Sleep disorder treatment effect evaluation system based on multiple modes

Abstract

The invention discloses a sleep disorder treatment effect evaluation system based on multiple modes, which relates to the technical field of sleep disorder treatment effect evaluation and comprises a multi-mode data acquisition module, a multi-mode data preprocessing module, a multi-mode evaluation module, a sleep disorder treatment monitoring module and a sleep disorder treatment evaluation module, the multi-modal data acquisition module is used for acquiring multi-modal data through acquisition equipment; the multi-modal evaluation module is used for acquiring evaluation indexes of the vital sign data, the questionnaire data and the sleep environment data on the sleep disorder treatment effect. A multi-modal data acquisition technology, a data preprocessing technology and a neural network model construction technology are closely combined with a modern information technology; the problems that in the prior art, the sleep disorder treatment effect cannot be evaluated in combination with multi-modal data, and monitoring of the evaluation effect of the multi-modal data is lacked are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep disorder treatment efficacy evaluation, and in particular to a sleep disorder treatment efficacy evaluation system based on multimodality. Background Art

[0002] Sleep disorders seriously affect people's physical and mental health and quality of life. They not only cause daytime sleepiness and inattention, but also increase health risks such as cardiovascular diseases in the long run. Traditional sleep disorder treatment efficacy evaluation methods have many limitations. On the one hand, relying solely on patient self-reports is highly subjective and affected by factors such as memory bias and emotions, making it difficult to accurately reflect the true sleep status; on the other hand, relying solely on a single physiological monitoring system cannot fully cover the various factors that affect sleep. With the development of science and technology, multimodal data collection and analysis technology has gradually matured. Wearable devices, smart home sensors, etc. can collect rich sleep-related data. The progress of machine learning and deep learning algorithms makes it possible to process and fuse these multi-source heterogeneous data. A multimodal sleep disorder treatment efficacy evaluation system has emerged, aiming to integrate multiple data, overcome the defects of traditional evaluation methods, achieve more accurate and comprehensive efficacy evaluation, and provide strong support for the adjustment of personalized treatment plans, thereby improving the efficacy of sleep disorder treatment and improving the quality of life of patients. Although the existing technology has made great progress in the direction of sleep disorder treatment efficacy evaluation, there are still some problems that need to be optimized. The existing sleep disorder treatment efficacy evaluation system based on multimodality cannot combine multimodal data to evaluate the effect of sleep disorder treatment. The lack of monitoring of the evaluation effect of multimodal data leads to low accuracy in the formulation of sleep treatment plans and failure to achieve the target sleep treatment effect. Summary of the invention

[0003] To achieve the above objectives, the present invention is implemented through the following technical solutions: a multimodal sleep disorder treatment efficacy evaluation system, comprising a multimodal data acquisition module, a multimodal data preprocessing module, a multimodal evaluation module, a sleep disorder treatment monitoring module and a sleep disorder treatment evaluation module, wherein each module is communicatively connected; The multimodal data acquisition module acquires multimodal data through an acquisition device, wherein the multimodal data includes vital sign data, questionnaire data and sleep environment data, and provides basic information support for the evaluation of the therapeutic effect of sleep disorders; The multimodal data preprocessing module performs preprocessing on the collected multimodal data, focusing on accurately extracting the sleep onset time, awakening times and sleep duration before and after the treatment of sleep disorders from the vital signs data, so as to provide key data support for the subsequent acquisition of the vital signs data for the evaluation index of the efficacy of the treatment of sleep disorders; The multimodal evaluation module is divided into a vital sign unit, a questionnaire unit and a sleep environment unit, wherein the vital sign unit, the questionnaire unit and the sleep environment unit are used to obtain evaluation indexes of vital sign data, questionnaire data and sleep environment data on the efficacy of sleep disorder treatment respectively; The sleep disorder treatment monitoring module constructs a sleep disorder treatment monitoring model through a neural network algorithm; The sleep disorder treatment evaluation module uses the sleep disorder treatment monitoring model to deeply mine the value of multimodal data, output an evaluation index of the multimodal data on the efficacy of sleep disorder treatment, and obtain the evaluation effect of the multimodal data on the efficacy of sleep disorder treatment.

[0004] A further improvement of the technical solution of the present invention is that the sleep monitoring data acquisition module acquires multimodal data through an acquisition device, including: The collection equipment includes a polysomnography monitor, an online questionnaire platform, a temperature sensor, a humidity sensor, a light intensity sensor, and a noise sensor; The vital sign data are the electroencephalogram before and after the sleep disorder treatment; the questionnaire data are the results of the self-assessment questionnaire on sleep quality; and the sleeping environment data include temperature, humidity, light intensity and noise.

