Multi-source physiological information comprehensive feedback type music physiotherapy system, method and equipment

Abstract

The invention discloses a multi-source physiological information comprehensive feedback type music physiotherapy system, method and equipment. The system comprises a biological signal acquisition module, a data processing and analysis module, a music generation and regulation module, a feedback control module, a user interaction module, a receiving and forwarding module and the like. A physiological signal of a user is collected in real time through wearable equipment; denoising the physiological signals and extracting features, and fusing to generate a comprehensive pressure index (CSI); personalized physiotherapy music is matched or generated according to the CSI, and physiotherapy parameters are dynamically adjusted; music physiotherapy is carried out, physiological feedback is monitored in real time, and loop optimization is carried out until the optimal music physiotherapy state is achieved; synchronous collection of multiple physiological indexes is achieved through the wearable device and the like, and the music physiotherapy effect is enhanced by establishing a multi-stage intervention system and dynamic closed-loop regulation and control.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-source physiological information integrated feedback music therapy, and in particular to a multi-source physiological information integrated feedback music therapy system, method and equipment. Background Art

[0002] Music therapy, as a non-drug intervention, has formed a relatively complete theoretical system in the field of mental health. Existing technologies mainly use a preset music library for one-way output. For example, the invention application with application number 202411206117.5 discloses a cloud computing-based music therapy remote monitoring method. Its solution focuses on the user's target EEG signal and performs music therapy remote monitoring on the user, achieving certain technical effects. However, its single biofeedback mechanism has significant limitations: 1) The physiological monitoring dimension is single, and relying solely on EEG signals cannot accurately identify complex psychological states (such as the intersection of post-traumatic stress and occupational burnout); 2) The intervention strategy is static, using fixed parameter mapping rules, which is difficult to adapt to the dynamic psychological fluctuations of people in high-stress occupations; 3) There is a lack of neurophysiological coordinated regulation, and EEG analysis and autonomic nervous function assessment are not integrated, resulting in a long anxiety relief cycle.

[0003] Therefore, there is an urgent need to build a multi-dimensional music intervention system tailored to the characteristics of high-stress occupations: through wearable devices, etc., to achieve simultaneous collection of multiple physiological indicators, and through the establishment of a multi-level intervention system and dynamic closed-loop regulation, the effect of music therapy can be enhanced. Summary of the Invention

[0004] In response to the above-mentioned problems, the purpose of the present invention is to provide a multi-source physiological information comprehensive feedback music therapy system, method and equipment to achieve multi-source physiological information collection, multi-level intervention, and dynamic closed-loop music therapy.

[0005] Embodiments of the present invention provide a multi-source physiological information comprehensive feedback music therapy system, method and device.

[0006] Aspect 1: A multi-source physiological information integrated feedback music therapy system, comprising:

[0007] Physiological signal acquisition module, used to collect multi-source physiological signals of users in real time;

[0008] The data processing and analysis module is used to perform denoising, feature extraction, and fusion analysis on the received multi-source physiological signals to generate a comprehensive stress index (CSI) of the user's physical and mental state;

[0009] Music generation and control module, used to generate therapeutic music based on the comprehensive stress index;

[0010] Feedback control module: real-time monitoring of user response to therapy music, forming a closed-loop control and optimizing therapy strategies;

[0011] User interaction module: provides a visual interface to display physiological signals, music parameters and treatment effects;

[0012] Receiving and forwarding module, used to receive, store and forward physical therapy data in real time, generate and push physical therapy task operation reports;

[0013] Among them, the biological signal acquisition module, data processing and analysis module, music generation and regulation module, feedback control module, user interaction module and receiving and forwarding module work together to perform physical therapy tasks.

[0014] Optionally, the biosignal acquisition module includes a multimodal sensor unit and a signal preprocessing unit, wherein:

[0015] A multimodal sensor unit that collects heart rate variability, brain waves, skin conductance, respiratory rate, and blood oxygen saturation signal data and supports wireless transmission;

[0016] The signal preprocessing unit uses wavelet transform and Kalman filtering to denoise the collected signals and extract effective physiological signal features.

[0017] Optionally, the data processing and analysis module includes a feature fusion unit and a state classification unit, wherein:

[0018] Feature fusion unit: converts effective physiological signals into comprehensive stress index (CSI) through weighted fusion algorithm;

[0019] State classification unit: Maps effective physiological signals and comprehensive stress index into state labels based on machine learning models.

