Wearable high school student fatigue monitoring system based on EEG and PPG

Abstract

The invention discloses a wearable fatigue monitoring system for senior high school students based on EEG and PPG, and relates to the field of human factor engineering, the system is based on electroencephalogram signals and pulse wave signals of senior high school students under daily learning and living conditions, the electroencephalogram signals and pulse wave characteristic values are fused and jointly input into a fatigue evaluation algorithm to obtain fatigue state evaluation of the senior high school students. The accuracy of the method reaches 80.6%, and powerful support is provided for guaranteeing physical and psychological healthy development of senior high school students.

Description

Technical Field

[0001] The present invention relates to the field of human factors engineering, and in particular to a wearable high school student fatigue monitoring system based on EEG and PPG signals. Background Art

[0002] With the rapid development of society and the increasing pressure of study, fatigue is becoming increasingly evident among contemporary high school students. In today's society, high school students face intense academic competition and various external distractions, often experiencing a state of high stress. Long periods of study, combined with psychological pressure and excessive use of electronic devices, are making high school students increasingly susceptible to fatigue and depression. This situation can easily impact their physical and mental well-being, leading to a decline in their learning and quality of life, such as sleep quality and classroom performance. Research data shows that the prevalence of chronic fatigue syndrome (CFS) among junior high school students is 4.23%, while the prevalence among high school students is 5.21%, both significantly higher than the general population. Furthermore, prolonged fatigue can severely impact the physical and mental health of high school students, even leading to psychological issues such as anxiety and depression. Therefore, in-depth research into the fatigue status of high school students during their daily studies and lives is crucial for safeguarding their physical and mental health. However, existing technologies often rely solely on a single signal to assess physical fatigue, resulting in low accuracy and a lack of consideration for the specific application scenarios of high school students, making them difficult to implement in practice. Summary of the Invention

[0003] To address the existing technical issues, the present invention aims to overcome the shortcomings of existing technologies and provide a high school student fatigue monitoring system based on EEG and PNG signals. This system utilizes a headband that collects EEG signals from high school students in their daily lives, and a wristwatch that collects pulse wave signals. The signals are aggregated via Bluetooth to the wristwatch's built-in microcontroller, which performs preliminary filtering. The collected signals are then uploaded to a server via the Internet of Things (IoT). The server extracts the effective eigenvalues ​​of the EEG and pulse wave signals and performs a fatigue assessment. Finally, the calculated results and assessment recommendations are displayed via the TCP / IP protocol on a guardian's associated mobile app and the wristwatch worn by the high school student. The system demonstrates high accuracy and feasibility.

[0004] One of the objectives of the present invention is to provide an EEG and PPG signal acquisition and transmission system suitable for high school students' daily learning and life scenarios. The specific technical solution is as follows:

[0005] Step S11: Use a headband to collect EEG signals and use a wristwatch to collect pulse wave signals.

[0006] Step S12: Use the Bluetooth protocol to transmit the EEG signals collected by the headband to the microcontroller in the wristwatch.

[0007] Step S13: Use the main control chip of the wristwatch to perform preliminary filtering on the collected physiological signals.

[0008] Step S14: Use the Internet of Things to upload physiological signals to the server and perform fatigue assessment.

[0009] Step S15: Use the TCP / IP protocol to transmit the evaluation results to the mobile app and then back to the watch.

[0010] Preferably, the headband's primary hardware components in step S11 include a tgam main control chip, an HC-05 Bluetooth communication module, and three dry electrodes; the wristwatch's primary hardware components include an ESP-WROOM-32 main control chip, a MAX30102 physiological signal sensor, a toggle power switch, a vibration motor, a TP4056 power module, a 3.7V lithium battery, and a 0.96-inch OLED display. The wristwatch uses the MAX30102 sensor to monitor PPPG signals using photoplethysmography.

[0011] Preferably, in step S12, the Bluetooth protocol IPV4 is used, and the master-slave mode is adopted, with the main control chip in the wristwatch serving as the master device to monitor the matching device online and receive messages.

[0012] Preferably, in step S13, a bandpass filter with a passband of 0.3 dB and a stopband of 10 dB is used to filter the signal.

[0013] Preferably, in step S14, the MQTT protocol is used to connect the hardware system to the server, and the evaluation method used by the server for fatigue evaluation is a real-time updated K-means algorithm.