[0005] Using polysomnography, electroencephalograms were collected before and after sleep disorder treatment, and vital sign data were obtained.

[0006] A further improvement of the technical solution of the present invention is that the process of obtaining questionnaire data by the sleep monitoring data collection module includes: Use the online questionnaire platform, enter the sleep quality self-assessment questionnaire questions into the online questionnaire platform according to the format requirements of the online questionnaire platform, set the number of answers to 10 questions, set the answer type to single-choice questions and the answer options to very consistent, relatively consistent, average, not quite consistent and very inconsistent, assign 5 points, 4 points, 3 points, 2 points and 1 point to the answer options of very consistent, relatively consistent, average, not quite consistent and very inconsistent respectively, generate a questionnaire link, and push the generated questionnaire link via SMS and WeChat public account. The online questionnaire platform stores the results of the sleep quality self-assessment questionnaire in the database to obtain the questionnaire data.

[0007] A further improvement of the technical solution of the present invention is that the process of obtaining sleep environment data by the sleep monitoring data acquisition module includes: Temperature sensors, humidity sensors, light intensity sensors and noise sensors are used to collect temperature, humidity, light intensity and noise to obtain sleeping environment data.

[0008] A further improvement of the technical solution of the present invention is that the multimodal data preprocessing module preprocesses the collected multimodal data to obtain the sleep onset time, awakening times and sleep duration before and after sleep disorder treatment, including: The collected multimodal data were cleaned, and a bandpass filter of 0.5 Hz to 100 Hz was used to filter out the low-frequency and high-frequency noise of the EEG. The temperature, humidity, light intensity, and noise were smoothed by the sliding average filter method. EEG waveforms include Wave, wave and waves, EEG The wave corresponds to the waking state, The wave corresponds to the light sleep state, The wave corresponds to the deep sleep state. The EEG wave was used as the sleep onset time point, and the sleep onset time corresponding to the EEG before and after the sleep disorder treatment was recorded; Monitor the EEG waveform and record the Wave to Wave, Wave to wave and Wave to The number of changes in the waveform of the wave was used to obtain the number of awakenings before and after the sleep disorder treatment; Recording brain wave waveform without change The corresponding time period of the wave was used to obtain the sleep duration before and after the sleep disorder treatment.

[0009] A further improvement of the technical solution of the present invention is that the process of obtaining the evaluation index of the therapeutic effect of sleep disorders by the vital sign data of the vital sign unit includes: According to the efficacy of sleep disorder treatment, weights of sleep onset time, awakening times and sleep duration are set; The process of calculating the improvement in sleep onset time, awakening frequency, and sleep duration includes: in, , and They are the degree of improvement in the time to fall asleep, the degree of improvement in the number of awakenings, and the degree of improvement in the length of sleep. and are the sleep onset time corresponding to the EEG before and after the treatment of sleep disorders, and The awakening times before and after sleep disorder treatment are respectively, based on the normal sleep duration of 9 hours a day. and The sleep durations before and after sleep disorder treatment were respectively; Since the time to fall asleep, the number of awakenings and the duration of sleep are obtained by analyzing the vital signs data, the evaluation index of the time to fall asleep, the number of awakenings and the duration of sleep on the efficacy of the treatment of sleep disorders is equal to the evaluation index of the vital signs data on the efficacy of the treatment of sleep disorders. The acquisition process is as follows: Among them, E is the evaluation index of vital sign data on the efficacy of sleep disorder treatment, , and are the weights of sleep time, awakening times and sleep duration, , and They are the degree of improvement in the time to fall asleep, the degree of improvement in the number of awakenings, and the degree of improvement in sleep duration.

[0010] A further improvement of the technical solution of the present invention is that the process of obtaining the evaluation index of the questionnaire data on the therapeutic effect of sleep disorders by the questionnaire unit includes: The process of selecting the highest and lowest scores in the sleep quality self-assessment questionnaire results and combining them with the actual sleep quality self-assessment questionnaire results to calculate the evaluation index of the questionnaire data for the efficacy of sleep disorder treatment is as follows: Among them, Q is the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment, The actual sleep quality self-assessment questionnaire results are as follows: The lowest score in the self-assessment questionnaire on sleep quality. It is the highest score in the self-assessment questionnaire on sleep quality.