[0020] Optionally, the music generation and control module includes a music library unit and a parameter mapping unit, wherein:

[0021] Music library unit, used to store various types of therapy music;

[0022] The parameter mapping unit is used to match therapeutic music according to state labels based on preset rules, or to synthesize personalized music in real time through a generative adversarial network (GAN).

[0023] Optionally, the feedback control module includes a real-time monitoring unit and a dynamic adjustment unit, wherein:

[0024] Real-time monitoring unit, used to continuously collect physiological signals during music therapy and calculate the therapy effect score;

[0025] The dynamic adjustment unit is used to dynamically switch music types or adjust personalized music parameters based on the therapy effect score.

[0026] Optionally, it also includes a cloud knowledge base module, which includes a knowledge base unit and a solution push unit.

[0027] The cloud library unit has a physical therapy knowledge base that is used to collect user physical therapy history data, physical therapy plans, expert experience, and physical therapy music;

[0028] The plan optimization unit is used to form a physiotherapy optimization reference plan based on the physiotherapy knowledge base data and push it to the user.

[0029] Optionally, the real-time feedback module includes a query monitoring unit, a word segmentation unit, a keyword extraction unit, and a solution push unit, wherein:

[0030] The inquiry monitoring unit is used to monitor questions raised by physical therapy users in real time and automatically capture the question text;

[0031] The word segmentation unit is used to segment the question text, identify the core content of the question, and find out the keywords;

[0032] Keyword extraction unit, used to extract keywords, perform intelligent matching, query and retrieve relevant solutions;

[0033] Solution push unit pushes problem solutions to physical therapy users.

[0034] A second aspect: A multi-source physiological information integrated feedback music therapy method, comprising the following steps:

[0035] S1, collect the user's physiological signals in real time through wearable devices;

[0036] S2, denoise the signal and extract features, and fuse them to generate a comprehensive stress index (CSI);

[0037] S3, matching or generating personalized therapy music based on CSI, and dynamically adjusting therapy parameters;

[0038] S4. Perform music therapy and monitor physiological feedback in real time, and optimize the cycle until the optimal music therapy state is achieved.

[0039] Optionally, the CSI classification types include: “relaxation”, “tension” and “anxiety”, and step S3 specifically includes:

[0040] If the CSI classification type is "anxiety", low-frequency alpha wave rhythm music and natural soundscapes are selected, with an initial volume no higher than 50dB; the CSI is reassessed every 2 minutes, and if the anxiety index drops by less than 10%, the alpha wave component sound is increased; if three consecutive adjustments are ineffective, the system switches to dynamic music generation mode and synthesizes a customized soundtrack through GAN.

[0041] A third aspect: An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method provided in the second aspect are implemented.

[0042] A fourth aspect: A non-transitory computer-readable storage medium having a computer program stored thereon, which implements the steps of the method provided in the second aspect when executed by a processor.

[0043] Beneficial effects of the present invention:

[0044] 1. The present invention constructs a comprehensive stress index (CSI) model through a weighted fusion algorithm of heart rate variability (HRV), electroencephalogram (EEG), skin conductance (SCL), respiratory rate (RR) and blood oxygen saturation (SpO2), thereby improving the recognition accuracy of complex psychological states. It adopts wavelet transform and Kalman filter denoising technology to effectively eliminate motion artifact interference and improve the signal-to-noise ratio of feature extraction. It can accurately identify multi-dimensional physiological states and realize the synchronous acquisition of multiple physiological indicators. By establishing a multi-level intervention system and dynamic closed-loop control, the effect of music therapy is enhanced.

[0045] 2. This invention establishes a dynamic mapping rule base between multiple physiological information and music parameters, enabling millisecond-level response adjustments to music parameters. Using a generative adversarial network (GAN) to synthesize personalized soundtracks, this method accelerates the rate of decrease in anxiety index. A closed-loop feedback mechanism implements iterative parameter optimization every two minutes, shortening the time to achieve optimal therapeutic state, enabling dynamic music intervention optimization, and further enhancing the effectiveness of music therapy.

