Method, device and wearable device for testing user state through wearable device

Abstract

This disclosure provides a method, apparatus, and wearable device for testing user status using a wearable device. The method includes: collecting multiple physiological signals from a user during a detection period; performing standardization processing on the multiple physiological signals using the user's individual physiological baseline to obtain multiple standardized features for the detection period; inputting the standardized features of the detection period into a pre-trained prediction model for evaluation to obtain an assessment result indicating whether the user is in an anxious state; and, in response to the assessment result indicating whether the user is in an anxious state, performing anxiety quantification scoring based on the multiple physiological signals and the individual physiological baseline to obtain an anxiety score value, thereby outputting the user status test result. Therefore, this disclosure, by introducing an individual physiological baseline to perform user status detection, can solve the evaluation error problem caused by individual physiological differences and improve the accuracy of user status detection results.

Description

Technical Field

[0001] This disclosure relates to the field of health monitoring technology, specifically to a method, apparatus, electronic device, and wearable device for testing a user's status using a wearable device. Background Technology

[0002] Anxiety disorders are a prevalent mental health problem worldwide, severely impacting people's quality of life. Traditional anxiety detection methods mainly rely on subjective questionnaires and clinical interviews, which have inherent limitations such as high subjectivity, poor real-time performance, and inability to conduct continuous monitoring, making it difficult to capture sudden anxiety states in everyday situations.

[0003] With the widespread adoption of wearable devices, objective anxiety detection based on physiological signals has become a research hotspot. Physiological signals (such as heart rate variability, skin conductance, and skin temperature) can directly reflect the activity of the autonomic nervous system and have a clear physiological correlation with anxiety states. However, existing technologies still have many shortcomings: First, they mostly focus on generalized stress or emotion recognition, lacking specific detection algorithms for anxiety and failing to effectively distinguish between anxiety and similar emotional states such as excitement or general stress. Second, some solutions use a single sensor to collect signals, making it difficult to capture the multidimensional physiological characteristics of anxiety, resulting in low detection accuracy. Third, related technologies mostly use fixed thresholds for judgment, failing to consider individual physiological differences and lacking a dynamically adaptive baseline model, leading to a high false alarm rate. Fourth, related technologies only achieve qualitative detection of anxiety states, lacking fine quantification of anxiety levels and failing to track dynamic changes in anxiety levels. Fifth, most devices only have monitoring functions, lacking real-time guidance mechanisms and failing to form a complete service loop.

[0004] Therefore, there is an urgent need for a user state detection technology that has the ability to specifically identify anxiety states, high-precision detection performance, and a personalized adaptive mechanism. Summary of the Invention

[0005] In view of the above, this disclosure provides a method, apparatus, electronic device, and wearable device for testing user status through wearable devices, which can achieve objective, specific, and personalized identification of user status, and improve the accuracy, timeliness, and practical application value of user status detection.

[0006] According to a first aspect of this disclosure, a method for testing a user's state using a wearable device is provided, comprising: collecting multiple physiological signals of the user under test corresponding to a detection period; performing standardization processing on the multiple physiological signals using the user's individual physiological baseline to obtain multiple standardized features of the detection period; inputting the multiple standardized features of the detection period into a pre-trained prediction model for evaluation to obtain an evaluation result of whether the user under test is in an anxious state; responding to the evaluation result of whether the user under test is in an anxious state, performing anxiety quantification scoring based on the multiple physiological signals and the individual physiological baseline to obtain an anxiety score value; and outputting a user state test result of the user under test based on the anxiety score value.

[0007] According to a second aspect of this disclosure, an apparatus for testing a user's state via a wearable device is provided, comprising: a data acquisition module configured to acquire multiple physiological signals of the user under test corresponding to a detection period; a data processing module configured to perform standardization processing on the multiple physiological signals using the user's individual physiological baseline to obtain multiple standardized features of the detection period; an anxiety assessment module configured to input the multiple standardized features of the detection period into a pre-trained prediction model for evaluation to obtain an assessment result indicating whether the user under test is in an anxious state; an anxiety scoring module configured to, in response to the assessment result indicating whether the user is in an anxious state, perform anxiety quantification scoring based on the multiple physiological signals and the individual physiological baseline to obtain an anxiety score value; and an output module configured to output the user state test result of the user under test based on the anxiety score value.

[0008] According to a third aspect of this disclosure, a wearable device is provided, including means as described in the second aspect, for testing a user's state via the means.

[0009] According to a fourth aspect of this disclosure, an electronic device is provided, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; the memory is used to store at least one executable instruction, the executable instruction causing the processor to perform an operation corresponding to the method for testing user status via a wearable device as described in the first aspect.

[0010] According to the technical solutions for testing user status through wearable devices provided in this disclosure, by introducing individual physiological baselines for standardization, physiological differences between different users are eliminated, improving the accuracy and universality of detection; by further performing quantitative scoring after identifying anxiety states, the degree of anxiety is precisely quantified, significantly enhancing the ability to dynamically monitor the user's psychological state, and improving the practical value of wearable devices and the user experience. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings.

[0012] Figure 1 This is a flowchart illustrating a method for testing a user's state using a wearable device, which is an exemplary embodiment of this disclosure.

[0013] Figure 2 This is a flowchart illustrating a method for testing a user's state using a wearable device, which is another exemplary embodiment of this disclosure.

[0014] Figure 3 This is a structural block diagram of an apparatus for testing a user's state via a wearable device, which is an exemplary embodiment of this disclosure.

[0015] Figure 4 A simplified structural diagram of an electronic device that is an exemplary embodiment of this disclosure. Detailed Implementation

[0016] To enable those skilled in the art to better understand the technical solutions in the embodiments of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and thoroughly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art should fall within the protection scope of this disclosure.

[0017] Reference is made to the accompanying drawings, which form part of the detailed description and illustrate exemplary embodiments. Furthermore, it should be understood that other embodiments may be utilized, and structural and / or logical changes may be made without departing from the scope of the claimed subject matter. It should also be noted that orientations and references (e.g., up, down, top, bottom, etc.) may be used merely to facilitate the description of features in the drawings. Therefore, the following detailed description is not to be construed in a limiting sense, and the scope of the claimed subject matter is defined only by the appended claims and their equivalents.

