Quantitative evaluation method for brain fatigue regulation effect of piano performance

Abstract

The application discloses a kind of piano performance brain fatigue regulation effect quantification evaluation method, including constructing test scene, respectively collecting the pulse wave signal and performance behavior data of subject wakefulness, fatigue, performance end;Extract physiological characteristics in pulse wave signal, and important evaluation and screening are carried out, and weight is determined according to importance;Extract performance behavior characteristics;Statistical relationship analysis is carried out to the change amount of performance behavior characteristics and physiological characteristics, constructs the regulation model of performance behavior characteristics to physiological characteristics weight, calculates weight regulation factor;The change amount of physiological characteristics is weighted fusion, and the output comprehensive fatigue regulation index is output.The application carries out multidimensional brain fatigue regulation effect evaluation by fusing physiological signal and performance behavior, overcomes the limitation that traditional evaluation relies on subjective report or single physiological index, realizes objective, continuous, quantifiable evaluation to piano performance intervention effect, reduces subjective evaluation deviation, improves scientificity and reliability.

Description

Technical Field

[0001] This invention belongs to the field of information processing technology, and specifically relates to a quantitative evaluation method for the effect of piano playing on mental fatigue regulation. Background Technology

[0002] Existing methods for assessing mental fatigue are mainly divided into two categories: subjective assessment and objective assessment. Subjective scales rely on the subject's self-perception, are easily influenced by subjective bias, and are difficult to achieve continuous and stable fatigue assessment. Objective assessments infer fatigue levels by measuring signals directly related to neural and physiological states. Common methods include electroencephalography (EEG), electrocardiography (ECG), and electrooculography (EOG). However, these signals usually require complex acquisition equipment and strict wearing conditions in practical applications, which limits the subject's freedom of movement and the experimental environment. In contrast, photoplethysmography (PPG) signals, as an easily obtainable physiological signal, can be acquired non-invasively through photoelectric sensing. It has advantages such as simple equipment, comfortable wear, and minimal interference with behavior. Furthermore, it contains rich information on heart rate variability and vascular morphodynamics, effectively reflecting the regulatory state of the autonomic nervous system, providing a highly promising solution for real-time and objective fatigue assessment.

[0003] Music intervention, as an easily implemented and widely accepted method, has demonstrated clear feasibility and a unique mechanism of action in alleviating mental fatigue. Music-related behaviors may regulate mental fatigue by influencing autonomic nervous system activity and attentional resource allocation. Among various music-related activities, music performance, compared to passive listening, is typically accompanied by richer perceptual, motor, and cognitive engagement. Piano playing, as a typical example of music performance, is characterized by strong subject adaptability and wide applicability, making it representative in studies of music-related behaviors and physiological states.

[0004] In existing fatigue-related research, fatigue detection methods and intervention methods are often independent, making it difficult to use fatigue assessment results for quantitative analysis of intervention effects. Especially in music-related behaviors or performance activities, although some studies focus on their impact on cognitive states, there is a lack of technical solutions for a unified quantitative assessment of fatigue state changes induced by music-related behaviors, incorporating changes in physiological signals. Therefore, this invention uses piano playing as a typical music performance scenario to quantitatively assess fatigue state and its changes under music intervention conditions, which has significant research significance and application value. Summary of the Invention

[0005] The main objective of this invention is to overcome the technical problems of neglecting intervention process modeling and lacking unified quantitative indicators for evaluating the relief effect in existing fatigue-related studies, and to provide a quantitative evaluation method for the effect of piano performance on mental fatigue regulation. This method analyzes the changes in pulse wave physiological characteristics before and after piano performance intervention, and introduces modeling of the regulatory effect of performance behavior characteristics on physiological responses to construct a continuous fatigue regulation index, thereby achieving an objective evaluation and quantitative description of the fatigue regulation effect under music intervention scenarios.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] One aspect of the present invention provides a method for quantitatively evaluating the effect of piano playing on mental fatigue regulation, comprising the following steps:

[0008] A test scenario was constructed, and pulse wave signals were collected from subjects in both awake and mentally fatigued states.

[0009] Behavioral data of the subjects during their piano playing and pulse wave signals after the performance were collected.

[0010] Physiological features, including heart rate variability and pulse wave morphology, were extracted from pulse wave signals during the waking state, mental fatigue state, and after the performance.

[0011] A classification model is established based on the random forest algorithm. The importance of physiological characteristics in the awake state and the mental fatigue state is evaluated. A set of physiological characteristics sensitive to mental fatigue is selected, and the corresponding physiological characteristic weights are determined according to the importance of each physiological characteristic in the classification model.

[0012] Based on the behavioral data, performance behavior features are extracted, including rhythmic stability features, performance continuity features, and performance load features.

[0013] A statistical relationship analysis was conducted on the changes in performance behavior characteristics and physiological characteristics sensitive to mental fatigue before and after performance. A moderating model of the weight of performance behavior characteristics on physiological characteristics was constructed, and the weight moderating factor was calculated.

[0014] Based on physiological feature weights and weight adjustment factors, the changes in physiological features in the set of physiological features sensitive to mental fatigue are weighted and fused to output a comprehensive fatigue regulation index for quantitatively characterizing the regulatory effect of piano performance intervention on mental fatigue.