[0046] 3. The layered therapy of the present invention adopts the synergistic mechanism of α-wave induced music and autonomic nervous system regulation. Through the technology of phase synchronization of music rhythm and physiological information, the parasympathetic nerve activity is enhanced, the deep relaxation state is quickly induced, the neurophysiological coordinated regulation is achieved, and the effect of music therapy is further enhanced.

[0047] 4. The cloud-based knowledge base of the present invention supports transfer learning of physical therapy cases, which shortens the adaptation time for new users. The intelligent hardware integration solution (wearable device + relaxation chair) enhances the effect of music intervention through vibration tactile feedback, improves user treatment compliance, and the real-time feedback module adopts semantic parsing technology to improve the accuracy of question response and significantly enhance the efficiency of human-computer interaction.

[0048] 5. Tests on high-stress groups showed that this system significantly reduced the scores on the Post-Traumatic Stress Disorder Scale and prolonged the duration of therapeutic effects, indicating that it has high application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the structure of the multi-source physiological information comprehensive feedback music therapy system of the present invention;

[0050] Figure 2 Schematic diagram of the process of the multi-source physiological information comprehensive feedback music therapy method of the present invention;

[0051] Figure 3 This is a schematic diagram of the machine learning model training and deployment process of the present invention;

[0052] Figure 4 Schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0053] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0054] The existing single physiological feedback mechanism has significant limitations. It cannot accurately identify complex psychological states, is difficult to adapt to the dynamic psychological fluctuations of people in high-stress occupations, lacks neurophysiological coordinated regulation, and does not integrate SCL or HRV analysis with autonomic nervous function assessment.

[0055] In order to solve the above problems, the present invention provides a multi-source physiological information comprehensive feedback music therapy system. Figure 1 This is a structural diagram of a music therapy system provided by an embodiment of the present invention, which includes a basic layer, a service layer and a display layer.

[0056] Among them, the basic layer provides the basic environment for the operation of the music therapy system, which is mainly physical resources, including computing resources, storage resources, network resources, time system resources and security protection.

[0057] The display layer connects to the client to visually display the operation status of music therapy equipment and therapy tasks, and provides services such as command interaction and problem consultation.

[0058] The service layer includes basic services and business services. The basic services mainly provide the software environment for the music therapy system, including middleware, in-memory database, relational database, time series database and other software operating environments.

[0059] The business services include a biosignal acquisition module, a data processing and analysis module, a music generation and control module, a feedback control module, a user interaction module, and a reception and forwarding module. This paper introduces the music therapy system based on the modules included in the business services. These modules work together under the control of the music therapy system to jointly perform music therapy tasks.

[0060] The physiological signal acquisition module includes a multimodal sensor unit and a signal preprocessing unit, wherein:

[0061] The multimodal sensor unit is used to collect multi-source data through devices such as heart rate belts, EEG caps, wearable sensors, and optical blood oxygen probe data, and supports wireless transmission.

[0062] The above-mentioned equipment can be used to collect multi-source data such as skin conductivity level (SCL) and heart rate variability (HRV). SCL is one of the important parameters for the control of the autonomic nervous system and is also an objective quantitative indicator reflecting the excitability of the human sympathetic nervous system.

[0063] HRV can reflect the autonomic nervous system's ability to regulate cardiac activity. The HRV value is affected by many factors, including emotions, exercise, environment, physiological conditions, etc. Emotional states such as emotional instability, anxiety, and tension may lead to a decrease in HRV, while a healthy lifestyle and exercise can increase the HRV value. In addition, anxiety, depression, excessive stress, etc. will affect HRV.

[0064] Optionally, when the device collects HRV, the sampling frequency is controllable, and HRV (heart rate variability) is evaluated by combining the extraction of the main wave of the pulse wave and the characteristics of the dicrotic wave with multiple indicators.

[0065] Optionally, devices such as an EEG cap, heart rate belt, bracelet, and ear / finger dual-mode acquisition sensors can be integrated into an integrated head-mounted device to reduce the burden of wearing.

[0066] Optionally, the data acquisition devices can be integrated into a smart relaxation chair with a built-in vibration motor. The low-frequency rhythm of the music triggers rhythmic vibrations on the back, enhancing the "music-body" synergy. Alternatively, a piezoelectric respiratory sensor can be embedded in the back of the relaxation chair to monitor the frequency of chest and abdominal movement in real time, cross-validating stress levels with data from wearable devices.