[0018] Numerous details are set forth in the following description. However, it will be apparent to those skilled in the art that the embodiments described herein can be practiced without these specific details. In some instances, well-known methods and apparatus are shown in block diagram form rather than in detail to avoid obscuring the embodiments described herein. Throughout this specification, references to “embodiment,” “one embodiment,” or “some embodiments” mean that a particular feature, structure, function, or characteristic described in connection with that embodiment is included in at least one embodiment herein. Therefore, the phrases “in an embodiment,” “in one embodiment,” or “some embodiments” appearing throughout this specification do not necessarily refer to the same embodiment. Furthermore, in one or more embodiments, particular features, structures, functions, or characteristics can be combined in any suitable manner. For example, a first embodiment can be combined with a second embodiment in any way that does not mutually exclude particular features, structures, functions, or characteristics associated with two embodiments.

[0019] As used in the description and appended claims, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0020] As described throughout this document and in the claims, a list of items connected by the terms “at least one of” or “one or more of” may mean any combination of the listed items. For example, the phrase “at least one of A, B, or C” may mean A; B; C; A and B; A and C; B and C; or A, B, and C.

[0021] In recent years, the widespread use of wearable devices has provided new possibilities for the objective detection of anxiety. Physiological signals (such as heart rate variability, skin conductance, and skin temperature) can reflect the activity state of the autonomic nervous system and are closely related to anxiety. Studies have shown that anxiety manifests as characteristic physiological responses such as decreased heart rate variability (especially a decrease in the high-frequency component HF-HRV), increased skin conductance, and changes in skin temperature.

[0022] However, existing technologies mainly suffer from the following problems: First, existing technologies mostly focus on stress detection or emotion recognition, lacking specific detection algorithms for anxiety, and cannot effectively distinguish anxiety from other emotional states (such as excitement or general stress); second, some existing solutions use only a single sensor (such as PPG or EDA only), resulting in low accuracy and difficulty in capturing the multidimensional physiological characteristics of anxiety; third, existing methods mostly use fixed thresholds, without considering individual differences and dynamic baselines, leading to a high false alarm rate; in addition, existing technologies lack fine-grained quantitative scoring of anxiety levels, making it difficult to track dynamic changes in anxiety levels.

[0023] Based on the various problems existing in the prior art, the embodiments of this disclosure provide a scheme for testing user status through wearable devices. This scheme achieves specific identification and high-precision quantitative assessment of anxiety state through multimodal feature fusion and individual baseline adaptation mechanism. At the same time, it constructs a closed-loop service from real-time monitoring to personalized guidance, which significantly improves the robustness and application value of detection.

[0024] The specific implementations of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating a method for testing a user's state using a wearable device, which is an exemplary embodiment of this disclosure.

[0025] The method described in this embodiment can be applied to various wearable devices to achieve user status detection in different scenarios. Wearable devices include, but are not limited to, smartwatches, smart bracelets, and smart rings. Alternatively, a combination of electronic devices and physiological signal sensing devices can be used to achieve the technical solution of this embodiment. For example, a combination of a smartphone and a medical patch can be used, where the medical patch collects the physiological signals of the user being tested, and the smartphone performs the core calculations.

[0026] refer to Figure 1 This embodiment mainly includes the following steps: Step 102: Collect multiple physiological signals from the tested user corresponding to the detection period.

[0027] In some embodiments, the acquired physiological signals include: photoplethysmography (PPG) signal, electrical conductance of skin (EDA) signal, skin temperature (SKT) signal, and acceleration (ACC) signal.

[0028] A photoplethysmography (PPG) sensor located on the wrist can be used to collect radial artery blood volume change signals through a green LED and photodiode with a wavelength of 520-570nm. The sensor's sampling frequency is configured to be 100Hz and the dynamic range is 16-bit to extract heart rate (HR) and heart rate variability (HRV) related features.

[0029] A constant voltage method (0.5V) can be used to measure skin conductivity via wrist electrodes, reflecting sweat gland activity and sympathetic nervous system activation levels. The sensor's sampling frequency can be set to 4Hz, and the measurement range can be set to 0.01-100μS.

[0030] A thermistor or infrared temperature sensor can be used to measure the skin temperature of the wrist. The sensor's sampling frequency can be set to 1Hz, the accuracy to ±0.1°C, and the measurement range to 20-45°C.

[0031] A triaxial accelerometer can be used to acquire motion acceleration signals, with a selectable range of ±8g and a selectable sampling frequency of 50Hz. The accelerometer is used to detect motion state and body position to help identify and filter out physiological signal interference caused by motion.

[0032] This embodiment, by limiting the types of physiological signals (PPG, EDA, temperature, acceleration), can provide a multimodal data foundation for subsequent feature extraction, thereby improving the robustness of anxiety recognition and the ability to capture multidimensional physiological responses.

[0033] Step 104: Using the individual physiological baseline of the tested user, perform standardization processing on multiple physiological signals to obtain multiple standardized features for the detection period.

[0034] In some embodiments, anxiety-specific feature extraction can be performed on the physiological signals collected during the detection period to obtain the physiological features of the detection period. In this embodiment, the physiological features may include heart rate variability (HRV) features, skin conductance (EDA) features, skin temperature features, and heart rate features.

[0035] Preferably, the collected physiological signals can be preprocessed before feature extraction, and anxiety-specific feature extraction can be performed based on the preprocessed physiological signals to obtain various physiological features.

[0036] The following is a brief introduction to the preprocessing of different physiological signals: For example, the preprocessing of photoplethysmography (PPG) signals mainly includes: A fourth-order Butterworth bandpass filter is used to clean the original PPG signal to effectively remove interference caused by baseline drift and high-frequency noise. An adaptive thresholding method is used to detect the pulse wave peak value. This threshold is dynamically calculated as the sum of the signal mean and 0.6 times the standard deviation within a sliding window to accurately locate heartbeat events. The original RR interval sequence is generated by calculating the time difference between adjacent peaks. A moving median filter is used to clean the RR interval sequence. The window length can be set to 5 RR intervals, and sampling data deviating from the median within ±20% of the window are removed to effectively eliminate outliers caused by motion artifacts or detection errors. An interpolation algorithm is used to convert the cleaned unequal-interval RR interval sequence into an equal-interval time sequence, with the resampling frequency set to 4Hz. This step provides the necessary data format for subsequent frequency domain analysis such as Fast Fourier Transform (FFT). Finally, a first-order polynomial fitting technique is used to identify and remove linear trend components in the resampled signal to ensure signal stationarity and improve the accuracy of spectral analysis.