[0015] As a preferred technical solution, the construction of the test scenario involves collecting pulse wave signals from subjects in both a conscious and mentally fatigued state, specifically as follows:

[0016] Subjects were kept in a seated, resting position, and pulse wave signals were collected in the conscious state using a pulse wave signal sensor.

[0017] By subjecting subjects to a continuous mental load through pre-set cognitive tasks, the subjects are made to experience mental fatigue.

[0018] After completing the cognitive task, the subjects' pulse wave signals were collected as pulse wave signals under mental fatigue state.

[0019] The pulse wave signals collected in both conscious and mentally fatigued states are labeled and stored.

[0020] As a preferred technical solution, the collection of behavioral data during the subject's piano playing process and pulse wave signals after the performance specifically includes:

[0021] Subjects were instructed to play the piano according to pre-set performance instructions;

[0022] During piano performance, performance behavior data is collected synchronously, including key trigger events generated during the performance and their corresponding time information;

[0023] The pulse wave signal of the subject was collected after the performance.

[0024] As a preferred technical solution, the pulse wave signal is preprocessed before extracting physiological characteristics from the pulse wave signal in the conscious state, mental fatigue state, and after the performance. Specifically, the preprocessing involves:

[0025] The pulse wave signal is low-pass filtered to remove high-frequency noise;

[0026] The baseline curve is fitted by cubic spline interpolation and then removed from the original pulse wave signal to obtain the baseline-drift-free pulse wave signal.

[0027] The pulse wave signal after baseline drift removal is periodically filtered to eliminate abnormal pulse wave periods;

[0028] Perform pulse wave feature point identification, including the starting point, the peak of the main wave, and the peak of the diphtheria wave.

[0029] As a preferred technical solution, the extraction of physiological characteristics from the awake state, the mental fatigue state, and the pulse wave signal after the performance specifically includes:

[0030] Based on the identified pulse wave feature points, multi-dimensional physiological features are extracted, including heart rate variability features and pulse wave morphological features, specifically:

[0031] Based on the identified main wave peak, a pulse cycle interval sequence is calculated to extract heart rate variability features, which include time-domain features and frequency-domain features. Simultaneously, pulse wave morphological features are calculated based on the time difference, amplitude difference, and waveform area of ​​the pulse wave feature points, which include time features, amplitude features, and area features.

[0032] As a preferred technical solution, the step of assessing the importance of physiological characteristics in the awake and mentally fatigued states, screening out a set of physiological characteristics sensitive to mental fatigue, and determining the corresponding physiological characteristic weights based on the importance of each physiological characteristic in the classification model, specifically:

[0033] Significance analysis was performed on the physiological characteristics of the awake and mentally fatigued states. If the physiological characteristic data conformed to a normal distribution, a T-test was used for significance analysis; otherwise, a Wilcoxon signed-rank test was used. Based on the significance analysis results, physiological characteristics that did not differ significantly between the awake and mentally fatigued states were removed.

[0034] The physiological characteristics selected through significance analysis were used as input, and the random forest algorithm was used to train the mental fatigue state discrimination model.

[0035] Feature selection is performed using recursive feature elimination to obtain the feature subset with the best classification effect, and the corresponding physiological feature weights are calculated based on the feature importance output by random forest.

[0036] As a preferred technical solution, the extraction of performance behavior features based on the behavioral data specifically includes:

[0037] Based on behavioral data, a time series of key presses during performance is constructed, specifically as follows:

[0038] ;

[0039] in, Indicates the first The time it takes for the button to be pressed. This indicates the total number of key events generated during the performance. A key event is a valid trigger event that meets the minimum press duration or force threshold.

[0040] Calculate the time interval sequence between adjacent key events, specifically as follows:

[0041] , ;

[0042] Based on the key press time interval sequence, rhythm stability features are extracted, specifically:

[0043] ;

[0044] in, The standard deviation of the time interval series. To prevent constants with zero denominators; rhythmic stability characteristics The larger the value, the more stable the playing rhythm;

[0045] Set time interval threshold When satisfied When this occurs, it is determined as a performance interruption event, and the performance continuity characteristic is defined as follows:

[0046] ;

[0047] in, The number of interruption events during the performance; performance continuity characteristics. The larger the value, the more continuous the performance.

[0048] Extracting performance load features, specifically:

[0049] ;

[0050] in, Duration of performance; characteristics of performance load The higher the value, the higher the performance load;

[0051] The extracted rhythmic stability features, performance continuity features, and performance load features are combined and normalized to construct a performance behavior feature vector, specifically:

[0052] ;

[0053] in, , , These are the characteristics of the performance behavior after normalization.

[0054] As a preferred technical solution, the statistical relationship analysis of the changes in performance behavior characteristics and physiological characteristics sensitive to mental fatigue before and after performance is performed to construct a model for the adjustment of the weights of performance behavior characteristics on physiological characteristics, and to calculate the weight adjustment factor, specifically as follows:

[0055] A set of physiological characteristics sensitive to mental fatigue Let the first The values ​​of each physiological characteristic before the performance were... The value after the performance The original changes in physiological characteristics are defined as: ;

[0056] The original variations in physiological characteristics are standardized, including orientation consistency transformation and scaling transformation, specifically as follows:

[0057] ;

[0058] ;

[0059] in, and They represent the first The mean and standard deviation of the changes in each physiological characteristic in the training sample;

[0060] The standardized changes in physiological characteristics are then fused to obtain the total changes in physiological characteristics.