[0014] Preferably, in step S15, the mobile app is designed to have two user groups: teachers and parents. On the teacher side, the teacher can view the real-time physical and mental fatigue status of all users in the class and their average classroom concentration; on the parent side, the parent can view the real-time fatigue status of the paired user and obtain the system's assessment and suggestions on the user's fatigue status over a period of time. In the event of a user's sudden accident, both the watch and the mobile app will sound an alarm.

[0015] A second object of the present invention is to provide a fatigue status assessment algorithm. The created fatigue assessment algorithm is applied to the above-mentioned high school student fatigue monitoring system based on EEG and PPG signals. The technical solution is as follows:

[0016] Step S21: Extract eigenvalues ​​from the original signal.

[0017] Step S22: Use the benchmark indicators to evaluate the user's fatigue, make limited corrections to the benchmark indicators based on the collected user's personal data, and generate user-personalized monitoring indicators.

[0018] Step S23: During real-time monitoring, the personalized monitoring indicators are updated in real time to obtain a fatigue assessment algorithm suitable for the individual user.

[0019] Preferably, in step S21, the extracted characteristic values ​​are heart rate, blood oxygen saturation, and alpha and theta waves of brain waves.

[0020] Preferably, in step S22, the indicator quantities are blood oxygen saturation, heart rate variability and α / θ value, and the benchmark indicators are determined by the average indicators of healthy people. After the user's personal data meets the requirements in the time dimension, the benchmark indicators are updated using the K-means algorithm based on the overall changes in the user indicators within the time period.

[0021] Preferably, in step S23, after the user's personal data changes and remains unchanged for a long time or forms a change trend, the indicator is updated in real time to form a K-means clustering center, and an algorithm suitable for the user is obtained: the user's real-time data is clustered using Euclidean distance, and the fatigue assessment score is obtained using normalization processing.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] This paper proposes a wearable fatigue monitoring system for high school students based on EEG and PPG. By combining EEG signals and pulse wave signals, using Internet of Things technology for data exchange, and utilizing clustering algorithms for fatigue assessment, it achieves an accuracy of 70.6% in a real learning environment. This result confirms that the method proposed in this paper has high accuracy and feasibility in fatigue monitoring of high school students, and provides a new method for fatigue assessment under the daily learning and living conditions of high school students.

[0024] This invention provides a method for assessing high school student fatigue based on EEG and PPG signals. By monitoring EEG signals, a student's mental state can be assessed in real time, enabling early detection of signs of mental fatigue. For example, when a student's brain is severely fatigued, the content of alpha waves increases while the content of beta waves decreases. Pulse wave signals can be used to assess a driver's cardiac state during fatigue driving assessment. Fatigue can lead to unstable heart rates, increasing cardiac stress and thus impacting learning ability. Real-time monitoring and analysis of pulse wave signals can detect abnormal changes in heart rate, providing timely warnings of physical fatigue and preventing overexertion. Comprehensive analysis of EEG and pulse wave signals provides a more comprehensive understanding of a student's level of physical and mental fatigue. Integrating these signals into fatigue algorithms can improve assessment accuracy. In real-life scenarios, fatigue monitoring systems can use this information to issue timely warnings, reminding students to rest and advising parents and teachers to take measures to avoid accidents caused by fatigue. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a flow chart of the wearable high school student fatigue monitoring system based on EEG and PPG provided in the present invention;

[0026] Figure 2 This is the overall structure diagram of the wearable high school student fatigue monitoring system based on EEG and PPG provided in the present invention;

[0027] Figure 3 This is a software interface diagram of the wearable high school student fatigue monitoring system based on EEG and PPG provided in the present invention;

[0028] Figure 4 This is a system rendering of the wearable high school student fatigue monitoring system based on EEG and PPG provided in the present invention; DETAILED DESCRIPTION

[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0030] For ease of understanding, the nouns or abbreviations mentioned below are first explained:

[0031] EEG: electroencephalogram.

[0032] PPG: Pulse wave signal.

