Student brain activity monitoring and learning progress adjusting system based on fNIRS

Abstract

The invention belongs to the technical field of educational nerves, and particularly relates to a student brain activity monitoring and learning progress adjusting system based on fNIRS, comprising: an fNIRS signal acquisition module configured to acquire blood oxygen signals of a prefrontal cortex and a parietal brain region of a student in real time through a multi-channel near-infrared light source and a detector; the data processing and analyzing module is used for carrying out preprocessing, feature extraction and cognitive state classification on the blood oxygen signal and outputting a cognitive load level and an attention state; the adaptive learning engine is used for learning content difficulty and rhythm based on a cognitive state dynamic adjustment strategy, and the dynamic adjustment strategy comprises adjusting task difficulty according to a cognitive load level and controlling learning rhythm according to an attention state; according to the method, blood oxygen signals (HbO / HbR) of brain areas such as the prefrontal lobe and the apical lobe are collected in real time through fNIRS, cognitive loads and attention states are quantified in combination with a machine learning algorithm (SVM + LSTM), and hysteresis and subjective deviation of a traditional method (such as test and observation) are avoided.

Description

Technical Field

[0001] The present invention belongs to the field of educational neurotechnology, and specifically is a student brain activity monitoring and learning progress adjustment system based on fNIRS. Background Art

[0002] Current assessments of student learning status in education primarily rely on subjective methods (such as classroom observations and test scores) or simple physiological indicators (such as eye tracking and heart rate monitoring). These methods are subject to lags, fail to reflect the brain's cognitive activity in real time, and struggle to accurately quantify deeper learning states, such as students' attention allocation and cognitive load.

[0003] Functional near-infrared spectroscopy (fNIRS) is a non-invasive brain imaging technique that indirectly reflects neural activity by detecting changes in blood oxygen concentration (oxyhemoglobin HbO and deoxyhemoglobin HbR) in the cerebral cortex. Compared to fMRI (which requires a fixed position) and EEG (which is susceptible to motion artifacts), fNIRS has the advantages of high portability, strong anti-interference capabilities, and low cost. It has been gradually applied in the fields of neuroscience, rehabilitation medicine, and human-computer interaction. In existing technologies, fNIRS is mostly used in clinical or basic research (such as depression diagnosis and brain-computer interfaces), but its systematic application in educational scenarios is not yet mature.

[0004] In recent years, a few studies have attempted to integrate brain science tools (such as EEG) with education, for example, by monitoring attention through brain waves and triggering reminders. However, these solutions have the following problems:

[0005] Signal limitations: EEG is susceptible to environmental noise and has difficulty locating specific functional areas of the brain (such as the prefrontal cortex, which is associated with higher-order cognition);

[0006] Lack of closed loop: Most systems only implement one-way monitoring and fail to dynamically link with teaching content;

[0007] Lack of universality: It does not take into account the impact of individual differences (such as age and cognitive characteristics) on brain signal interpretation.

[0008] Therefore, to address the above problems, a student brain activity monitoring and learning progress adjustment system based on fNIRS is proposed. Summary of the Invention

[0009] (1) Technical problems solved

[0010] In view of the shortcomings of the existing technology, the present invention provides a student brain activity monitoring and learning progress adjustment system based on fNIRS to solve the problems raised in the background technology.

[0011] (2) Technical solution

[0012] To achieve the above objectives, the present invention provides the following technical solution: a student brain activity monitoring and learning progress adjustment system based on fNIRS, comprising:

[0013] fNIRS signal acquisition module: configured to collect blood oxygen signals (HbO and HbR) of students' prefrontal cortex and parietal brain regions in real time through multi-channel near-infrared light sources and detectors;

[0014] Data processing and analysis module: pre-processes blood oxygen signals, extracts features, and classifies cognitive states, outputting cognitive load levels and attention states;

[0015] Adaptive learning engine: Dynamically adjusts learning content difficulty and pace based on cognitive state. This includes adjusting task difficulty based on cognitive load level and controlling learning pace based on attention state.

[0016] User interaction module: provides a visual monitoring interface for teachers and real-time feedback for students.

[0017] Preferably, the fNIRS signal acquisition module includes:

[0018] Dual-wavelength (760nm and 850nm) near-infrared light source, sampling frequency ≥10Hz;

[0019] The flexible headband integrates an optode array with an optode spacing of 3 cm, covering the prefrontal and parietal cortex;

[0020] Motion artifact suppression unit with built-in accelerometer.

[0021] Preferably, the preprocessing of the blood oxygen signal in the data processing and analysis module includes:

[0022] High-frequency noise and low-frequency drift were removed through a band-pass filter (0.01-0.2 Hz);

[0023] The wavelet transform algorithm combined with accelerometer data eliminates motion artifacts;

[0024] Changes in HbO and HbR concentrations were calculated based on the modified Beer-Lambert law.

[0025] Preferably, the feature extraction includes:

[0026] Extract the time domain features (mean, slope, variance) and frequency domain power spectral density of the prefrontal HbO signal;

[0027] Attention concentration was assessed by changes in the HbO / HbR ratio in the parietal area.

[0028] Preferably, the adaptive learning engine adopts a model that integrates support vector machine (SVM) and long short-term memory network (LSTM), with input parameters including blood oxygen signal dynamic curve, task type and individual historical data, and output as cognitive load level (low / medium / high) and attention state (distracted / concentrated).