[0061] Based on multiple performance samples, the statistical relationship between performance behavior characteristics and changes in physiological characteristics was analyzed. The total change in physiological characteristics was used as the dependent variable and performance behavior characteristics were used as the independent variable to establish a regression model and obtain the contribution coefficient of each performance behavior characteristic.

[0062] A comprehensive performance quality score is obtained by weighting and summing the performance behavior characteristics based on the contribution coefficient;

[0063] The overall performance quality score is mapped using an adjustment function to obtain a weight adjustment factor.

[0064] As a preferred technical solution, the changes in various physiological characteristics are weighted and fused based on physiological characteristic weights and weight adjustment factors to output a comprehensive fatigue regulation index for quantitatively characterizing the regulatory effect of piano performance intervention on mental fatigue, specifically as follows:

[0065] Based on the established baseline weights of fatigue-sensitive physiological characteristics, and combined with a weight adjustment factor, the adjusted weights of the physiological characteristics are calculated as follows:

[0066] ;

[0067] in, and The first Adjusted weights and baseline weights of physiological characteristics, This is a weighting adjustment factor calculated based on the performance quality score;

[0068] Based on the adjusted physiological characteristic weights, the standardized physiological characteristic changes are weighted and fused to output a mental fatigue regulation index, specifically:

[0069] ;

[0070] in, For the first Adjusted weights of physiological characteristics For the first The amount of change after standardization of a physiological characteristic.

[0071] Another aspect of the present invention provides a quantitative evaluation system for the effect of piano performance on mental fatigue regulation, applied to the above-mentioned quantitative evaluation method for the effect of piano performance on mental fatigue regulation, including a pulse wave signal acquisition module, a feature extraction module, a weight evaluation module, a weight adjustment factor calculation module, and a comprehensive fatigue regulation index calculation module.

[0072] The pulse wave signal acquisition module is used to construct test scenarios and collect pulse wave signals when the subject is awake and mentally fatigued; it also collects behavioral data of the subject during piano playing and pulse wave signals after the performance.

[0073] The feature extraction module is used to extract physiological features from pulse wave signals in the waking state, mental fatigue state, and after the performance, including heart rate variability features and pulse wave morphological features; based on the behavioral data, performance behavior features are extracted, including rhythm stability features, performance continuity features, and performance load features.

[0074] The weight evaluation module is used to establish a classification model based on the random forest algorithm, evaluate the importance of physiological features in the awake state and the mental fatigue state, screen out the set of physiological features sensitive to mental fatigue, and determine the corresponding physiological feature weights according to the importance of each physiological feature in the classification model.

[0075] The weight adjustment factor calculation module is used to perform statistical relationship analysis on the changes in performance behavior characteristics and physiological characteristics sensitive to mental fatigue before and after performance, construct a model for the adjustment of the weight of physiological characteristics by performance behavior characteristics, and calculate the weight adjustment factor.

[0076] The comprehensive fatigue regulation index calculation module is used to perform weighted fusion of the changes in physiological features in the set of physiological features sensitive to mental fatigue based on the weights of physiological features and weight adjustment factors, and output a comprehensive fatigue regulation index to quantitatively characterize the regulatory effect of piano performance intervention on mental fatigue.

[0077] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0078] (1) This invention proposes a multi-dimensional method for evaluating the effect of mental fatigue regulation by integrating physiological signals and performance behavior, overcoming the limitations of traditional evaluation methods that rely on subjective reports or single physiological indicators. This invention constructs a mental fatigue regulation index by simultaneously collecting pulse wave signals and piano performance behavior data, achieving an objective, continuous, and quantifiable evaluation of the intervention effect on piano performance. This method effectively reduces the bias caused by subjective evaluation and improves the scientificity and reliability of fatigue relief effect evaluation.

[0079] (2) This invention innovatively introduces a dynamic adjustment mechanism for the weight of physiological characteristics based on performance behavior characteristics, thereby achieving personalized and adaptive optimization of the assessment model. Traditional assessment models often use fixed weights to integrate multiple features, making it difficult to adapt to the varying effects of different performance qualities on fatigue relief. This invention establishes a statistical relationship model between performance behavior characteristics and changes in physiological characteristics, calculates a weight adjustment factor based on performance quality, and dynamically adjusts the contribution of each physiological characteristic in the comprehensive assessment, thus more accurately reflecting the actual effect of a single performance intervention and enhancing the explanatory power and applicability of the assessment model.

[0080] (3) This invention collects pulse wave signals, extracts heart rate variability features and pulse wave morphological features, uses a random forest model to evaluate the importance of each physiological feature in fatigue discrimination, and combines a recursive feature elimination method to screen out the feature subset that is most sensitive to mental fatigue, thereby improving the accuracy and robustness of the fatigue regulation quantification model.

[0081] (4) The present invention uses pulse wave signal as the main physiological signal source. The sensing method is simple, non-invasive and comfortable to wear. It does not require complex electrode arrangement or professional medical equipment, which reduces the threshold for system deployment and use. It is suitable for various application scenarios such as laboratory environment, training scenario and daily health monitoring. It can obtain quantitative evaluation results of the brain fatigue regulation effect in a low-cost, convenient and fast way.