[0011] A further improvement of the technical solution of the present invention is that the process of obtaining the evaluation index of the sleep environment data on the efficacy of the sleep disorder treatment by the sleep environment unit includes: Set the sleeping environment temperature threshold, sleeping environment humidity threshold, sleeping environment light intensity threshold and sleeping environment noise threshold, and set the weights of temperature, humidity, light intensity and noise according to the treatment effect of sleep disorders; The calculation process of the improvement degree of temperature, humidity, light intensity and noise on the therapeutic effect of sleep disorders is as follows: in, , , and The improvement degree of temperature, humidity, light intensity and noise on the treatment effect of sleep disorders respectively. , , and are temperature, humidity, light intensity and noise, , , and They are respectively the sleeping environment temperature threshold, the sleeping environment humidity threshold, the sleeping environment light intensity threshold and the sleeping environment noise threshold; The process of calculating the evaluation index of the effect of sleep environment data on the treatment of sleep disorders by using the improvement degree of temperature, humidity, light intensity and noise on the treatment effect of sleep disorders and the weights of temperature, humidity, light intensity and noise respectively includes: in, It is an evaluation index of the sleep environment data on the efficacy of sleep disorder treatment. , , and are the weights of temperature, humidity, light intensity and noise respectively, , , and These are the degrees of improvement in the therapeutic effects of sleep disorders by temperature, humidity, light intensity and noise respectively.

[0012] A further improvement of the technical solution of the present invention is that the sleep disorder treatment monitoring module, the process of constructing a sleep disorder treatment monitoring model includes: The multimodal data and its evaluation index of the therapeutic efficacy of sleep disorders were used as a data set and divided into a training set and a test set in a ratio of 7:3; MLP is selected as the neural network structure. The input layer includes six neurons, which receive vital signs data, questionnaire data, temperature, humidity, light intensity and noise. The hidden layer is equipped with Sigmoid function and MSE function. The output layer includes three neurons, which output the evaluation index of the treatment effect of sleep disorders based on vital signs data, questionnaire data and sleep environment data respectively. The neural network model is trained using the training set data, the learning rate is set to 0.01, and the number of iterations is set to 1000. Through iterative training, the nonlinear relationship between the vital sign data and the evaluation index of the efficacy of the treatment of sleep disorders, the nonlinear relationship between the questionnaire data and the evaluation index of the efficacy of the treatment of sleep disorders, and the nonlinear relationship between the sleep environment data and the evaluation index of the efficacy of the treatment of sleep disorders are learned. The output value of the output layer is mapped to the [0,1] interval using the Sigmoid function to obtain the neural network model; The test set data is input into the trained neural network model, and the MSE function is used to evaluate the error between the output value of the neural network model and the actual value. The parameters of the neural network model are adjusted according to the evaluation results, the performance of the neural network model is optimized, and a sleep disorder treatment monitoring model is obtained.

[0013] A further improvement of the technical solution of the present invention is that the process of obtaining the evaluation effect of multimodal data on the therapeutic effect of sleep disorders by the sleep disorder treatment evaluation module includes: Combined with the sleep disorder treatment monitoring model, the evaluation effect of multimodal data on the efficacy of sleep disorder treatment was analyzed. When the evaluation index of vital signs data on the efficacy of sleep disorder treatment was lower than 0.3, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was low; when the evaluation index of vital signs data on the efficacy of sleep disorder treatment was between 0.3 and 0.6, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was medium; when the evaluation index of vital signs data on the efficacy of sleep disorder treatment was higher than 0.6, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was high. When the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is lower than 0.3, the questionnaire data has a low evaluation effect on the efficacy of sleep disorder treatment; when the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is between 0.3 and 0.6, the questionnaire data has a medium evaluation effect on the efficacy of sleep disorder treatment; when the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is higher than 0.6, the questionnaire data has a high evaluation effect on the efficacy of sleep disorder treatment; When the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is lower than 0.3, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is low; when the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is between 0.3 and 0.6, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is medium; when the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is higher than 0.6, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is high.