[0067] Wireless transmission is supported between each acquisition device and the music therapy system, and dual-mode communication can be achieved via Wi-Fi or Bluetooth. After the device is bound, the connection will be automatically remembered to ensure stability under high-intensity use.

[0068] The signal preprocessing unit uses wavelet transform and Kalman filtering to denoise the collected signals and extract effective physiological signal features.

[0069] The collected physiological signals are usually interfered by various noises, such as breathing sounds, muscle movement noise and environmental noise, which will seriously affect the analysis and diagnosis of physiological signals.

[0070] Wavelet transform is a time-frequency domain analysis method with excellent time-frequency localization properties. It can effectively separate the different frequency components of a signal and is widely used in the field of physiological signal denoising. The wavelet transform physiological signal denoising method includes signal sampling, wavelet decomposition, noise reduction processing, and signal reconstruction.

[0071] The data processing and analysis module includes a feature fusion unit and a state classification unit, which are used to denoise, extract features and perform fusion analysis on the received multi-source physiological signals to generate a comprehensive stress index of the user's physical and mental state.

[0072] Feature fusion unit: converts physiological signals into comprehensive stress index (CSI) through weighted fusion algorithm;

[0073] For example: by collecting data such as heart rate variability (HRV), skin conductance (SCL), electroencephalogram (EEG), respiratory rate (RR) and blood oxygen saturation (SpO2), the comprehensive stress index (CSI) is calculated through weighted fusion algorithm.

[0074] Specifically,

[0075]

[0076] Among them, w i (t) is the weight parameter, S i (t) is the normalized value of each collected data.

[0077] The state classification unit maps effective physiological signals and the comprehensive stress index into state label values based on a machine learning model. This machine learning model can employ models such as support vector machines (SVMs) or long-term time-lapse (LSTM) models. First, feature engineering techniques are used to convert effective physiological signals and the comprehensive computed stress index (CSI) into vector features. This meticulous feature engineering constructs an efficient and powerful vector feature set, providing a rich information foundation for the machine learning model to accurately capture the inherent patterns and regularities of the data.

[0078] The data after feature engineering is input into the machine learning model for model training. During the training process, the model algorithm will learn the mapping relationship between the data based on the input vector features and sample labels, and build a prediction model. It will then be trained, tested, and evaluated based on the training set, test set, and validation set.

[0079] The model is evaluated to check whether its performance meets the requirements. Evaluation indicators usually include accuracy, recall rate, and F1 score.

[0080] Specifically, the model training and deployment steps are as follows: Figure 3 Shown, including:

[0081] S11. Use feature engineering to convert effective physiological signals and comprehensive calculated stress index (CSI) into feature vectors;

[0082] S12. Select an appropriate machine learning model;

[0083] S13. Use the extracted feature vectors and other relevant features to perform model training, testing, and evaluation;

[0084] S14. Obtain the optimized model and deploy it to the music therapy system.

[0085] In the embodiments of the present invention, feature engineering plays a vital role. It is a bridge connecting the comprehensive computational stress index (CSI) and the machine learning model, and has a key impact on improving model performance, accuracy and generalization ability.

[0086] Feature engineering can include data preprocessing, feature construction, feature conversion, feature encoding, feature selection and other techniques. By converting data into features that can better represent the underlying problem, machine learning performance can be improved.

[0087] Through the state classification unit, the effective physiological signals and comprehensive stress index are mapped into state labels, including state label values such as "relaxation", "tension" and "anxiety".

[0088] The music generation and control module includes a music library unit and a parameter mapping unit, which is used to generate therapeutic music parameters, including rhythm, pitch, timbre and volume, according to the comprehensive stress index.

[0089] The music library unit is used to store various types of therapeutic music; the music library unit architecture adopts a layered design including a multi-dimensional music classification system such as the basic treatment layer, cultural adaptation layer and generational preference layer.

[0090] The basic therapeutic music includes:

[0091] Alpha wave induction music (8-12Hz), sine wave base music, superimposed natural white noise (rain / waves), etc., focus on soothing basic physical therapy.