[0082] (5) By standardizing and weighting the changes in physiological characteristics before and after the performance, this invention avoids the interference of individual baseline differences on the assessment results, making the assessment results more reflective of the relative change trend caused by the piano performance intervention, and improving the comparability and stability between different subjects and in multiple intervention processes. Attached Figure Description

[0083] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0084] Figure 1 This is a flowchart of a method for quantitatively evaluating the effect of piano playing on mental fatigue regulation, as disclosed in Embodiment 1 of the present invention.

[0085] Figure 2 This is a flowchart of the preprocessing of pulse wave signals in Embodiment 1 of the present invention;

[0086] Figure 3 It is the pulse wave signal after low-pass filtering in Embodiment 2 of the present invention;

[0087] Figure 4It is the pulse wave signal after baseline removal in Embodiment 2 of the present invention. Detailed Implementation

[0088] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0089] Example 1:

[0090] This invention discloses a method for quantitatively evaluating the effect of piano playing on mental fatigue regulation, such as... Figure 1 As shown, the specific steps are as follows:

[0091] S1. Construct a test scenario and collect pulse wave signals when the subject is awake and under mental fatigue. The mental fatigue state is obtained by continuously applying mental load to the subject through a pre-set cognitive task. The specific steps are as follows:

[0092] S101. In a suitable temperature and quiet environment, the subject is required to remain seated and at rest, with their fingers naturally placed or lightly pressed on the sensing area of ​​the pulse wave signal sensor, so that the pulse wave signal sensor can stably acquire the pulse wave signal, thereby collecting the pulse wave signal of the subject in a conscious state. The sampling rate of the pulse wave sensor is set to 200Hz.

[0093] S102. Set up a cognitive task. In this embodiment, the cognitive task consists of a Stroop task and an N-back task. The Stroop task is specifically set as follows: a colored word is presented in the center of the screen. The font color of the word may or may not match its meaning. The subject must ignore the meaning and quickly determine the displayed color of the word, responding by pressing a key. The N-back task is specifically set as follows: a sequence of numbers (0–9) is presented sequentially in the center of the screen. The subject must determine whether the currently appearing number is the same as the Nth number that appeared previously, where N can be 2 or 3. The subject completes the cognitive task continuously for 20 minutes, recording the reaction time and accuracy throughout. After the task, the subject must immediately fill out the Karolinska Sleepiness Scale for self-evaluation. Based on subjective feelings and task performance, it is determined whether the subject has reached a state of mental fatigue. If not, the subject is subjected to another set of cognitive tasks to continuously apply mental load until the subject reaches a state of mental fatigue.

[0094] S103. After the subject completes the cognitive task and confirms that he is in a state of mental fatigue, the subject's pulse wave signal is collected again as pulse wave signal data in the state of mental fatigue.

[0095] S104. The pulse wave signals collected in the conscious state and the mental fatigue state are labeled and stored for subsequent construction of fatigue state discrimination model and feature screening.

[0096] S2. Guide the subject to perform a piano piece, simultaneously collecting behavioral data related to the performance during the process, and collecting the subject's pulse wave signal again after the performance to characterize the physiological state after the intervention. Specific steps are as follows:

[0097] S201. After collecting pulse wave signals under mental fatigue, the piano score information to be played is presented to the subject through the display. Each note is arranged from left to right according to the preset playing order, and the starting time and corresponding playing time of each note are indicated by sliding the moving indicator line from left to right, thereby guiding the performer to complete the piano performance according to the score.

[0098] S202. During piano performance, behavioral data related to the performance process is collected synchronously, recording key trigger events and their corresponding time information (specifically, including the key trigger time and key number). In this embodiment, the collection of performance behavior data is accomplished using an electronic piano device in a laboratory. During playing, the scanning circuit board inside the electronic piano can monitor the physical state of each key in real time, record the key number and the time of key press, and transmit and store it to a computer as the basic data for subsequent performance behavior feature extraction.

[0099] S203. After the performance, the subject's pulse wave signal was collected again as pulse wave signal data of the physiological state after the intervention.

[0100] S3. Preprocess the pulse wave signals acquired in steps S1 and S2, and extract multi-dimensional physiological features, including heart rate variability and pulse wave morphology. The pulse wave preprocessing process is as follows: Figure 2 As shown, the specific steps are as follows:

[0101] S301. The pulse wave signal is subjected to low-pass filtering to suppress high-frequency noise components introduced by environmental interference or sensors. The effective range of the pulse wave signal is generally below 10Hz. Therefore, this invention uses a finite impulse response (FIR) low-pass filter with a cutoff frequency of 10Hz to filter out the high-frequency noise in the pulse wave signal.

[0102] S302. Fit the baseline curve using cubic spline interpolation and remove the baseline curve from the original pulse wave signal to obtain the baseline-drift-free pulse wave signal. This step can remove the ultra-low frequency signals attached to the pulse wave caused by human breathing movements or changes in human posture.

[0103] S303. Perform periodic screening on the baseline-removed signal. Calculate the Pearson correlation coefficient between each period to evaluate the signal similarity between adjacent periods, remove abnormal pulse wave periods, and thus obtain a high-quality pulse wave signal.

[0104] S304. Perform pulse wave feature point identification, including the starting point, the main wave peak point, and the dicrotic wave peak point;

[0105] S305. Based on the processed pulse wave signal and the identified pulse wave feature points, extract multi-dimensional physiological features, including heart rate variability features and pulse wave morphological features. The heart rate variability features include time-domain features and frequency-domain features, while the pulse wave morphological features include amplitude features, time features, and area features. These features together constitute a multi-dimensional set of physiological features.