[0037] For example, the preprocessing of electrical conductance (EDA) signals mainly includes: First, the original EDA signal is filtered using a third-order Butterworth low-pass filter, with a cutoff frequency set to 1 Hz. This step aims to effectively suppress high-frequency noise and electromyographic interference, providing a smooth signal basis for subsequent analysis. Then, the filtered signal is decomposed using the cvxEDA algorithm to decouple it into two independent physiological components: (1) skin conductance level (SCL): a slowly changing baseline component extracted by a sparse non-negative deconvolution method, used to characterize the individual's overall arousal level; (2) skin conductance response (SCR): a transient component corresponding to a rapid impulse response. In this embodiment, a peak detection threshold of 0.01 μS can be set to identify and locate transient emotional activation events triggered by specific stimuli.

[0038] Subsequently, refined parameters can be extracted from the detected SCR, including: peak amplitude, rise time (defined as the time required for the signal to rise from 10% to 90% of the peak value, in order to improve the accuracy of the rise time measurement and eliminate data jitter anomalies), recovery time (defined as the time required for the signal to decay from the peak value to 50%, characterizing the fall time characteristics), and SCR frequency per unit time (i.e., peak count / minute).

[0039] Finally, pseudo-SCRs with amplitudes smaller than a preset amplitude threshold (e.g., 0.01 μS) or rise times shorter than a preset rise threshold (e.g., 0.5 s) can be removed to ensure that the extracted SCR features have clear physiological significance and analytical reliability.

[0040] For example, preprocessing for skin temperature (SKT) signals mainly includes: First, a sliding window (window length for example, 60 seconds) is used to smooth the raw skin temperature signal, thereby smoothing short-term temperature fluctuations and suppressing high-frequency noise interference. Then, based on the filtered signal, the slope of temperature change within a preset time window (window length for example, 5 minutes) can be calculated to quantify the dynamic trend of temperature change.

[0041] In some embodiments, before performing feature extraction on the physiological signals collected during the detection period, the activity state of each physiological signal can be determined based on the acceleration signal of each physiological signal during the detection period; and based on the activity state of each physiological signal, each physiological signal belonging to the motion state can be removed from the detection period.

[0042] For example, based on time-series data collected by a triaxial accelerometer, threshold judgment logic and machine learning classification algorithms (such as random forest models) can be integrated to classify the current activity patterns of the tested user, accurately identifying various typical states including sitting, walking, and strenuous exercise. A data confidence labeling mechanism is also established. When the algorithm determines that the user (tested user) is in a state of strenuous exercise, the system automatically marks the physiological signal data for that period as a low-confidence interval, or selectively skips it in subsequent analysis, thereby effectively avoiding the interference of motion artifacts on the evaluation of physiological indicators and ensuring the accuracy of the analysis results.

[0043] The implementation process of anxiety-specific feature extraction for different preprocessed signals is as follows: In this embodiment, anxiety-specific feature extraction can be performed on the photoplethysmography (PPG) signal to obtain heart rate variability (HRV) features, specifically including: (1) Temporal characteristics (window length for example, 5 minutes), which mainly include: SDNN (Standard Deviation of Normal RR Interval): Used to characterize the overall level of heart rate variability, which tends to decrease under anxious conditions.

[0044] RMSSD (Root Mean Square of RR Interval Difference): Used to reflect the activity level of the parasympathetic nervous system, it is significantly reduced during anxiety.

[0045] pNN50 (the percentage of adjacent RR intervals greater than 50ms): This indicator reflects the anxiety state, and its value decreases when the person is anxious.

[0046] (2) Frequency domain features (window length, for example, 5 minutes): A Hanning window can be applied to the RR interval sequence to effectively reduce spectral leakage. Frequency domain analysis can be completed using Fast Fourier Transform (FFT) or Welch power spectral density estimation. The frequency domain features mainly include: HF-HRV (high-frequency power, 0.15-0.4Hz): used to reflect parasympathetic nerve activity (vagal tone), it is significantly reduced by more than 40% during anxiety, making it a key indicator for identifying anxiety.

[0047] LF-HRV (low-frequency power, 0.04-0.15Hz): This indicator reflects the combined regulatory effect of the sympathetic and parasympathetic nervous systems and may increase in anxious states.

[0048] LF / HF ratio: Used as an indicator of sympathetic-parasympathetic balance, calculated using the following formula: ( This represents the percentage of low-frequency power. (This is a percentage of high-frequency power). When this ratio is higher than a certain threshold (e.g., 2.0), it indicates that the sympathetic nervous system is in a state of overactivation.

[0049] Total power (TP, 0.04-0.4Hz): The overall energy level used to characterize heart rate variability, which tends to decrease during anxiety.

[0050] In some embodiments, normalization may be performed to eliminate the influence of total power on the high-frequency and low-frequency power analysis results. The normalization calculation formula can be expressed as Formula 1 below: (Formula 1) In formula 1, Represents the high-frequency normalized value. This represents the low-frequency normalized value. This indicates extremely low frequency.

[0051] In this embodiment, anxiety-specific feature extraction can be performed on the electrical conductance (EDA) signal to obtain EDA features, specifically including: Mean SCL: This represents the average skin conductance (SCL) level over a specific time period. This indicator reflects an individual's overall physiological arousal level over a period of time, and typically rises significantly under anxiety or stress.

[0052] SCR Frequency: This represents the number of peak skin conductance response (SCR) signals detected per unit of time (usually per minute). This frequency is a key indicator of the frequency of mood fluctuations and increases significantly in anxious states (e.g., more than 3 times / minute).

[0053] SCR Amplitude: This represents the average amplitude of all valid SCR peaks. The strength of the amplitude is positively correlated with the intensity of emotional activation; in an anxious state, the amplitude of the SCR response elicited by a stimulus usually increases.

[0054] SCR Rise Time: This reflects the time required for the SCR waveform to rise from its initial response point to its peak. This characteristic reflects the activation speed of the sympathetic nervous system; in a state of anxiety, the activation process is more rapid, resulting in a shorter rise time.

[0055] SCL Coefficient of Variation (CV): The ratio of the SCL standard deviation to the SCL mean, used to quantify the relative volatility of the SCL signal over a period of time. This indicator effectively reflects the instability of autonomic nervous system activity; its volatility typically increases in anxious states.