[0014] The beneficial effects of the present invention are as follows: compared with the traditional multimodal sleep disorder treatment efficacy evaluation system, the multimodal data acquisition technology, data preprocessing technology, and neural network model construction technology in the method of the present invention are closely combined with modern information technology, accurately capturing vital sign data, questionnaire data, and sleep environment data, obtaining the electroencephalogram before and after sleep disorder treatment, the electroencephalogram, sleep quality self-assessment questionnaire results, temperature, humidity, light intensity, and noise data, achieving real-time and comprehensive monitoring of multimodal data in the process of sleep disorder treatment, performing data cleaning and filtering on the collected multimodal data, and constructing a sleep disorder treatment system. The monitoring model calculates the evaluation index of the multimodal data on the efficacy of sleep disorder treatment, analyzes the evaluation effect, and solves the problem that an existing multimodal sleep disorder treatment efficacy evaluation system cannot evaluate the sleep disorder treatment effect in combination with multimodal data, lacks monitoring of the multimodal data evaluation effect, resulting in low accuracy in the formulation of sleep treatment plans and failure to achieve the target sleep treatment effect. The present invention ensures that the dynamic monitoring standards for a multimodal sleep disorder treatment efficacy evaluation system can be refined within a more precise range, so that the monitored data becomes a more accurate indicator under the same conditions. The development and application of this method significantly enhances the level of intelligence in the multimodal sleep disorder treatment efficacy evaluation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0016] Figure 1 The present invention is a block diagram of a multimodal sleep disorder treatment efficacy evaluation system. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] like Figure 1As shown, the present invention provides a sleep disorder treatment efficacy evaluation system based on multimodality, including a multimodal data acquisition module, a multimodal data preprocessing module, a multimodal evaluation module, a sleep disorder treatment monitoring module and a sleep disorder treatment evaluation module, wherein each module is communicatively connected; The multimodal data acquisition module acquires multimodal data through acquisition equipment. The multimodal data includes vital sign data, questionnaire data and sleep environment data, providing basic information support for the evaluation of the efficacy of sleep disorder treatment; The multimodal data preprocessing module performs preprocessing on the collected multimodal data, focusing on accurately extracting the sleep onset time, number of awakenings, and sleep duration before and after treatment of sleep disorders from the vital signs data, providing key data support for the subsequent evaluation index of the efficacy of sleep disorder treatment using vital signs data; The multimodal evaluation module is divided into a vital sign unit, a questionnaire unit and a sleep environment unit, wherein the vital sign unit, the questionnaire unit and the sleep environment unit are used to obtain evaluation indexes of the vital sign data, the questionnaire data and the sleep environment data for the treatment efficacy of sleep disorders respectively; The sleep disorder treatment monitoring module builds a sleep disorder treatment monitoring model through a neural network algorithm; The sleep disorder treatment evaluation module uses the sleep disorder treatment monitoring model to deeply explore the value of multimodal data, output the evaluation index of multimodal data on the efficacy of sleep disorder treatment, and obtain the evaluation effect of multimodal data on the efficacy of sleep disorder treatment.

[0019] Preferably, the sleep monitoring data acquisition module acquires multimodal data through an acquisition device, including: The collection equipment includes a polysomnography monitor, an online questionnaire platform, a temperature sensor, a humidity sensor, a light intensity sensor, and a noise sensor; The vital sign data are the electroencephalogram before and after the sleep disorder treatment; the questionnaire data are the results of the self-assessment questionnaire on sleep quality; and the sleeping environment data include temperature, humidity, light intensity and noise.

[0020] Using polysomnography, electroencephalograms were collected before and after sleep disorder treatment, and vital sign data were obtained.

[0021] Preferably, the sleep monitoring data collection module, the process of obtaining questionnaire data includes: Use an online questionnaire platform. According to the format requirements of the online questionnaire platform, enter the questions of the self-assessment questionnaire on sleep quality into the online questionnaire platform. Set the number of questions to 10, the question type to single-choice questions, and the answer options to very consistent, relatively consistent, average, less consistent, and very inconsistent. Assign 5 points, 4 points, 3 points, 2 points, and 1 point to very consistent, relatively consistent, average, less consistent, and very inconsistent in the answer options respectively. Generate a questionnaire link and push the generated questionnaire link via text message and WeChat official account. The online questionnaire platform stores the results of the self-assessment questionnaire on sleep quality in the database and obtains the questionnaire data.

[0022] Preferably, in the sleep monitoring data acquisition module, the process of obtaining sleep environment data includes: Utilize temperature sensors, humidity sensors, light intensity sensors, and noise sensors to collect temperature, humidity, light intensity, and noise, and obtain sleep environment data.