[0033] The method for monitoring high school students' fatigue driving based on EEG signals and pulse wave signals described in the present invention is aimed at the problem that high school students may experience physical and mental fatigue due to heavy academic workload during their daily study and life, thereby affecting their physical and mental health development. By integrating two physiological parameters, EEG signals and pulse wave signals, the fatigue state is evaluated from more dimensions. Compared with traditional evaluation methods, different features are extracted to improve the accuracy and reliability of the evaluation. Therefore, the example data of the evaluation method provided by the present invention are derived from four subjects. The subjects are male and female students aged between 19 and 21 years old, with a height range of 160-187 cm, a weight range of 45-90 kg, and no history of systemic diseases. They also get enough sleep and rest the night before the test and are prohibited from consuming stimulating beverages such as alcohol and coffee within 24 hours to ensure the authenticity and validity of the test data.

[0034] The testing process is:

[0035] The test involved four participants divided into two groups, one male and one female. Each test lasted two hours. The first group completed a set of low-intensity physical and mental tasks, namely jogging and reciting, with a 1:1 task-to-task ratio, without any long breaks (over 15 minutes). The second group remained well-rested. Both groups completed the Fatigue Severity Scale (FSS) and took a reaction time test after the test. The test was conducted twice, with a one-day interval between each session.

[0036] During the specific testing process, the detailed work flow of the system is as follows:

[0037] The first step is to collect EEG and pulse wave signals.

[0038] In this embodiment, the EEG sensor is placed at FP1 (on the left forehead). This area of ​​the forehead has less hair, allowing for more accurate and clear EEG signals. Furthermore, its proximity to the glasses facilitates blink detection, minimizing blink signal interference in the calculated data. For the pulse wave signal, the PPG sensor is placed on the inside of the wrist.

[0039] The second step is to transmit the EEG signal to the main control chip in the watch.

[0040] In this embodiment, because collecting EEG signals consumes significant energy, the headband uses a Bluetooth module to transmit the raw signals to the wristwatch's main control chip, which then handles initial data processing and all IoT functions. For Bluetooth transmission, the hardware module uses the HC-05 module, employing the IPv4 protocol and master-slave mode. Upon startup, the watch is configured as the Bluetooth server, continuously listening for Bluetooth messages from a specific device ID.

[0041] The third step is to preliminarily process the EEG signal and pulse wave signal.

[0042] In this embodiment, the wristwatch is responsible for preliminary signal processing. For EEG signals, this initial processing primarily involves filtering. Alpha waves (8-13 Hz) and theta waves (4-7 Hz) have specific frequency ranges. A Chebyshev filter with a passband of 4-13 Hz, a 0.3 dB passband, and a 50 dB band-stop was designed using MATLAB. For pulse wave signals, initial processing involves both filtering and processing the raw data to obtain blood oxygen saturation and heart rate. This is accomplished using the MAX30102's built-in photoplethysmography algorithm.

[0043] The fourth step is to upload the physiological signals to the server and perform fatigue assessment.

[0044] In this embodiment, the watch first connects to Wi-Fi to gain internet connectivity. It then establishes a channel with the server using the MQTT protocol with a specified key and transmits messages. Regarding data processing, EEG signals are random. Unlike deterministic signals, random signals lack a clear spectrum, making them difficult to decompose into a series of distinct sinusoidal waves via Fourier transform. Therefore, AMRA (autoregressive moving average) analysis is used. The AMRA model calculates the power spectrum of the measured EEG signal. A baseline index is obtained by normalizing the EEG and pulse wave signals using the variation range of healthy individuals. When the initial data collection reaches an hour, the baseline index is initially corrected. This is done by subtracting the baseline index from the average value of the collected signal, multiplying it by an update factor of 0.2, and adding the resultant value to the baseline index.

[0045] Step 5: Send the fatigue assessment results to the mobile app.

[0046] In this embodiment, after obtaining the fatigue assessment results, the server transmits the data via the TCP protocol to the mobile app logged in with the account bound to the corresponding device.

[0047] In addition, this system also has an emergency handling function. When the user's related indicators show abnormal characteristics, an alarm will be issued. For example, if the heart rate suddenly increases, the account bound to the device will be alerted. At the same time, the vibration motor on the watch will start vibrating to remind students to pay attention to their health.

[0048] In this embodiment, the Fatigue Severity Scale (FFS) and reaction time test are used as objective estimates of the test subject's fatigue level. The Fatigue Severity Scale score ranges from 9 (no fatigue) to 63 (severe fatigue) as an estimate of physical fatigue, and the increase in reaction time ranges from 5% (no obvious fatigue) to 25% (obvious fatigue) as an estimate of mental fatigue. The scores are normalized to a percentage scale of 0-100, with fatigue level decreasing in descending order.