[0092] Theta wave deep relaxation (4-7Hz) binaural beats, combined with low-frequency cello (<200Hz), for deep relaxation;

[0093] Gamma pulse waves (around 40Hz) combined with mixed clock ticking sounds (rhythm error <5ms) achieve a focus-enhancing effect;

[0094] The cultural adaptation layer adapts to regional characteristics, such as local opera (Huangmei Opera singing frequency analysis library)

[0095] The generational preference layer can adapt music templates according to users born in the 2000s or 1990s.

[0096] The entire music library unit implements a dynamic music management mechanism, intelligently labels various types of music, and dynamically updates them.

[0097] The parameter mapping unit is used to match therapeutic music according to state labels based on preset rules, or to synthesize personalized music in real time through a generative adversarial network (GAN).

[0098] A state label-music parameter mapping rule table can be established, and a standardized mapping library can be established based on the correlation research between physiological state labels (relaxation / tension / anxiety) and music acoustic parameters.

[0099] For example: Anxiety state is associated with: low-frequency rhythm (60-80BPM), natural soundscape (rain / flowing water), and alpha wave superposition (8-12Hz); tension state is associated with: gradual rhythm acceleration (80-100BPM), and string music crescendo to guide attention shift; relaxation state is associated with: stable medium-frequency rhythm (70-90BPM), and a lyric-free chanting tone.

[0100] At the same time, dynamic rules are introduced to optimize parameter thresholds based on the user's comprehensive stress index, support customized preferences, and generate adversarial network (GAN) architecture.

[0101] The input of the GAN generator is the current user's comprehensive stress index (CSI index) and the user (age / gender / medical history), and the output is a sequence of music parameters (rhythm, pitch, and timbre combination); the GAN discriminator evaluates the effectiveness of the generated music based on physical therapy effect data (such as the degree of decrease in the anxiety index) and synthesizes customized soundtracks (such as the fusion of natural sounds and low-frequency rhythms preferred by the user).

[0102] The feedback control module includes a real-time monitoring unit and a dynamic adjustment unit, which monitor the user's response to the therapy music in real time, forming a closed-loop control and optimizing the therapy strategy;

[0103] Real-time monitoring unit, used to continuously collect physiological signals during music therapy and calculate the therapy effect score;

[0104] Calculate real-time relative pressure index CSI norm , the formula is:

[0105]

[0106] Among them, CSI0 is the pressure index base state (simulated valley value), CSI max is the stress index state (simulated peak value), and CSI(t) is the real-time comprehensive calculated stress index.

[0107] Real-time relative pressure index (CSI) norm , which can realize the scoring of physical therapy effects.

[0108] The dynamic adjustment unit is used to dynamically switch music types or adjust personalized music parameters based on the therapy effect score.

[0109] The user interaction module includes an emergency handling unit and a display viewing unit, which are used to handle emergency situations, provide a visual interface, and display physiological signals, music parameters and treatment effects.

[0110] Emergency response unit, used to receive and handle emergencies during physical therapy.

[0111] The display unit dynamically displays various data during the treatment process. The system performs statistical analysis on the treatment results and presents them visually in a variety of chart styles. For example, it can perform a simple analysis of data such as treatment time, music type, and the Calculated Stress Index (CSI), and then present them more intuitively in the form of a line chart, allowing users to directly see the results of these data and quickly draw conclusions.

[0112] The receiving and forwarding module includes a data receiving unit and a format output unit, which are used to receive, store and forward physical therapy data in real time, and generate and push physical therapy task operation reports.

[0113] Receive data units, based on the therapy data, the system allows to filter out specific data from the database and export it to different file formats, such as Excel, CSV, PDF, etc.

[0114] The format output unit outputs standardized format reports for further analysis, viewing or data sharing.

[0115] Furthermore, the system also includes a cloud knowledge base module, which includes a knowledge base unit and a solution push unit.

[0116] The knowledge base unit has a physical therapy knowledge base that is used to collect user physical therapy history data, physical therapy plans, expert experience, and physical therapy music;

[0117] The plan push unit is used to form a physiotherapy optimization reference plan based on the physiotherapy knowledge base data and push it to the user.

[0118] Furthermore, the system also includes a real-time feedback module including a query monitoring unit, a word segmentation unit, a keyword extraction unit and a solution push unit, which is used to monitor questions raised by users in real time, automatically capture question texts, perform word segmentation on question texts, identify the core content of questions, and find keywords; and based on keywords, perform intelligent matching, query and retrieve relevant solutions; and push problem solutions to users.