[0023] Preferably, in the multimodal data preprocessing module, the process of preprocessing the collected multimodal data and obtaining the sleep onset time, number of awakenings, and sleep duration before and after sleep disorder treatment includes: Perform data cleaning on the collected multimodal data. Use a band-pass filter with a frequency range of 0.5 Hz to 100 Hz to filter out low-frequency and high-frequency noise in the electroencephalogram. Smooth the temperature, humidity, light intensity, and noise through the moving average filtering method; The electroencephalogram waveform includes waves, waves and waves. In the electroencephalogram, waves correspond to the waking state, waves correspond to the light sleep state, waves correspond to the deep sleep state. Take waves as the sleep onset time point and record the sleep onset time corresponding to the electroencephalogram before and after sleep disorder treatment; Monitor the waveform of the electroencephalogram and record the number of waveform changes from waves to waves, waves to waves, and waves to waves to obtain the number of awakenings before and after sleep disorder treatment; Record the time period corresponding to waves when the electroencephalogram waveform does not change to obtain the sleep duration before and after sleep disorder treatment.

[0024] Preferably, in the vital sign unit, the process of obtaining the evaluation index of the efficacy of sleep disorder treatment based on vital sign data includes: According to the efficacy of sleep disorder treatment, the weights of sleep onset time, awakening times and sleep duration are set; The process of calculating the improvement in sleep onset time, awakening frequency, and sleep duration includes: in, , and They are the improvement degree of falling asleep time, the improvement degree of awakening times and the improvement degree of sleep duration. and are the sleep onset time corresponding to the EEG before and after the treatment of sleep disorders, and The awakening times before and after sleep disorder treatment are respectively, based on the normal sleep duration of 9 hours a day. and The sleep durations before and after sleep disorder treatment were respectively; Since the time to fall asleep, the number of awakenings and the duration of sleep are obtained by analyzing the vital signs data, the evaluation index of the time to fall asleep, the number of awakenings and the duration of sleep on the efficacy of the treatment of sleep disorders is equal to the evaluation index of the vital signs data on the efficacy of the treatment of sleep disorders. The acquisition process is as follows: Among them, E is the evaluation index of vital sign data on the efficacy of sleep disorder treatment, , and are the weights of sleep time, awakening times and sleep duration, , and They are the degree of improvement in the time to fall asleep, the degree of improvement in the number of awakenings, and the degree of improvement in sleep duration.

[0025] Preferably, the process of obtaining the evaluation index of the efficacy of the questionnaire data on the treatment of sleep disorders in the questionnaire unit includes: The process of selecting the highest and lowest scores in the sleep quality self-assessment questionnaire results and combining them with the actual sleep quality self-assessment questionnaire results to calculate the evaluation index of the questionnaire data for the efficacy of sleep disorder treatment is as follows: Among them, Q is the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment, The actual sleep quality self-assessment questionnaire results are as follows: The lowest score in the self-assessment questionnaire on sleep quality. It is the highest score in the self-assessment questionnaire on sleep quality.

[0026] Preferably, the process of obtaining the evaluation index of the sleep environment data on the efficacy of the sleep disorder treatment by the sleep environment unit includes: Set the sleeping environment temperature threshold, sleeping environment humidity threshold, sleeping environment light intensity threshold and sleeping environment noise threshold, and set the weights of temperature, humidity, light intensity and noise according to the treatment effect of sleep disorders; The calculation process of the improvement degree of temperature, humidity, light intensity and noise on the therapeutic effect of sleep disorders is as follows: in, , , and The improvement degree of temperature, humidity, light intensity and noise on the treatment effect of sleep disorders respectively. , , and are temperature, humidity, light intensity and noise, , , and They are respectively the sleeping environment temperature threshold, the sleeping environment humidity threshold, the sleeping environment light intensity threshold and the sleeping environment noise threshold; The process of calculating the evaluation index of the effect of sleep environment data on the treatment of sleep disorders by using the improvement degree of temperature, humidity, light intensity and noise on the treatment effect of sleep disorders and the weights of temperature, humidity, light intensity and noise respectively includes: in, It is an evaluation index of the sleep environment data on the efficacy of sleep disorder treatment. , , and are the weights of temperature, humidity, light intensity and noise respectively, , , and These are the degrees of improvement in the therapeutic effects of sleep disorders by temperature, humidity, light intensity and noise respectively.

