periods within the set time period, as well as the depression assessment values ​​of different credible signals, and combining the number of credible signals whose depression assessment values ​​of the user to be screened do not meet the requirements within the set time, as well as the depression assessment values ​​of different credible signals to derive the comprehensive depression assessment value of the user to be screened.

3. The depression quantitative assessment system according to claim 2, characterized in that: When the credibility of the electrocardiogram data meets the requirements, the depression assessment value of the user to be screened is determined; when the number of credible signals whose depression assessment value of the user to be screened does not meet the requirements is greater than a preset number, the comprehensive depression assessment value of the user to be screened is determined.

4. The depression quantitative assessment system according to claim 2, characterized in that: The value range of the set time is at least over 24 hours.

5. The depression quantitative assessment system according to claim 2, characterized in that: The electrocardiogram data is divided into reliable cycles and abnormal cycles according to waveform characteristics, specifically including: The cardiac cycles of the electrocardiogram data are divided according to the waveform characteristics of the electrocardiogram data, and whether the user to be screened is in a state of intense activity during the cardiac cycle is determined according to the waveform characteristics of different cardiac cycles of the electrocardiogram data. If so, the cardiac cycle is determined to be an abnormal cycle, and if not, the cardiac cycle is determined to be a credible cycle.

6. The depression quantitative assessment system according to claim 2, characterized in that: When there are continuous trusted cycles greater than a set number threshold or the duration of continuous trusted cycles is greater than a preset time threshold in the electrocardiogram data of the user to be screened, the electrocardiogram data corresponding to the continuous trusted cycles greater than the set number threshold or the continuous trusted cycles greater than the preset time threshold are taken as trusted signals.

7. The depression quantitative assessment system according to claim 2, characterized in that: The specific method for evaluating the credibility of the electrocardiogram data is: Obtaining the number of trustworthy cycles of the user to be screened within the set time, and determining whether the credibility of the electrocardiogram data meets the requirement according to the number of trustworthy cycles, if so, proceeding to the next step, if not, determining that the credibility of the electrocardiogram data does not meet the requirement; Obtaining the time length and number of the credible signals of the to-be-screened user within the set time, and determining whether the credibility of the electrocardiogram data meets the requirements according to the time length and number of the credible signals of the to-be-screened user within the set time, if yes, proceeding to the next step, if not, determining that the credibility of the electrocardiogram data does not meet the requirements; Evaluate the credibility of the trusted cycle of the user to be screened according to the number of trusted cycles and the number of abnormal cycles of the user to be screened within the set time; Determine a stable and credible signal according to the duration of the credible signal of the user to be screened within the set time, and evaluate the credibility of the credible signal of the user to be screened according to the time length and number of the stable and credible signals, the time length and signal of the credible signal; The credibility of the electrocardiogram data of the user to be screened is evaluated according to the credibility of the credible signal of the user to be screened and the credibility of the credible cycle.

8. The depression quantitative assessment system according to claim 7, characterized in that: When the duration of the credible signal is greater than a set time reference value, the credible signal is determined to be a stable credible signal, wherein the duration of the stable credible signal is determined based on the sum of the durations of all credible signals of the user to be screened within a set time.

9. The depression quantitative assessment system according to claim 2, characterized in that: The method for determining the depression evaluation value of the user to be screened is: Data preprocessing: preprocessing the credible signal through a baseline removal filter and a low-pass filter, and performing peak transformation on the preprocessed data; Feature recognition and extraction of feature parameters: The peak transformed data is subjected to the threshold method to find the R wave position of the credible signal. After the R wave position is found, the front and rear windows are set at the corresponding positions in the preprocessed data, and the start and end positions of the P, Q, R, S, and T waves and the corresponding peak positions are found by the threshold method. Heart rate variability parameter calculation: heart rate variability parameters are calculated according to the start and end positions of the P, Q, R, S, and T waves of the credible signal and the corresponding peak positions to obtain the time domain analysis parameters and frequency domain analysis parameters of the credible signal; Depression evaluation value output: According to the time domain analysis parameters and frequency domain analysis parameters of the credible signal, an evaluation model based on a gradient enhancement algorithm is used to determine the depression evaluation value of the user to be screened.

10. The depression quantitative assessment system according to claim 2, characterized in that: The method for determining the comprehensive evaluation value of depression of the user to be screened is: S41: obtaining the number of credible signals whose depression evaluation values ​​of the to-be-screened user do not meet the requirements in multiple test cycles within a set time period and the depression evaluation values ​​of different credible signals, and determining the initial depression evaluation value of the test cycle according to the number of credible signals whose depression evaluation values ​​in each test cycle do not meet the requirements and the depression evaluation values ​​of different credible signals; S42 determines whether the number of test cycles in which the initial depression assessment value of the to-be-screened user does not meet the requirements within the set time period is greater than the preset number of cycles. If so, proceed to step S44; if not, proceed to step S43; S43 determines the initial depression assessment value of the user to be screened at the set time according to the number of credible signals whose depression assessment value of the user to be screened at the set time does not meet the requirements and the depression assessment values ​​of different credible signals, and judges whether the initial depression assessment value of the user to be screened at the set time meets the requirements, and if so, determines the comprehensive depression assessment value of the user to be screened according to the initial depression assessment value of the user to be screened at the set time; S44 determines the comprehensive depression assessment value of the user to be screened based on the number of test cycles in which the initial depression assessment value of the user to be screened within the set time period does not meet the requirements, the initial depression assessment values ​​of different test cycles, and the initial depression assessment value of the user to be screened at the set time.

Images