[0119] Inquiry monitoring unit, the music therapy system can provide a WeChat intelligent robot similar to QQ Xiaobing, which is used to monitor the questions raised by users through WeChat in real time. The music therapy system uses the wxauto library to capture the chat history of the WeChat chat window. If there is @current robot, the text will be automatically captured.

[0120] The word segmentation unit uses Pangu word segmentation technology to decompose the question, find out the keywords, perform word segmentation processing, and identify the core content of the question. If the semantic understanding is wrong, it can also call AI large models such as Wenxin Yiyan to re-analyze the semantics.

[0121] The keyword extraction unit extracts keywords from questions for intelligent matching, and uses keywords to query and retrieve relevant solutions.

[0122] The solution push unit is responsible for pushing specific solutions to users and providing real-time feedback to solve user problems.

[0123] like Figure 2 As shown, the multi-source physiological information comprehensive feedback music therapy method based on the above system includes the following steps:

[0124] S1, collect the user's physiological signals in real time through wearable devices;

[0125] S2, denoise the signal and extract features, and fuse them to generate a comprehensive stress index (CSI);

[0126] S3, matching or generating personalized therapy music based on CSI, and dynamically adjusting therapy parameters;

[0127] S4. Perform music therapy and monitor physiological feedback in real time, and optimize the cycle until the optimal music therapy state is achieved.

[0128] The CSI classification types include: "relaxed", "tense" and "anxious". Step S3 specifically includes:

[0129] If the CSI classification type is "anxiety", low-frequency rhythm music and natural soundscapes are selected, with an initial volume not higher than 50dB; the CSI is reassessed every 2 minutes, and if the anxiety index drops by less than 10%, the alpha wave component sound is increased; if three consecutive adjustments are ineffective, the system switches to dynamic music generation mode and synthesizes a customized soundtrack through GAN.

[0130] Application example: The user wears an EEG cap, a heart rate belt, a bracelet, and ear / finger dual-mode acquisition sensors. The system collects data such as heart rate variability (HRV), skin conductance (SCL), electroencephalogram (EEG), respiratory rate (RR), and blood oxygen saturation (SpO2) in real time.

[0131] A weighted fusion algorithm was used to convert the CSI into a comprehensive calculation. After preprocessing, the CSI was calculated as "anxiety." The music generation module matched the preset alpha wave music (frequency 8-12Hz) with the volume set to 50dB. After two minutes of playback, the CSI dropped by 8%. The system automatically increased the alpha wave volume to 55dB. After five minutes, the CSI dropped to the "relaxation" threshold, concluding the treatment and generating a report.

[0132] The present invention also provides an electronic device, Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, the electronic device may include: a processor, a communications interface, a memory, and a communication bus, wherein the processor, the communications interface, and the memory communicate with each other via the communication bus. The processor may call logic instructions in the memory, for example, to execute the following method:

[0133] S1, collect the user's physiological signals in real time through wearable devices;

[0134] S2, denoise the signal and extract features, and fuse them to generate a comprehensive stress index (CSI);

[0135] S3, matching or generating personalized therapy music based on CSI, and dynamically adjusting therapy parameters;

[0136] S4. Perform music therapy and monitor physiological feedback in real time, and optimize the cycle until the optimal music therapy state is achieved.

[0137] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0138] An embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method provided in each of the above embodiments is implemented, for example, including:

[0139] S1, collect the user's physiological signals in real time through wearable devices;

[0140] S2, denoise the signal and extract features, and fuse them to generate a comprehensive stress index (CSI);

[0141] S3, matching or generating personalized therapy music based on CSI, and dynamically adjusting therapy parameters;

[0142] S4. Perform music therapy and monitor physiological feedback in real time, and optimize the cycle until the optimal music therapy state is achieved.