[0027] Preferably, the sleep disorder treatment monitoring module, the process of constructing a sleep disorder treatment monitoring model includes: The multimodal data and its evaluation index of the therapeutic efficacy of sleep disorders were used as a data set and divided into a training set and a test set in a ratio of 7:3; MLP is selected as the neural network structure. The input layer includes six neurons, which receive vital signs data, questionnaire data, temperature, humidity, light intensity and noise. The hidden layer is equipped with Sigmoid function and MSE function. The output layer includes three neurons, which output the evaluation index of the treatment effect of sleep disorders based on vital signs data, questionnaire data and sleep environment data respectively. The neural network model is trained using the training set data, the learning rate is set to 0.01, and the number of iterations is set to 1000. Through iterative training, the nonlinear relationship between the vital sign data and the evaluation index of the efficacy of the treatment of sleep disorders, the nonlinear relationship between the questionnaire data and the evaluation index of the efficacy of the treatment of sleep disorders, and the nonlinear relationship between the sleep environment data and the evaluation index of the efficacy of the treatment of sleep disorders are learned. The output value of the output layer is mapped to the [0,1] interval using the Sigmoid function to obtain the neural network model; The test set data is input into the trained neural network model, and the MSE function is used to evaluate the error between the output value of the neural network model and the actual value. The parameters of the neural network model are adjusted according to the evaluation results, the performance of the neural network model is optimized, and a sleep disorder treatment monitoring model is obtained.

[0028] Preferably, the sleep disorder treatment evaluation module, the process of obtaining multimodal data for evaluating the efficacy of sleep disorder treatment includes: Combined with the sleep disorder treatment monitoring model, the evaluation effect of multimodal data on the efficacy of sleep disorder treatment was analyzed. When the evaluation index of vital signs data on the efficacy of sleep disorder treatment was lower than 0.3, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was low; when the evaluation index of vital signs data on the efficacy of sleep disorder treatment was between 0.3 and 0.6, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was medium; when the evaluation index of vital signs data on the efficacy of sleep disorder treatment was higher than 0.6, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was high. When the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is lower than 0.3, the questionnaire data has a low evaluation effect on the efficacy of sleep disorder treatment; when the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is between 0.3 and 0.6, the questionnaire data has a medium evaluation effect on the efficacy of sleep disorder treatment; when the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is higher than 0.6, the questionnaire data has a high evaluation effect on the efficacy of sleep disorder treatment; When the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is lower than 0.3, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is low; when the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is between 0.3 and 0.6, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is medium; when the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is higher than 0.6, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is high.

[0029] Firstly, the electroencephalogram (EEG) before and after the treatment of sleep disorders, the results of the self-assessment questionnaire on sleep quality, temperature, humidity, light intensity and noise were collected by using a polysomnography monitor, an online questionnaire platform, a temperature sensor, a humidity sensor, a light intensity sensor and a noise sensor. Secondly, the collected multimodal data were preprocessed, and the vital signs data were used to obtain the time to fall asleep, the number of awakenings and the length of sleep before and after the treatment of sleep disorders. Then, the evaluation indexes of the therapeutic efficacy of sleep disorders using the vital signs data, the questionnaire data and the sleeping environment data were obtained by calculation. Next, a sleep disorder treatment monitoring model was constructed using a neural network algorithm. Finally, the sleep disorder treatment monitoring model was used to obtain the evaluation effect of the multimodal data on the therapeutic efficacy of sleep disorders.

[0030] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A sleep disorder treatment efficacy evaluation system based on multimodality, comprising a multimodal data acquisition module, a multimodal data preprocessing module, a multimodal evaluation module, a sleep disorder treatment monitoring module and a sleep disorder treatment evaluation module, wherein: Each module is connected in communication, characterized by: The multimodal data acquisition module acquires multimodal data through an acquisition device, wherein the multimodal data includes vital sign data, questionnaire data, and sleep environment data; The multimodal data preprocessing module preprocesses the collected multimodal data and uses the vital sign data to obtain the time of falling asleep, the number of awakenings and the length of sleep before and after the sleep disorder treatment; The multimodal evaluation module is divided into a vital sign unit, a questionnaire unit and a sleep environment unit, wherein the vital sign unit, the questionnaire unit and the sleep environment unit are used to obtain evaluation indexes of vital sign data, questionnaire data and sleep environment data on the efficacy of sleep disorder treatment respectively; The sleep disorder treatment monitoring module constructs a sleep disorder treatment monitoring model through a neural network algorithm; The sleep disorder treatment evaluation module uses a sleep disorder treatment monitoring model to obtain multimodal data to evaluate the efficacy of sleep disorder treatment.

2. The multimodal sleep disorder treatment efficacy evaluation system according to claim 1, characterized in that: The sleep monitoring data acquisition module acquires multimodal data through an acquisition device, including: The collection equipment includes a polysomnography monitor, an online questionnaire platform, a temperature sensor, a humidity sensor, a light intensity sensor, and a noise sensor; The vital sign data are the electroencephalograms before and after the sleep disorder treatment; the questionnaire data are the results of the self-assessment questionnaire on sleep quality; the sleep environment data include temperature, humidity, light intensity and noise. The polysomnography monitor is used to collect the electroencephalograms before and after the sleep disorder treatment to obtain the vital sign data.