[0056] In this embodiment, the following skin temperature features can be extracted from the preprocessed skin temperature signal to quantify the physiological responses related to anxiety: Mean temperature: This represents the average skin temperature within a specific time window (e.g., 5 minutes). This indicator characterizes the overall thermal level of the body surface during this period and is a fundamental parameter for assessing thermal comfort and physiological baseline status.

[0057] Temperature drop rate: This quantifies the rate at which skin temperature decreases by calculating the slope of temperature change over time. This indicator is a key feature for identifying anxiety-induced vasoconstriction responses, where reduced peripheral blood flow in an anxious state leads to a significant drop in skin temperature (e.g., a rate of decrease of -0.5°C / 10 minutes).

[0058] Temperature standard deviation: This reflects the dispersion or variability of the skin temperature signal within the analysis window. This indicator can be used to assess the stability of the thermoregulatory system. In emotional states such as anxiety, instability of the autonomic nervous system may lead to an increase in the standard deviation of skin temperature.

[0059] In this embodiment, heart rate features related to anxiety state can be extracted from the preprocessed photoplethysmography (PPG) signal, as follows: Mean Heart Rate (HR): This measure represents the average number of heartbeats within a specific window (e.g., 5 minutes). This metric is a key parameter reflecting sympathetic nervous system excitability. Excluding exercise interference, anxiety or stress typically activates the sympathetic nervous system, leading to a significant increase in mean heart rate (e.g., exceeding 80 bpm).

[0060] Heart rate trend: By calculating the slope of a linear regression of heart rate data within an analysis window (e.g., 5 minutes), the dynamic direction and rate of heart rate changes are quantified. This feature can effectively reflect a sustained increase or instability in heart rate under anxiety.

[0061] In some embodiments, the physiological characteristics of the detection period can be standardized based on the mean and standard deviation of an individual's physiological baseline to obtain standardized characteristics for that period. This data standardization mechanism can eliminate individual differences, thereby improving the generalization ability of subsequent prediction models.

[0062] For example, the calculation of the standardized features satisfies the following formula 2: (Formula 2) In Formula 2, Indicates standardized features, Indicates physiological characteristics during the detection period. This represents the average value of an individual's physiological baseline. The standard deviation represents the baseline of an individual's physiological condition.

[0063] This embodiment uses data standardization to map current physiological characteristics to individual baseline distributions, which can eliminate the influence of unit and individual differences, enhance the comparability of characteristics of different users at different times, and provide standardized data for model input.

[0064] Step 106: Input multiple standardized features of the detection period into the pre-trained prediction model to perform evaluation and obtain the evaluation result of whether the tested user is in an anxious state.

[0065] In some embodiments, a sliding detection window can be used to acquire standardized features during the detection period. A prediction model, based on pre-trained label states, performs predictions on the standardized features of each detection window, outputting the probability distribution of each detection window corresponding to each label state. By employing a sliding window and continuous window probability judgment mechanism, false triggers caused by instantaneous fluctuations are avoided, improving the stability and robustness of anxiety state determination, making it suitable for real-time or near-real-time detection scenarios.

[0066] In some embodiments, the window length of the detection window can be set to 5 minutes.

[0067] In this embodiment, the labeled states of the prediction model include at least an anxious state. For example, the labeled states of the prediction model may include four categories: anxious state, calm state, general stress state, and excited state.

[0068] In some embodiments, the prediction model includes a classifier constructed based on the CatBoost algorithm. Specifically, the classifier is configured to receive normalized features of each detection window and predict the probability of it belonging to each labeled state, thereby outputting a probability distribution for each detection window corresponding to each labeled state. By employing the CatBoost algorithm to perform anxiety detection, the physiological representation patterns specific to anxiety states can be deeply coupled during the model training phase, thereby effectively distinguishing subtle differences between anxiety states and general stress or excitement states, significantly improving the robustness and reliability of classification prediction results.

[0069] In some embodiments, the classifier tree depth can be 6 to 8 layers, thereby achieving a balance between model complexity and overfitting, ensuring prediction efficiency while maintaining high classification accuracy, which is suitable for deployment in resource-constrained wearable devices.

[0070] In some embodiments, during the training of the classifier, the learning rate can be set to 0.03, the L2 regularization coefficient to 3.0, and the number of iterations to 500 to 1000 rounds. However, this is not a limitation, and those skilled in the art can arbitrarily adjust the above model parameters based on actual model training needs, and this disclosure does not impose any restrictions on this.

[0071] In some embodiments, a weighted average calculation can be performed on the distribution probability of the target window corresponding to the anxiety state based on the distribution probability of the anxiety state corresponding to the two adjacent detection windows of the target window, and the distribution probability of the target window corresponding to the anxiety state can be updated based on the calculation result. Here, the target window can be any one of the detection windows.

[0072] Specifically, in continuous physiological signal detection, the predicted anxiety state probability for a single detection window may fluctuate due to transient artifacts (such as motion interference or signal loss). Since human anxiety states are typically temporally continuous, a neighborhood-weighted averaging mechanism is introduced to fuse the probability distributions of the target window with its preceding and following adjacent windows. This leverages temporal correlation to suppress high-frequency noise, resulting in a smoother, more physiologically consistent anxiety probability curve.

[0073] For example, when the nth detection window is determined as the target window, the probability distribution values ​​of the (n-1)th detection window (hereinafter referred to as the preceding window) and the (n+1)th detection window (hereinafter referred to as the subsequent window) corresponding to the anxiety state can be obtained respectively. By taking a weighted average of the probability distribution values ​​of the anxiety state corresponding to the three consecutive detection windows (for example, assigning the weights of the three detection windows to 0.2, 0.3, and 0.5), the probability distribution value of the anxiety state corresponding to the target window is updated based on the weighted average calculation result, thereby ensuring that the detected anxiety state has a reasonable duration on the time axis.

[0074] In some embodiments, if the probability distribution of anxiety states in at least two consecutive detection windows is greater than a preset probability threshold (e.g., the probability distribution is greater than 0.6), an assessment result can be obtained that the tested user is in an anxiety state.

[0075] Step 108: In response to the assessment results of whether the tested user is in an anxious state, perform anxiety quantification scoring based on multiple physiological signals and individual physiological baselines to obtain an anxiety score value.