[0106] Specifically, based on the identified main wave peak, a pulse cycle interval sequence is calculated to extract heart rate variability features; at the same time, pulse wave morphological features are calculated based on the time difference, amplitude difference, and waveform area of ​​the pulse wave feature points.

[0107] S4. Based on the physiological characteristics of the awake and mentally fatigued states, a classification model is established using random forest. The importance of these physiological characteristics is assessed, and a set of physiological characteristics sensitive to mental fatigue is selected. The corresponding basic weights are then determined based on the importance of each physiological characteristic in the model. The specific steps are as follows:

[0108] S401. Conduct a significance analysis on physiological characteristics under awake and mental fatigue states to evaluate the effectiveness of each physiological characteristic in fatigue detection. Specifically, perform a difference test on the value distribution of the same physiological characteristic under awake and mental fatigue states. If the physiological characteristic data conforms to a normal distribution, use a T-test to test the significance of each pair of data; for data that does not conform to a normal distribution, use a Wilcoxon signed-rank test. Based on the significance analysis results, remove physiological characteristics that do not show significant differences between awake and mental fatigue states.

[0109] S402. Using the candidate physiological features selected through significance analysis as input features and the corresponding physiological state labels (awake state or mental fatigue state) as output labels, construct a random forest classification model.

[0110] S403. Feature selection is performed using a recursive feature elimination method to obtain the feature subset with the best classification effect. The importance scores of the selected target physiological features in the random forest model are normalized, and the normalized values ​​are used as the basic weights for each feature in subsequent evaluations. The fatigue-sensitive physiological features and basic weights selected in this step will be used in step S7 to build a fatigue regulation quantification model.

[0111] S5. Based on the piano performance behavior data collected in step S2, extract performance behavior features, including rhythmic stability features, performance continuity features, and performance load features. The specific steps are as follows:

[0112] S501. During piano performance, record the timestamp of each valid key press event to construct a performance key press time sequence:

[0113] , (Formula 1);

[0114] in, Indicates the first The time it takes for the button to be pressed. This represents the total number of key events generated during the performance. Each key event is a valid trigger event that meets the minimum press duration or force threshold. Based on this, the time interval sequence between adjacent key events is calculated:

[0115] , (Formula 2);

[0116] in, ;

[0117] S502. Extract rhythmic stability features to characterize the consistency of time intervals between adjacent keys during performance, reflecting the stability of the performance behavior in terms of time structure. Rhythmic stability is calculated as follows:

[0118] , (Formula 3);

[0119] in, The standard deviation of the time interval series. To prevent constants with a denominator of zero. The larger the value, the more stable the playing rhythm;

[0120] S503. Extract performance continuity features to characterize whether there are obvious interruptions, pauses, or discontinuous operations during the performance, reflecting the degree of continuity of the performance behavior. Set a time interval threshold. When satisfied When this occurs, it is determined as a performance interruption event, and the performance continuity is defined as follows:

[0121] , (Formula 4);

[0122] in, This represents the number of interruption events during the performance. The larger the value, the more continuous the performance.

[0123] S504. Extract performance load features to characterize the operational intensity per unit time during performance, reflecting the level of motor and cognitive load generated by the performance behavior on the subject. Performance load is calculated as follows:

[0124] , (Formula 5);

[0125] in, The duration of the performance. The higher the value, the higher the performance load;

[0126] S505. Combine the extracted rhythmic stability features, performance continuity features, and performance load features, and perform normalization to construct a performance behavior feature vector:

[0127] , (Formula 6);

[0128] , , The normalized performance behavior features are used as the input to the subsequent weighted adjustment model to analyze the moderating effect of performance behavior on changes in physiological characteristics.

[0129] S6. Analyze the statistical relationship between performance behavior characteristics and changes in fatigue-sensitive physiological characteristics before and after performance, construct a moderating model of the weights of performance behavior characteristics on physiological characteristics, and calculate the weight moderating factors. The specific steps are as follows:

[0130] S601. Regarding the set of fatigue-sensitive physiological characteristics obtained in step S4. Calculate the changes in physiological characteristics before and after piano playing, assuming the first... The values ​​of each physiological characteristic before the performance were... The value after the performance Then its original change is defined as:

[0131] , (Formula 7);

[0132] The changes in physiological characteristics were standardized, including orientation consistency transformation and scaling transformation. Since different physiological characteristics may exhibit different trends during fatigue relief, orientation consistency transformation was introduced to ensure semantic consistency of the changes in each physiological characteristic in subsequent analyses.

[0133] , (Formula 8);

[0134] After processing, an increase in the values ​​of all physiological characteristic changes indicates a shift towards relief from fatigue. To eliminate differences in the dimensions and numerical ranges of different physiological characteristics, a scaling transformation is performed on the changes in physiological characteristics:

[0135] , (Formula 9);

[0136] in, and They represent the first The mean and standard deviation of each physiological characteristic change in the training sample. To characterize the overall physiological changes caused by the piano performance intervention, the standardized physiological characteristic changes are fused to obtain the total physiological characteristic change.