3. The multimodal sleep disorder treatment efficacy evaluation system according to claim 2, characterized in that: The sleep monitoring data collection module, the process of obtaining questionnaire data includes: Use the online questionnaire platform, enter the sleep quality self-assessment questionnaire questions into the online questionnaire platform according to the format requirements of the online questionnaire platform, set the number of answers to 10 questions, set the answer type to single-choice questions and the answer options to very consistent, relatively consistent, average, not quite consistent and very inconsistent, assign 5 points, 4 points, 3 points, 2 points and 1 point to the answer options of very consistent, relatively consistent, average, not quite consistent and very inconsistent respectively, generate a questionnaire link, and push the generated questionnaire link via SMS and WeChat public account. The online questionnaire platform stores the results of the sleep quality self-assessment questionnaire in the database to obtain the questionnaire data.

4. The multimodal sleep disorder treatment efficacy evaluation system according to claim 3, characterized in that: The sleep monitoring data acquisition module acquires sleep environment data in the following process: Temperature sensors, humidity sensors, light intensity sensors and noise sensors are used to collect temperature, humidity, light intensity and noise to obtain sleeping environment data.

5. The multimodal sleep disorder treatment efficacy evaluation system according to claim 4, characterized in that: The multimodal data preprocessing module preprocesses the collected multimodal data to obtain the sleeping time, awakening times and sleeping duration before and after the sleep disorder treatment, including: The collected multimodal data were cleaned, and a bandpass filter of 0.5 Hz to 100 Hz was used to filter out the low-frequency and high-frequency noise of the EEG. The temperature, humidity, light intensity, and noise were smoothed by the sliding average filter method. EEG waveforms include Wave, wave and waves, EEG The wave corresponds to the waking state, The wave corresponds to the light sleep state. The wave corresponds to the deep sleep state. The EEG wave was used as the sleep onset time point, and the sleep onset time corresponding to the EEG before and after the sleep disorder treatment was recorded; Monitor the EEG waveform and record the Wave to Wave, Wave to wave and Wave to The number of changes in the waveform of the wave was used to obtain the number of awakenings before and after the sleep disorder treatment; Recording brain wave waveform without change The corresponding time period of the wave was used to obtain the sleep duration before and after the sleep disorder treatment.

6. The multimodal sleep disorder treatment efficacy evaluation system according to claim 5, characterized in that: The process of obtaining the evaluation index of the efficacy of the treatment of sleep disorders by the vital sign data of the vital sign unit includes: According to the efficacy of sleep disorder treatment, weights of sleep onset time, awakening times and sleep duration are set; The process of calculating the improvement in sleep onset time, awakening frequency, and sleep duration includes: in, , and They are the improvement degree of falling asleep time, the improvement degree of awakening times and the improvement degree of sleep duration. and are the sleep onset time corresponding to the EEG before and after the treatment of sleep disorders, and The awakening times before and after sleep disorder treatment are respectively, based on the normal sleep duration of 9 hours a day. and The sleep durations before and after sleep disorder treatment were respectively; Since the time to fall asleep, the number of awakenings and the duration of sleep are obtained by analyzing the vital signs data, the evaluation index of the time to fall asleep, the number of awakenings and the duration of sleep on the efficacy of the treatment of sleep disorders is equal to the evaluation index of the vital signs data on the efficacy of the treatment of sleep disorders. The acquisition process is as follows: Among them, E is the evaluation index of vital sign data on the efficacy of sleep disorder treatment, , and are the weights of sleep time, awakening times and sleep duration, , and They are the degree of improvement in the time to fall asleep, the degree of improvement in the number of awakenings, and the degree of improvement in sleep duration.

7. The multimodal sleep disorder treatment efficacy evaluation system according to claim 6, characterized in that: The process of obtaining the evaluation index of the efficacy of the questionnaire data on the treatment of sleep disorders by the questionnaire unit includes: The process of selecting the highest and lowest scores in the sleep quality self-assessment questionnaire results and combining them with the actual sleep quality self-assessment questionnaire results to calculate the evaluation index of the questionnaire data for the efficacy of sleep disorder treatment is as follows: Among them, Q is the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment, The actual sleep quality self-assessment questionnaire results are as follows: The lowest score in the self-assessment questionnaire on sleep quality. It is the highest score in the self-assessment questionnaire on sleep quality.