[0076] In some embodiments, a heart rate variability factor can be calculated based on the heart rate variability characteristics and the individual physiological baseline of the heart rate variability characteristics during the detection period; a skin conductance factor can be calculated based on the skin conductance characteristics during the detection period; a heart rate factor can be calculated based on the heart rate characteristics and the individual physiological baseline of the heart rate characteristics during the detection period; a skin temperature factor can be calculated based on the temperature drop rate in the skin temperature characteristics during the detection period; and an anxiety score can be calculated based on the heart rate variability factor, skin conductance factor, heart rate factor, and skin temperature factor.

[0077] For example, the formula for calculating the anxiety score can be expressed as the following formula 3: (Formula 3) In formula 3, Indicates the anxiety score. Indicates the factors affecting heart rate variability. Indicates factors affecting skin electrical conductance. Indicates factors affecting skin temperature. Indicates factors affecting heart rate. Indicates high-frequency power characteristics, Individual physiological baseline representing high-frequency power characteristics, express Frequency characteristics express Maximum frequency, Indicates the characteristics of average heart rate. An individual physiological baseline representing heart rate characteristics.

[0078] In formula 3, , , and All are preset weights, for , , and The weighting ratio is not limited in this embodiment; it is sufficient that the sum of the four weights equals 1.

[0079] This embodiment calculates influencing factors and scores them comprehensively based on four physiological characteristics: HRV, skin conductance, heart rate, and temperature. It achieves a multi-dimensional physiological fusion anxiety quantification index, which is more sensitive and specific than a single index.

[0080] Step 110: Based on the anxiety score, output the user status test results of the tested user.

[0081] In some embodiments, the current emotional state of the test user can be classified and determined based on the calculated anxiety quantification score, and the user state test results, including anxiety score, emotion category, and other information, can be output to the test user through a wearable device.

[0082] In some embodiments, based on a calculated anxiety score, a corresponding guidance prompt can be determined from a preset configuration file, and the determined guidance prompt can be output through a wearable device. Optionally, the guidance prompt output through the wearable device may include, but is not limited to, breathing training instructions, muscle relaxation instructions, soothing music, etc.

[0083] In this embodiment, a higher anxiety score indicates a more severe level of anxiety. Personalized guidance prompts can be provided based on the anxiety score.

[0084] For example: when the anxiety score falls within the range of 31-60, deep breathing exercises can be guided, prompting users to pause their current task and take a short break; when the anxiety score falls within the range of 60-80, light exercise can be recommended (e.g., 5-10 minutes of walking or stretching), 5 minutes of guided mindfulness meditation can be provided, or it can be suggested to stay away from the source of anxiety (e.g., turn off the computer, leave the stressful environment); when the anxiety score falls within the range of 81-100, an early warning vibration and pop-up reminder can be triggered immediately, guiding users to implement emergency breathing control (e.g., perform the 4-7-8 breathing method for 3 minutes), and recommending contacting emergency contacts or mental health counseling hotlines; at the same time, anxiety episodes are automatically recorded to support subsequent tracking and retrospection of user status changes.

[0085] In some embodiments, personalized guidance prompts may be provided based on anxiety scores, the user's activity status, and / or the time of day. For example, when the user is asleep during the nighttime hours, guidance prompts adapted to the sleep context may be provided, such as playing soothing breathing guidance audio (at a low volume) and gentle meditation guidance to help the user fall back asleep.

[0086] This embodiment matches corresponding guidance prompts based on anxiety scores to provide personalized and actionable behavioral guidance based on user status detection results, thereby enhancing the practicality of wearable devices and the user experience.

[0087] In summary, this embodiment eliminates individual user differences and improves the accuracy and universality of user state detection by fusing multimodal physiological signals and introducing individual physiological baselines for standardization. The use of a categorical gradient boosting tree algorithm to predict anxiety states effectively distinguishes anxiety from other states, enhancing the reliability of the prediction results. By outputting user state test results including anxiety scores and corresponding guidance prompts, it not only achieves refined quantification of anxiety levels but also constructs a closed-loop service from "anxiety detection" to "relief guidance," significantly enhancing the practical value of wearable devices.

[0088] Figure 2 This is a flowchart illustrating an exemplary embodiment of the individual physiological baseline construction method of this disclosure. The individual physiological baseline of the tested user in this embodiment mainly includes two stages: a baseline establishment step and a baseline update step. The following will describe these stages in conjunction with... Figure 2 Provide a detailed description.

[0089] Step 202: Collect multiple baseline physiological signals of the test user in a resting state, calculate the average value and standard deviation of the physiological characteristics of the multiple baseline physiological signals, and establish the individual physiological baseline of the test user.

[0090] In practical applications, a baseline data collection period of 3-7 days can be conducted when the user first uses the device to record the baseline physiological signal characteristics in a calm state, thereby establishing the individual physiological baseline of the tested user.

[0091] In some embodiments, multiple baseline physiological signals of the test subject in a resting state can be collected. Preprocessing is performed on the multiple baseline physiological signals to obtain multiple preprocessed signals. Feature extraction is performed on the multiple preprocessed signals to obtain multiple physiological features. The mean and standard deviation of the multiple physiological features are calculated, and based on the mean, standard deviation, and a preset multiple of the standard deviation of the multiple physiological features, the individual physiological baseline of the test subject is calculated.

[0092] For example, the signal acquisition cycle for the baseline establishment step is 3 to 7 days.

[0093] In this step, the acquisition of baseline physiological signals, signal pre-processing, and feature extraction can be referred to the relevant descriptions in steps 102 and 104 above, and will not be repeated here.

[0094] In some embodiments, the individual physiological baseline of the tested user can satisfy the following formula 4: (Formula 4) In Formula 4, Indicates an individual's physiological baseline. This represents the average value. For a preset multiple (e.g., (Value is 1.5) It represents the standard deviation.

[0095] This embodiment ensures the reproducibility and standardization of individual physiological baselines by acquiring baseline physiological signals under quiescent conditions and performing signal preprocessing, feature extraction, and statistical calculations. The individual physiological baseline is calculated using the signal mean and standard deviation, balancing baseline stability and individual variability. An adjustable K value allows the individual physiological baseline to be adapted to detection scenarios with varying sensitivity requirements.