[0137] S602. Based on multiple performance samples, analyze the statistical relationship between performance behavior characteristics and changes in physiological characteristics. Using the total change in physiological characteristics as the dependent variable and the normalized performance behavior characteristics obtained in step S5 as the independent variable, establish a regression model to obtain the regression coefficients of each performance behavior characteristic, which are the contribution coefficients of each performance behavior characteristic to the total change in physiological characteristics.

[0138] S603. Based on the contribution coefficients obtained from the regression analysis, a weighted summation of the performance behavior characteristics is performed to obtain a comprehensive performance quality score. ;

[0139] S604. Obtain the weighting adjustment factor by mapping the overall performance quality score through an adjustment function. The weight adjustment factor is based on 1 and fluctuates within a preset range, and the adjustment function must ensure that... Tiny changes will not cause The dramatic changes in the sigmoid function enhance the model's stability. Based on this principle, this embodiment selects a translation-scaling variant of the sigmoid function as the adjustment function to calculate the global weight adjustment factor. :

[0140] , (Formula 10);

[0141] in, As a regulation function, this weighting factor is used to dynamically adjust the weights of different physiological characteristics in the fatigue regulation quantification model, reflecting the impact of performance quality on the evaluation of regulation effect.

[0142] S7. Based on the basic weights of fatigue-sensitive physiological characteristics determined in step S4, and combined with the weight adjustment factors obtained in step S6, the changes in each physiological characteristic are weighted and fused to output a comprehensive fatigue regulation index, which is used to quantitatively characterize the regulatory effect of piano performance intervention on mental fatigue. The specific steps are as follows:

[0143] S701, Regarding the set of fatigue-sensitive physiological characteristics obtained in step S4. and the corresponding basic weights Combined with the weight adjustment factor obtained in step S6 The adjusted physiological characteristic weights are calculated as follows:

[0144] , (Formula 11);

[0145] in, and The first Adjusted weights and baseline weights of physiological characteristics, This is a weighting adjustment factor calculated based on the performance quality score;

[0146] S702. Based on the adjusted physiological characteristic weights, the standardized physiological characteristic changes are weighted and fused to output the mental fatigue adjustment index:

[0147] , (Formula 12);

[0148] in, For the first Adjusted weights of physiological characteristics For the first The standardized change in a physiological characteristic. The magnitude of this modulation index is used to characterize the degree to which piano performance intervention modulates mental fatigue. When When, it indicates that after piano performance intervention, the overall physiological state of the subjects showed a change towards fatigue relief compared to before the performance; when When the threshold is reached, it indicates that the overall physiological state has not been alleviated or there is a trend of worsening fatigue. The mental fatigue regulation index can be used as a quantitative result of the effect of a single performance intervention, or it can be used for comparative analysis of the effects of multiple performance interventions.

[0149] Example 2:

[0150] This invention discloses a method for quantitatively evaluating the effect of piano playing on mental fatigue regulation, the specific steps of which are as follows:

[0151] S1. Refer to the corresponding steps in Example 1, which will not be repeated here;

[0152] S2. Refer to the corresponding steps in Example 1, which will not be repeated here;

[0153] In this embodiment, to verify the difference in the effect of piano performance intervention on regulating mental fatigue compared to natural recovery, a resting control condition was also set up. Specifically, after reaching a state of mental fatigue and collecting pulse wave signals under mental fatigue conditions, some subjects performed piano performance intervention according to step S2, while other subjects remained in a seated resting state for the same duration as the piano performance intervention, without performing any other intervention activities.

[0154] Pulse wave signals were collected from both the piano performance intervention group and the resting control group after the intervention for subsequent processing and analysis.

[0155] S3. Preprocess the pulse wave signals acquired in steps S1 and S2, and extract multidimensional physiological features, including heart rate variability features and pulse wave morphological features. The specific steps are as follows:

[0156] S301, The pulse wave signal after filtering using an FIR low-pass filter is as follows: Figure 3 As shown;

[0157] S302, the pulse wave signal after baseline drift removal is as follows: Figure 4 As shown;

[0158] S303. Periodically filter the baseline-removed signals;

[0159] S304. Identify pulse wave periodic feature points;

[0160] S305. Refer to the corresponding steps in Example 1, which will not be repeated here.

[0161] The pulse wave signals collected under the control conditions were also preprocessed and had their physiological features extracted according to the method described in step S3, and were used for comparative analysis with the physiological feature changes under the piano performance intervention conditions. The resting control data were used only for experimental verification and were not involved in the construction of the mental fatigue regulation index. The quantitative evaluation method for the mental fatigue regulation effect of this invention can be implemented independently without relying on control group data.

[0162] S4. Refer to the corresponding steps in Example 1, which will not be repeated here.

[0163] S5. Refer to the corresponding steps in Example 1, which will not be repeated here.

[0164] S6. Refer to the corresponding steps in Example 1, which will not be repeated here.

[0165] S7. Refer to the corresponding steps in Example 1, which will not be repeated here.

[0166] Example 3:

[0167] In this embodiment, a quantitative evaluation system for the effect of piano playing on mental fatigue regulation is provided. The system includes a pulse wave signal acquisition module, a feature extraction module, a weight evaluation module, a weight regulation factor calculation module, and a comprehensive fatigue regulation index calculation module.

[0168] The pulse wave signal acquisition module is used to construct test scenarios and collect pulse wave signals when the subject is awake and mentally fatigued; it also collects behavioral data of the subject during piano playing and pulse wave signals after the performance.