[0143] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0144] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0145] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A multi-source physiological information integrated feedback music therapy system, characterized in that: include: Physiological signal acquisition module, used to collect multi-source physiological signals of users in real time; The data processing and analysis module is used to perform denoising, feature extraction, and fusion analysis on the received multi-source physiological signals to generate a comprehensive stress index (CSI) of the user's physical and mental state; Music generation and control module, used to generate therapeutic music based on the comprehensive stress index; Feedback control module: real-time monitoring of user response to therapy music, forming a closed-loop control and optimizing therapy strategies; User interaction module: provides a visual interface to display physiological signals, music parameters and treatment effects; Receiving and forwarding module, used to receive, store and forward physical therapy data in real time, generate and push physical therapy task operation reports; Among them, the biological signal acquisition module, data processing and analysis module, music generation and regulation module, feedback control module, user interaction module and receiving and forwarding module work together to perform physical therapy tasks.

2. The comprehensive feedback music therapy system according to claim 1, characterized in that: The biological signal acquisition module includes a multimodal sensor unit and a signal preprocessing unit, wherein: A multimodal sensor unit that collects heart rate variability, brain waves, skin conductance, respiratory rate, and blood oxygen saturation signal data and supports wireless transmission; The signal preprocessing unit uses wavelet transform and Kalman filtering to denoise the collected signals and extract effective physiological signal features.

3. The comprehensive feedback music therapy system according to claim 2, characterized in that: The data processing and analysis module includes a feature fusion unit and a state classification unit, wherein: Feature fusion unit: converts effective physiological signals into comprehensive stress index (CSI) through weighted fusion algorithm; State classification unit: Maps effective physiological signals and comprehensive stress index into state labels based on machine learning models.

4. The comprehensive feedback music therapy system according to claim 3, characterized in that: The music generation and control module includes a music library unit and a parameter mapping unit, wherein: Music library unit, used to store various types of therapy music; The parameter mapping unit is used to match therapeutic music according to state labels based on preset rules, or to synthesize personalized music in real time through a generative adversarial network (GAN).

5. The comprehensive feedback music therapy system according to claim 4, characterized in that: The feedback control module includes a real-time monitoring unit and a dynamic adjustment unit, wherein: Real-time monitoring unit, used to continuously collect physiological signals during music therapy and calculate the therapy effect score; The dynamic adjustment unit is used to dynamically switch music types or adjust personalized music parameters based on the therapy effect score.

6. The comprehensive feedback music therapy system according to claim 1, characterized in that: It also includes a cloud knowledge base module, which includes a knowledge base unit and a solution push unit. The cloud library unit has a physical therapy knowledge base that is used to collect user physical therapy history data, physical therapy plans, expert experience, and physical therapy music; The plan optimization unit is used to form a physiotherapy optimization reference plan based on the physiotherapy knowledge base data and push it to the user.

7. The comprehensive feedback music therapy system according to claim 1, characterized in that: The real-time feedback module also includes an inquiry monitoring unit, a word segmentation unit, a keyword extraction unit, and a solution push unit, wherein: The inquiry monitoring unit is used to monitor questions raised by physical therapy users in real time and automatically capture the question text; The word segmentation unit is used to segment the question text, identify the core content of the question, and find out the keywords; Keyword extraction unit, used to extract keywords, perform intelligent matching, query and retrieve relevant solutions; Solution push unit pushes problem solutions to physical therapy users.

8. A multi-source physiological information integrated feedback music therapy method based on the system according to any one of claims 1 to 7, characterized in that: Including steps: S1, collect the user's physiological signals in real time through wearable devices; S2, denoise the signal and extract features, and fuse them to generate a comprehensive stress index (CSI); S3, matching or generating personalized therapy music based on CSI, and dynamically adjusting therapy parameters; S4. Perform music therapy and monitor physiological feedback in real time, and optimize the cycle until the optimal music therapy state is achieved.

9. The multi-source physiological information comprehensive feedback music therapy method according to claim 8, characterized in that: The CSI classification types include: "relaxed", "tense" and "anxious", and step S3 specifically includes: If the CSI classification type is "anxiety," low-frequency alpha wave rhythm music and natural soundscapes are selected, with an initial volume no higher than 50dB. The CSI is reassessed every 2 minutes, and if the anxiety index decreases by less than 10%, the alpha wave component is increased. If three consecutive adjustments are ineffective, the system switches to dynamic music generation mode and synthesizes a customized soundtrack using GAN.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of a multi-source physiological information comprehensive feedback music therapy method as described in any one of claims 8 or 9 are implemented.

Images