[0096] Step 204: Dynamically collect multiple dynamic physiological signals of the tested user using the baseline update window. Based on the preset time decay weight and the sampling timestamp of each dynamic physiological signal within the sliding window, calculate the temporal weight of each dynamic physiological signal within the baseline update window. Update the individual physiological baseline based on the signal feature value of each dynamic physiological signal and the temporal weight of each dynamic physiological signal.

[0097] For example, a sliding window (with a window length of, for example, 7 days) can be used to dynamically update individual physiological baselines to adapt to long-term changes in users' physiological states. Time decay weights can be introduced, with more recent data having a higher weight, while the weight of data from 7 days ago can linearly decay to 0.5.

[0098] During the baseline update step, data marked as occurring during anxiety episodes can be excluded to avoid contaminating the baseline.

[0099] In some embodiments, the acceleration signal and acquisition timestamp of each dynamic physiological signal within the baseline update window can be identified to determine the activity state and activity period of each dynamic physiological signal; based on the activity state and activity period of each dynamic physiological signal, the individual context to which each dynamic physiological signal belongs is determined. Based on the individual context to which each dynamic physiological signal belongs and the physiological characteristic values ​​of each dynamic physiological signal, the individual physiological baseline is adjusted respectively to establish an individual context baseline for the tested user corresponding to different individual contexts. In this embodiment, each activity state includes at least one of the following: a motor state, a calm state, and a sleep state.

[0100] For example, the nighttime baseline HF-HRV is typically higher than the daytime baseline, and the heart rate baseline during walking is higher than that while sitting. For instance, if the accelerometer detects that the user (the tested user) has been lying down for more than 15 minutes and determines that the user is in a sleep-preparation state, the system can invoke the nighttime contextual baseline to perform anxiety detection, thereby improving the accuracy of anxiety detection. By introducing individual contextual baselines (exercise, calm, sleep, etc.), anxiety assessment can distinguish the physiological response characteristics under different activity states, avoiding false positives or false negatives caused by different activity states, and improving the model's predictive accuracy.

[0101] In summary, this embodiment achieves dynamic adaptation and high-precision modeling of the baseline, significantly reducing the false alarm rate. Specifically, by collecting physiological signals in a calm state during the baseline establishment period (3-7 days), an initial normal range consistent with the user's actual physiological condition is established. During the baseline update period, time decay weights and a sliding window are used to automatically adjust the baseline according to long-term changes in the user's physiological state (such as changes in physical condition and seasonal changes), avoiding detection failures caused by baseline aging. Furthermore, this embodiment enables fine-grained, scenario-based detection by establishing independent contextual baselines for different activity states such as exercise, calm, and sleep, further improving the reliability of anxiety state detection results.

[0102] Figure 3 A simplified structural diagram of an apparatus for testing a user's state via a wearable device, which is an exemplary embodiment of this disclosure.

[0103] like Figure 3 As shown, the device 300 in this embodiment mainly includes: a data acquisition module 302, a data processing module 304, an anxiety assessment module 306, an anxiety scoring module 308, and an output module 310.

[0104] The data acquisition module 302 is configured to acquire multiple physiological signals of the tested user corresponding to the detection period; The data processing module 304 is configured to perform standardized processing on the multiple physiological signals using the individual physiological baseline of the tested user to obtain multiple standardized features of the detection period. The anxiety assessment module 306 is configured to input multiple standardized features of the detection period into a pre-trained prediction model to perform an assessment and obtain an assessment result on whether the tested user is in an anxious state. Anxiety scoring module 308 is configured to perform anxiety quantification scoring based on the multiple physiological signals and the individual physiological baseline in response to the assessment result of whether the tested user is in an anxious state, and obtain an anxiety score value.

[0105] The output module 310 is configured to output the user status test results of the tested user based on the anxiety score.

[0106] In some embodiments, the plurality of physiological signals include: photoplethysmography (PPG) signal, skin conductance signal, skin temperature signal, and acceleration signal.

[0107] In some embodiments, the device 300 further includes a baseline maintenance module (not shown) configured to obtain the individual physiological baseline of the tested user in the following manner: a baseline establishment step, which involves acquiring multiple reference physiological signals of the tested user in a resting state, calculating the average value and standard deviation of the physiological characteristics of the multiple reference physiological signals, to establish the individual physiological baseline of the tested user; and a baseline update step, which involves dynamically acquiring multiple dynamic physiological signals of the tested user using a baseline update window, calculating the temporal weight of each dynamic physiological signal within the baseline update window based on a preset time decay weight and the sampling timestamp of each dynamic physiological signal within the sliding window, and updating the individual physiological baseline based on the signal characteristic value of each dynamic physiological signal and the temporal weight of each dynamic physiological signal.

[0108] In some embodiments, the baseline maintenance module is further configured to: collect multiple baseline physiological signals of the test user in a calm state; perform preprocessing on the multiple baseline physiological signals to obtain multiple preprocessed signals; perform anxiety-specific feature extraction on the multiple preprocessed signals to obtain the multiple physiological features; calculate the mean and standard deviation of the multiple physiological features; and calculate the individual physiological baseline of the test user based on the mean, standard deviation, and a preset multiple of the standard deviation of the multiple physiological features.

[0109] In some embodiments, the baseline maintenance module is further configured such that: the signal acquisition period for the baseline establishment step is 3 to 7 days; and the individual physiological baseline of the tested user satisfies the following formula:

[0110] in, Indicates an individual's physiological baseline. This represents the average value. For preset multiples, It represents the standard deviation.

[0111] In some embodiments, the baseline maintenance module is further configured to: identify the acceleration signal and acquisition timestamp of each dynamic physiological signal within the baseline update window to determine the activity state and activity period of each dynamic physiological signal; determine the individual context to which each dynamic physiological signal belongs based on the activity state and activity period of each physiological signal; adjust the individual physiological baseline based on the individual context to which each dynamic physiological signal belongs and the physiological characteristic value of each dynamic physiological signal, and establish an individual context baseline for the tested user corresponding to different individual contexts; wherein each activity state includes at least one of the following: movement state, calm state, and sleep state.

[0112] In some embodiments, the data processing module 304 is configured to: perform feature extraction on the physiological signals collected during the detection period to obtain the physiological features of the detection period, the physiological features including: heart rate variability features, skin conductance features, skin temperature features, and heart rate features; and perform standardization processing on the physiological features of the detection period based on the mean and standard deviation of the individual physiological baseline to obtain the standardized features of the detection period. The standardized feature satisfies the following formula definition:

[0113] in, This indicates the standardized feature. This indicates the physiological characteristics of the detection period. This represents the average value of the individual's physiological baseline. The standard deviation represents the individual's physiological baseline.