[0169] The feature extraction module is used to extract physiological features from pulse wave signals in the waking state, mental fatigue state, and after the performance, including heart rate variability features and pulse wave morphological features; based on the behavioral data, performance behavior features are extracted, including rhythm stability features, performance continuity features, and performance load features.

[0170] The weight evaluation module is used to establish a classification model based on the random forest algorithm, evaluate the importance of physiological features in the awake state and the mental fatigue state, screen out the set of physiological features sensitive to mental fatigue, and determine the corresponding physiological feature weights according to the importance of each physiological feature in the classification model.

[0171] The weight adjustment factor calculation module is used to perform statistical relationship analysis on the changes in performance behavior characteristics and physiological characteristics sensitive to mental fatigue before and after performance, construct a model for the adjustment of the weight of physiological characteristics by performance behavior characteristics, and calculate the weight adjustment factor.

[0172] The comprehensive fatigue regulation index calculation module is used to perform weighted fusion of the changes in physiological features in the set of physiological features sensitive to mental fatigue based on the weights of physiological features and weight adjustment factors, and output a comprehensive fatigue regulation index to quantitatively characterize the regulatory effect of piano performance intervention on mental fatigue.

[0173] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure can be divided into different functional modules to complete all or part of the functions described above. This system is a quantitative evaluation method for the effect of piano playing on mental fatigue regulation applied to the above embodiments.

[0174] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0175] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for quantitatively evaluating the effect of piano playing on mental fatigue regulation, characterized in that, Includes the following steps: A test scenario was constructed, and pulse wave signals were collected from subjects in both awake and mentally fatigued states. Behavioral data of the subjects during their piano playing and pulse wave signals after the performance were collected. Physiological features, including heart rate variability and pulse wave morphology, were extracted from pulse wave signals during the waking state, mental fatigue state, and after the performance. A classification model is established based on the random forest algorithm. The importance of physiological characteristics in the awake state and the mental fatigue state is evaluated. A set of physiological characteristics sensitive to mental fatigue is selected, and the corresponding physiological characteristic weights are determined according to the importance of each physiological characteristic in the classification model. Based on the behavioral data, performance behavior features are extracted, including rhythmic stability features, performance continuity features, and performance load features. A statistical relationship analysis was conducted on the changes in performance behavior characteristics and physiological characteristics sensitive to mental fatigue before and after performance. A moderating model of the weight of performance behavior characteristics on physiological characteristics was constructed, and the weight moderating factor was calculated. Based on physiological feature weights and weight adjustment factors, the changes in physiological features in the set of physiological features sensitive to mental fatigue are weighted and fused to output a comprehensive fatigue regulation index for quantitatively characterizing the regulatory effect of piano performance intervention on mental fatigue.

2. The method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, The constructed test scenario involves collecting pulse wave signals from subjects in both awake and mentally fatigued states, specifically: Subjects were kept in a seated, resting position, and pulse wave signals were collected in the conscious state using a pulse wave signal sensor. By subjecting subjects to a continuous mental load through pre-set cognitive tasks, the subjects are made to experience mental fatigue. After completing the cognitive task, the subjects' pulse wave signals were collected as pulse wave signals under mental fatigue state. The pulse wave signals collected in both conscious and mentally fatigued states are labeled and stored.

3. The method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, The collection of behavioral data during the subject's piano playing process and pulse wave signals after the performance specifically includes: Subjects were instructed to play the piano according to pre-set performance instructions; During piano performance, performance behavior data is collected synchronously, including key trigger events generated during the performance and their corresponding time information; The pulse wave signal of the subject was collected after the performance.

4. The method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, Before extracting physiological characteristics from pulse wave signals in the conscious state, mental fatigue state, and after a performance, the pulse wave signals are preprocessed, specifically as follows: The pulse wave signal is low-pass filtered to remove high-frequency noise; The baseline curve is fitted by cubic spline interpolation and then removed from the original pulse wave signal to obtain the baseline-drift-free pulse wave signal. The pulse wave signal after baseline drift removal is periodically filtered to eliminate abnormal pulse wave periods; Perform pulse wave feature point identification, including the starting point, the peak of the main wave, and the peak of the diphtheria wave.

5. A method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 4, characterized in that, The extraction of physiological characteristics from the awake state, mental fatigue state, and pulse wave signal after the performance specifically includes: Based on the identified pulse wave feature points, multi-dimensional physiological features are extracted, including heart rate variability features and pulse wave morphological features, specifically: Based on the identified main wave peak, a pulse cycle interval sequence is calculated to extract heart rate variability features, which include time-domain features and frequency-domain features. Simultaneously, pulse wave morphological features are calculated based on the time difference, amplitude difference, and waveform area of ​​the pulse wave feature points, which include time features, amplitude features, and area features.

6. The method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, The importance of physiological characteristics in the awake and mentally fatigued states is assessed to obtain a set of physiological characteristics sensitive to mental fatigue. The weights of these physiological characteristics are then determined based on their importance in the classification model. Specifically: Significance analysis was performed on the physiological characteristics of the awake and mentally fatigued states. If the physiological characteristic data conformed to a normal distribution, a T-test was used for significance analysis; otherwise, a Wilcoxon signed-rank test was used. Based on the significance analysis results, physiological characteristics that did not differ significantly between the awake and mentally fatigued states were removed. The physiological characteristics selected through significance analysis were used as input, and the random forest algorithm was used to train the mental fatigue state discrimination model. Feature selection is performed using recursive feature elimination to obtain the feature subset with the best classification effect, and the corresponding physiological feature weights are calculated based on the feature importance output by random forest.