[0049] The test results showed that the average accuracy of the two tests was 80.6%, which was calculated as 1-|system-assessed fatigue score-objective estimate of fatigue level| / objective estimate of fatigue level.

Claims

1. The EEG and PPG signal acquisition and transmission system suitable for high school students' daily learning and life scenes is characterized by: It includes the following five steps: Step S11: Use a headband to collect EEG signals and use a wristwatch to collect pulse wave signals.

2. Step S12: Use the Bluetooth protocol to transmit the EEG signals collected by the headband to the microcontroller in the wristwatch.

3. Step S13: Use the wristwatch's main control chip to perform preliminary filtering on the collected physiological signals.

4. Step S14: Use the Internet of Things to upload physiological signals to the server and perform fatigue assessment.

5. Step S15: Use TCP / IP protocol to transmit the evaluation results to the mobile app and then back to the watch.

6. The EEG and PPG signal acquisition and transmission system according to claim 1, characterized in that: In step S11, the headband's main hardware components include the tgam main control chip, a BLE Bluetooth communication module, and three dry electrodes. The wristwatch's main hardware components include the ESP-WROOM-32 main control chip, a MAX30102 physiological signal sensor, a toggle power switch, a vibration motor, a TP4056 power module, a 3.7V lithium battery, and a 0.96-inch OLED display. The wristwatch uses the MAX30102 sensor to monitor PPPG signals using photoplethysmography.

7. The electric vehicle fatigue driving evaluation method based on EMG and ECG signals according to claim 1, characterized in that: In step S12, the Bluetooth protocol IPV6 is used, and the master-slave mode is adopted. The main control chip in the watch acts as the master device to monitor the matching device online and receive messages.

8. The electric vehicle fatigue driving evaluation method based on EMG and ECG signals according to claim 1, characterized in that: In step S13, a bandpass filter with a passband of 0.3 dB and a stopband of 10 dB is used to filter the signal.

9. The electric vehicle fatigue driving evaluation method based on EMG and ECG signals according to claim 1, characterized in that: In step S14, the MQTT protocol is used to connect the hardware system to the server, and the evaluation method used by the server for fatigue evaluation is the real-time updated K-means algorithm.

10. The electric vehicle fatigue driving evaluation method based on EMG and ECG signals according to claim 1, characterized in that: In step S15, the mobile app is designed to have two user groups: teachers and parents. On the teacher side, you can view the real-time physical and mental fatigue of all users in the class and check their average classroom concentration. On the parent side, you can view the real-time fatigue status of paired users and receive system assessments and recommendations on the user's fatigue status over time. In the event of a user's sudden accident, both the watch and the mobile app will sound an alarm.

11. The second part provides a fatigue status assessment algorithm. The fatigue assessment algorithm is applied to the above-mentioned high school student fatigue monitoring system based on EEG and PPG signals, and includes the following three steps: Step S21: extracting characteristic values ​​from the original signal.

12. Step S22: Use the baseline indicators to evaluate user fatigue, and make limited corrections to the baseline indicators based on the collected user personal data to generate user personalized monitoring indicators.

13. Step S23: During real-time monitoring, the personalized monitoring indicators are updated in real time to obtain a fatigue assessment algorithm suitable for the individual user.

14. The EEG and PPG signal acquisition and transmission system according to claim 1, characterized in that: In step S21, the extracted feature values ​​are heart rate, blood oxygen saturation, and alpha and theta waves of the brain waves.

15. The EEG and PPG signal acquisition and transmission system according to claim 1, characterized in that: In step S22, the indicator quantities are blood oxygen saturation, heart rate variability and α / θ value. The benchmark indicators are determined by the average indicators of healthy people. After the user's personal data meets the requirements in the time dimension, the benchmark indicators are updated using the K-means algorithm based on the overall changes in the user indicators within the time period.

16. The EEG and PPG signal acquisition and transmission system according to claim 1, characterized in that: In step S23, after the user's personal data changes and remains unchanged for a long time or forms a change trend, the indicators are updated in real time to form a K-means clustering center, and an algorithm suitable for the user is obtained: the user's real-time data is clustered using Euclidean distance, and the fatigue assessment score is obtained using normalization processing.

Images