[0114] In some embodiments, the data processing module 304 is configured to: before performing feature extraction on the physiological signals collected during the detection period, determine the activity state of each physiological signal based on the acceleration signal of each physiological signal during the detection period; and based on the activity state of each physiological signal, remove each physiological signal belonging to the motion state from the detection period.

[0115] In some embodiments, the anxiety assessment module 306 is configured to: acquire standardized features in the detection period by sliding the detection window; perform prediction on the standardized features of each detection window based on the pre-trained label states using the prediction model, and output the distribution probability of each detection window corresponding to each label state, wherein each label state includes at least the anxiety state; if the distribution probability of the anxiety state in at least two consecutive detection windows is greater than a preset probability threshold, an assessment result is obtained that the tested user is in an anxiety state.

[0116] In some embodiments, the prediction model includes a classifier configured to predict the probability that the standardized features of each detection window belong to each labeled state based on a class gradient boosting tree algorithm, and output the distribution probability of each detection window corresponding to each labeled state; wherein each labeled state includes a calm state, an anxious state, a general stress state, and an excited state.

[0117] In some embodiments, the classifier has a tree depth of 6 to 8 layers.

[0118] In some embodiments, the anxiety assessment module 306 is configured to: after obtaining the distribution probability of each detection window corresponding to each label state, perform a weighted average calculation on the distribution probability of the target window corresponding to the anxiety state based on the distribution probability of the two adjacent detection windows of the target window corresponding to the anxiety state, and update the distribution probability of the target window corresponding to the anxiety state based on the calculation result; wherein, the target window is any one of the detection windows.

[0119] In some embodiments, the detection window has a window length of 5 minutes.

[0120] In some embodiments, the anxiety scoring module 308 is configured to: calculate a heart rate variability factor based on the heart rate variability characteristics and the individual physiological baseline of the heart rate variability characteristics during the detection period; calculate a skin conductance factor based on the skin conductance characteristics during the detection period; calculate a heart rate factor based on the heart rate characteristics and the individual physiological baseline of the heart rate characteristics during the detection period; calculate a skin temperature factor based on the skin temperature characteristics during the detection period; and calculate the anxiety score based on the heart rate variability factor, the skin conductance factor, the heart rate factor, and the skin temperature factor.

[0121] In some embodiments, the anxiety scoring module 308 is configured to: after obtaining the anxiety score, determine a guidance prompt corresponding to the anxiety score from a preset configuration file; and output the guidance prompt.

[0122] Another embodiment of this disclosure also provides a wearable device including the apparatus as described in the foregoing embodiments for detecting user status via the apparatus.

[0123] Reference Figure 4 The diagram shows a schematic representation of an electronic device according to an embodiment of the present disclosure. The specific embodiments of the present disclosure do not limit the specific implementation of the electronic device.

[0124] like Figure 4 As shown, the electronic device in this embodiment may include: a processor 402, a communications interface 404, a memory 406, and a communications bus 408.

[0125] The processor 402, communication interface 404, and memory 406 communicate with each other via communication bus 408.

[0126] Communication interface 404 is used to communicate with other electronic devices or servers.

[0127] The processor 402 is used to execute program 410, specifically to perform the relevant steps in the above-described method embodiment for testing user status through a wearable device.

[0128] Specifically, program 410 may include program code that includes computer operation instructions.

[0129] Processor 402 may be a CPU, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement embodiments of this disclosure. The smart device may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0130] Memory 406 is used to store program 410. Memory 406 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0131] Program 410 may include multiple computer instructions. Specifically, program 410 may use multiple computer instructions to cause processor 402 to perform the operation corresponding to the method for testing user status through wearable device described in any of the foregoing multiple method embodiments.

[0132] The specific implementation of each step in procedure 410 can be found in the corresponding descriptions of the steps and units in the above method embodiments, and has corresponding beneficial effects, which will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the devices and modules described above can be referred to the corresponding process descriptions in the foregoing method embodiments, and will not be repeated here.

[0133] This disclosure also provides a computer storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in any of the foregoing method embodiments. The computer storage medium includes, but is not limited to, compact disc read-only memory (CD-ROM), random access memory (RAM), floppy disk, hard disk, or magneto-optical disk.

[0134] This disclosure also provides a computer program product, including computer instructions that instruct a computing device to perform operations corresponding to the method for testing a user's state through a wearable device described in any of the above embodiments.

[0135] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this disclosure can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this disclosure.

[0136] The methods described above according to embodiments of this disclosure can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code originally stored on a remote recording medium or a non-transitory machine-readable medium and subsequently stored on a local recording medium, downloaded over a network. Thus, the methods described herein can be stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA)). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., Random Access Memory (RAM), Read-Only Memory (ROM), Flash Memory, etc.) capable of storing or receiving software or computer code, implementing the methods described herein when the software or computer code is accessed and executed by the computer, processor, or hardware. Furthermore, when a general-purpose computer accesses code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for performing the methods shown herein.

[0137] Those skilled in the art will recognize that the units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments disclosed herein.

[0138] The above embodiments are only used to illustrate the embodiments of this disclosure, and are not intended to limit the embodiments of this disclosure. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of this disclosure. Therefore, all equivalent technical solutions also fall within the scope of the embodiments of this disclosure, and the patent protection scope of the embodiments of this disclosure should be defined by the claims.

Claims

1. A method for testing a user's state using a wearable device, comprising: Multiple physiological signals from the tested user corresponding to the detection period were collected; Using the individual physiological baseline of the tested user, the multiple physiological signals are standardized to obtain multiple standardized features for the detection period. Multiple standardized features of the detection period are input into a pre-trained prediction model for evaluation to obtain an assessment result of whether the tested user is in an anxious state. In response to the assessment result of whether the tested user is in an anxious state, an anxiety quantification score is performed based on the multiple physiological signals and the individual physiological baseline to obtain an anxiety score value; Based on the anxiety score, the user status test results of the tested user are output.

2. The method according to claim 1, wherein, The multiple physiological signals include: photoplethysmography (PPG) signal, skin conductance signal, skin temperature signal, and acceleration signal.