7. The method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, The extraction of performance behavior features based on the behavioral data specifically includes: Based on behavioral data, a time series of key presses during performance is constructed, specifically as follows: ； in, Indicates the first The time it takes for the button to be pressed. This indicates the total number of key events generated during the performance. A key event is a valid trigger event that meets the minimum press duration or force threshold. Calculate the time interval sequence between adjacent key events, specifically as follows: ， ； Based on the key press time interval sequence, rhythm stability features are extracted, specifically: ； in, The standard deviation of the time interval series. To prevent constants with zero denominators; rhythmic stability characteristics The larger the value, the more stable the playing rhythm; Set time interval threshold When satisfied When this occurs, it is determined as a performance interruption event, and the performance continuity characteristic is defined as follows: ； in, The number of interruption events during the performance; performance continuity characteristics. The larger the value, the more continuous the performance. Extracting performance load features, specifically: ； in, Duration of performance; characteristics of performance load The higher the value, the higher the performance load; The extracted rhythmic stability features, performance continuity features, and performance load features are combined and normalized to construct a performance behavior feature vector, specifically: ； in, , , These are the characteristics of the performance behavior after normalization.

8. A method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, The statistical relationship between performance behavior characteristics and changes in physiological characteristics sensitive to mental fatigue before and after performance is analyzed. A moderating model of the influence of performance behavior characteristics on the weights of physiological characteristics is constructed, and the weight moderating factors are calculated. Specifically: A set of physiological characteristics sensitive to mental fatigue Let the first The values ​​of each physiological characteristic before the performance were... The value after the performance The original changes in physiological characteristics are defined as: ； The original variations in physiological characteristics are standardized, including orientation consistency transformation and scaling transformation, specifically as follows: ； ； in, and They represent the first The mean and standard deviation of the changes in each physiological characteristic in the training sample; The standardized changes in physiological characteristics are then fused to obtain the total changes in physiological characteristics. Based on multiple performance samples, the statistical relationship between performance behavior characteristics and changes in physiological characteristics was analyzed. The total change in physiological characteristics was used as the dependent variable, and the performance behavior characteristics were used as the independent variable. A regression model was established to obtain the contribution coefficient of each performance behavior characteristic. A comprehensive performance quality score is obtained by weighting and summing the performance behavior characteristics based on the contribution coefficient; The overall performance quality score is mapped using an adjustment function to obtain a weight adjustment factor.

9. A method for quantitatively evaluating the effect of piano playing on mental fatigue regulation according to claim 1, characterized in that, The method involves weighting and fusing the changes in various physiological characteristics based on physiological feature weights and weighting adjustment factors, and outputting a comprehensive fatigue regulation index to quantitatively characterize the effect of piano performance intervention on mental fatigue. Specifically: Based on the established baseline weights of fatigue-sensitive physiological characteristics, and combined with a weight adjustment factor, the adjusted weights of the physiological characteristics are calculated as follows: ； in, and The first Adjusted weights and baseline weights of physiological characteristics, This is a weighting adjustment factor calculated based on the performance quality score; Based on the adjusted physiological characteristic weights, the standardized physiological characteristic changes are weighted and fused to output a mental fatigue regulation index, specifically: ； in, For the first Adjusted weights of physiological characteristics For the first The amount of change after standardization of a physiological characteristic.

10. A quantitative evaluation system for the effect of piano playing on mental fatigue regulation, characterized in that, A quantitative evaluation method for the effect of piano playing on mental fatigue regulation, applicable to any one of claims 1-9, includes a pulse wave signal acquisition module, a feature extraction module, a weight evaluation module, a weight adjustment factor calculation module, and a comprehensive fatigue regulation index calculation module; The pulse wave signal acquisition module is used to construct test scenarios and collect pulse wave signals when the subject is awake and mentally fatigued; it also collects behavioral data of the subject during piano playing and pulse wave signals after the performance. The feature extraction module is used to extract physiological features from pulse wave signals in the waking state, mental fatigue state, and after the performance, including heart rate variability features and pulse wave morphological features. Based on the behavioral data, performance behavior features are extracted, including rhythmic stability features, performance continuity features, and performance load features. The weight evaluation module is used to establish a classification model based on the random forest algorithm, evaluate the importance of physiological features in the awake state and the mental fatigue state, screen out the set of physiological features sensitive to mental fatigue, and determine the corresponding physiological feature weights according to the importance of each physiological feature in the classification model. The weight adjustment factor calculation module is used to perform statistical relationship analysis on the changes in performance behavior characteristics and physiological characteristics sensitive to mental fatigue before and after performance, construct a model for the adjustment of the weight of physiological characteristics by performance behavior characteristics, and calculate the weight adjustment factor. The comprehensive fatigue regulation index calculation module is used to perform weighted fusion of the changes in physiological features in the set of physiological features sensitive to mental fatigue based on the weights of physiological features and weight adjustment factors, and output a comprehensive fatigue regulation index to quantitatively characterize the regulatory effect of piano performance intervention on mental fatigue.

Images