8. The multimodal sleep disorder treatment efficacy evaluation system according to claim 7, characterized in that: The process of obtaining the evaluation index of the sleep environment data on the efficacy of the sleep disorder treatment by the sleep environment unit includes: Set the sleeping environment temperature threshold, sleeping environment humidity threshold, sleeping environment light intensity threshold and sleeping environment noise threshold, and set the weights of temperature, humidity, light intensity and noise according to the treatment effect of sleep disorders; The calculation process of the improvement degree of temperature, humidity, light intensity and noise on the therapeutic effect of sleep disorders is as follows: in, , , and The improvement degree of temperature, humidity, light intensity and noise on the treatment effect of sleep disorders respectively. , , and are temperature, humidity, light intensity and noise, , , and They are respectively the sleeping environment temperature threshold, the sleeping environment humidity threshold, the sleeping environment light intensity threshold and the sleeping environment noise threshold; The process of calculating the evaluation index of the effect of sleep environment data on the treatment of sleep disorders by using the improvement degree of temperature, humidity, light intensity and noise on the treatment effect of sleep disorders and the weights of temperature, humidity, light intensity and noise respectively includes: in, It is an evaluation index of the sleep environment data on the efficacy of sleep disorder treatment. , , and are the weights of temperature, humidity, light intensity and noise respectively, , , and These are the degrees of improvement in the therapeutic effects of sleep disorders by temperature, humidity, light intensity and noise respectively.

9. The multimodal sleep disorder treatment efficacy evaluation system according to claim 8, characterized in that: The sleep disorder treatment monitoring module constructs a sleep disorder treatment monitoring model, including: The multimodal data and its evaluation index of the therapeutic efficacy of sleep disorders were used as a data set and divided into a training set and a test set in a ratio of 7:3; MLP is selected as the neural network structure. The input layer includes six neurons, which receive vital signs data, questionnaire data, temperature, humidity, light intensity and noise. The hidden layer is equipped with Sigmoid function and MSE function. The output layer includes three neurons, which output the evaluation index of the treatment effect of sleep disorders based on vital signs data, questionnaire data and sleep environment data respectively. The neural network model is trained using the training set data, the learning rate is set to 0.01, and the number of iterations is set to 1000. Through iterative training, the nonlinear relationship between the vital sign data and the evaluation index of the efficacy of the treatment of sleep disorders, the nonlinear relationship between the questionnaire data and the evaluation index of the efficacy of the treatment of sleep disorders, and the nonlinear relationship between the sleep environment data and the evaluation index of the efficacy of the treatment of sleep disorders are learned. The output value of the output layer is mapped to the [0,1] interval using the Sigmoid function to obtain the neural network model; The test set data is input into the trained neural network model, and the MSE function is used to evaluate the error between the output value of the neural network model and the actual value. The parameters of the neural network model are adjusted according to the evaluation results, the performance of the neural network model is optimized, and a sleep disorder treatment monitoring model is obtained.

10. The multimodal sleep disorder treatment efficacy evaluation system according to claim 9, characterized in that: The sleep disorder treatment evaluation module obtains the multimodal data for evaluating the effect of sleep disorder treatment, including: Combined with the sleep disorder treatment monitoring model, the evaluation effect of multimodal data on the efficacy of sleep disorder treatment was analyzed. When the evaluation index of vital signs data on the efficacy of sleep disorder treatment was lower than 0.3, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was low; when the evaluation index of vital signs data on the efficacy of sleep disorder treatment was between 0.3 and 0.6, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was medium; when the evaluation index of vital signs data on the efficacy of sleep disorder treatment was higher than 0.6, the evaluation effect of vital signs data on the efficacy of sleep disorder treatment was high. When the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is lower than 0.3, the questionnaire data has a low evaluation effect on the efficacy of sleep disorder treatment; when the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is between 0.3 and 0.6, the questionnaire data has a medium evaluation effect on the efficacy of sleep disorder treatment; when the evaluation index of the questionnaire data on the efficacy of sleep disorder treatment is higher than 0.6, the questionnaire data has a high evaluation effect on the efficacy of sleep disorder treatment; When the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is lower than 0.3, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is low; when the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is between 0.3 and 0.6, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is medium; when the evaluation index of sleep environment data on the efficacy of sleep disorder treatment is higher than 0.6, the efficacy of sleep environment data on the efficacy of sleep disorder treatment is high.

Images