3. The method of claim 2, wherein, The individual physiological baseline of the tested users was obtained through the following methods: The baseline establishment step involves collecting multiple baseline physiological signals from the tested user in a resting state, calculating the mean and standard deviation of the physiological characteristics of the multiple baseline physiological signals, and establishing the individual physiological baseline of the tested user. The baseline update step involves dynamically acquiring multiple dynamic physiological signals of the tested user using a baseline update window. Based on a preset time decay weight and the sampling timestamp of each dynamic physiological signal within the sliding window, the temporal weight of the multiple dynamic physiological signals within the baseline update window is calculated. Based on the signal feature value of each dynamic physiological signal and the temporal weight of each dynamic physiological signal, the individual physiological baseline is updated.

4. The method according to claim 3, wherein, The baseline establishment steps include: Multiple baseline physiological signals of the tested user were collected in a calm state; Preprocessing is performed on the multiple reference physiological signals to obtain multiple preprocessed signals; Anxiety-specific feature extraction is performed on the multiple preprocessed signals to obtain the multiple physiological features; Calculate the mean and standard deviation of the aforementioned physiological characteristics; The individual physiological baseline of the tested user is calculated based on the mean, standard deviation, and a preset multiple of the standard deviation of the multiple physiological characteristics.

5. The method according to claim 4, wherein, The signal acquisition period for the baseline establishment step is 3 to 7 days. The individual physiological baseline of the tested users satisfies the following formula: wherein, denotes the average value, is a preset multiple, denotes the standard deviation.

6. The method according to claim 3, further comprising: The acceleration signal and acquisition timestamp of each dynamic physiological signal within the baseline update window are identified to determine the activity state and activity period of each dynamic physiological signal; Based on the activity state and activity period of each dynamic physiological signal, the individual context to which each dynamic physiological signal belongs is determined; Based on the individual context to which each dynamic physiological signal belongs and the physiological characteristic value of each dynamic physiological signal, the individual physiological baseline is adjusted respectively to establish the individual context baseline of the tested user corresponding to different individual contexts; Each activity state includes at least one of the following: a dynamic state, a calm state, or a sleep state.

7. The method of claim 2, wherein, The method utilizes the individual physiological baseline of the tested user to perform standardized processing on the multiple physiological signals, obtaining multiple standardized features for the detection period, including: Feature extraction is performed on the physiological signals collected during the detection period to obtain the physiological features of the detection period, including: heart rate variability features, skin conductance features, skin temperature features, and heart rate features; Based on the mean and standard deviation of the individual physiological baseline, the physiological characteristics of the detection period are standardized to obtain the standardized characteristics of the detection period. The standardized feature satisfies the following formula definition: wherein, represents the standardized feature, represents a physiological feature of the detection period, represents a mean value of the individual physiological baseline, represents a standard deviation of the individual physiological baseline.

8. The method of claim 7, wherein, Before performing feature extraction on the physiological signals collected during the detection period, the method further includes: Based on the acceleration signals of each physiological signal during the detection period, the activity state of each physiological signal is determined; Based on the activity state of each physiological signal, physiological signals belonging to the motion state are removed from the detection period.

9. The method of claim 7, wherein, In response to the assessment result that the tested user is in an anxious state, an anxiety quantification score is performed based on the physiological signals and the individual's physiological baseline to obtain an anxiety score value, including: Based on the heart rate variability characteristics and individual physiological baselines of the heart rate variability characteristics during the detection period, the heart rate variability influencing factor is calculated. Based on the skin conductivity characteristics during the detection period, calculate the skin conductivity influencing factor; Based on the heart rate characteristics during the detection period and the individual physiological baseline of the heart rate characteristics, the heart rate influencing factor is calculated; Based on the skin temperature characteristics during the detection period, calculate the skin temperature influencing factor; The anxiety score is calculated based on the heart rate variability factor, the skin conductance factor, the heart rate factor, and the skin temperature factor.

10. The method of claim 1, wherein, The step of inputting multiple standardized features of the detection period into a pre-trained prediction model for evaluation to obtain an assessment result of whether the tested user is in an anxious state includes: Standardized features within the detection period are obtained by sliding the detection window. The prediction model performs predictions on the standardized features of each detection window based on the pre-trained label states, and outputs the distribution probability of each detection window corresponding to each label state, wherein each label state includes at least the anxiety state. If the probability distribution of anxiety states in at least two consecutive detection windows is greater than a preset probability threshold, an assessment result is obtained that the tested user is in an anxious state.

11. The method of claim 10, wherein, The prediction model includes: The classifier is configured to predict the probability that the normalized features of each detection window belong to each label state based on the class gradient boosting tree algorithm, and output the distribution probability of each detection window corresponding to each label state. The label states include calm state, anxious state, general stress state, and excited state.

12. The method of claim 11, wherein, The classifier has a tree depth of 6 to 8 levels.

13. The method according to claim 10, wherein, After obtaining the probability distribution of each detection window corresponding to each label state, the method further includes: Based on the distribution probability of anxiety states corresponding to two adjacent detection windows of the target window, a weighted average calculation is performed on the distribution probability of anxiety states corresponding to the target window, and the distribution probability of anxiety states corresponding to the target window is updated based on the calculation result. The target window can be any one of the detection windows.

14. The method according to claim 1, wherein, After obtaining the anxiety score, the method further includes: The system determines guidance prompts corresponding to the anxiety score from a preset configuration file. The aforementioned guidance prompt will be displayed.

15. An apparatus for testing a user's state via a wearable device, comprising: The data acquisition module is configured to acquire multiple physiological signals of the tested user corresponding to the detection period. The data processing module is configured to perform standardized processing on the multiple physiological signals using the individual physiological baseline of the tested user to obtain multiple standardized features of the detection period. Anxiety assessment module is configured to input multiple standardized features of the detection period into a pre-trained prediction model to perform assessment and obtain an assessment result of whether the tested user is in an anxious state. Anxiety scoring module is configured to perform anxiety quantification scoring based on the multiple physiological signals and the individual physiological baseline in response to the assessment result of whether the test user is in an anxious state, and obtain an anxiety score value. The output module is configured to output the user status test results of the tested user based on the anxiety score.

16. A wearable device, comprising the means of claim 15, for testing a user's state via the means.

17. An electronic device comprising: The processor, the communication interface, the memory, and the communication bus are provided, and the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform an operation corresponding to the method for testing a user's state through a wearable device as described in any one of claims 1 to 14.

Images