[0029] Preferably, the dynamic adjustment strategy includes:

[0030] When high cognitive load is detected, reduce the difficulty of the learning content or insert a review of knowledge points;

[0031] When attention is distracted, shorten the duration of a single study session and trigger interactive sessions;

[0032] Dynamically adjust intervention thresholds based on individual resting-state baseline data.

[0033] Preferably, the user interaction module includes:

[0034] The teacher's terminal displays the whole brain blood oxygen thermogram and the list of students who need intervention in real time;

[0035] The student side embeds an attention value progress bar and a neural feedback reward mechanism.

[0036] A learning method based on fNIRS student brain activity monitoring and learning progress adjustment system, characterized by comprising the following steps:

[0037] Initial baseline calibration: Establishing an individual signal-cognitive state mapping model through resting-state measurements and standardized tasks;

[0038] Real-time monitoring and dynamic adjustment: triggering learning content switching, rhythm control or interactive intervention based on changes in blood oxygen signals;

[0039] Long-term learning path optimization: Generate neural data reports and recommend personalized learning plans.

[0040] Preferably, the dynamic adjustment includes:

[0041] If the prefrontal HbO concentration exceeded 150% of the baseline value and the variance increased, it was determined to be cognitive overload and the difficulty of the questions was reduced;

[0042] If the parietal HbO / HbR ratio was continuously below the threshold of 80%, the attention activation task was initiated.

[0043] A student brain activity monitoring and learning progress adjustment system based on fNIRS can be integrated into smart classrooms or online education platforms, supporting simultaneous monitoring of multiple students and cross-disciplinary task adaptation.

[0044] (3) Beneficial effects

[0045] Compared with the existing technology, the present invention provides a student brain activity monitoring and learning progress adjustment system based on fNIRS, which has the following beneficial effects:

[0046] 1. This invention uses fNIRS to collect real-time blood oxygenation signals (HbO / HbR) from brain regions such as the frontal and parietal lobes. It then combines this with a machine learning algorithm (SVM+LSTM) to quantify cognitive load and attentional states, avoiding the lag and subjective bias of traditional methods (such as testing and observation). Compared to EEG (which is susceptible to motion artifacts) and fMRI (which requires a fixed position), fNIRS equipment is lightweight and resistant to motion interference, making it suitable for the dynamic classroom environment.

[0047] 2. This invention triggers intervention strategies in real time based on changes in brain signals (e.g., reducing difficulty during high-load situations, initiating interaction during distracted situations), forming a closed-loop "monitoring-analysis-adjustment" approach that significantly improves learning efficiency. Dynamically adjusting the judgment threshold based on individual resting-state baseline data avoids a "one-size-fits-all" approach and adapts to the cognitive characteristics of different students.

[0048] 3. The present invention can be integrated into smart classrooms, online education platforms or special education (such as ADHD intervention) to support simultaneous monitoring of multiple students.

[0049] 4. The present invention enhances learning motivation and reduces resistance caused by adjusting strategies through attention progress bars and neurofeedback rewards (such as virtual badges). BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0051] Figure 1 It is a system structure framework diagram of the present invention;

[0052] Figure 2 It is a flow chart of the learning method of the system of the present invention. DETAILED DESCRIPTION

[0053] 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 any creative efforts shall fall within the scope of protection of the present invention.

[0054] Specific examples are given below.

[0055] Example 1: fNIRS-based student brain activity monitoring and learning progress adaptive adjustment system;

[0056] System hardware composition:

[0057] This example uses a portable fNIRS device (model: NIRx NIRSport2) as the core monitoring device, with the following specific configuration:

[0058] Light sources and detectors: 16 light sources (760nm / 850nm dual wavelength) and 16 detectors, arranged at 3cm intervals, covering the prefrontal cortex (Fp1 / Fp2 / Fz) and parietal areas (P3 / P4);

[0059] Motion compensation module: integrated 3-axis accelerometer (sampling rate 100Hz) for motion artifact correction;

[0060] Data transmission: Real-time data transmission to the processing terminal (tablet / server) via Bluetooth 5.0;

[0061] System workflow:

[0062] Step 1: Individualized calibration (before class);

[0063] Students wearing the device complete:

[0064] 5-minute resting-state measurement (eyes closed and relaxed); standardized cognitive task (10 arithmetic problems of increasing difficulty);

[0065] System establishment:

[0066] Baseline blood oxygen level (resting mean HbO: 12.5 ± 2.1 μmol / L); individual cognitive load threshold (e.g., an HbO increase > 35% is considered high load);

[0067] Step 2: Real-time classroom monitoring (taking mathematics class as an example);

[0068] Scenario A: Explanation of complex geometry problems;

[0069] Detected:

[0070] The frontal HbO concentration increased by 42% (surpassing the threshold) within 90 seconds; the parietal HbO / HbR ratio decreased to 0.85 (baseline: 1.2);

[0071] System response:

[0072] The teacher's end pops up an alert: "Student No. 3 is cognitively overloaded"; a simplified version of the problem (without auxiliary lines and added steps) is automatically pushed; the student's end displays a 30-second problem-solving animation;

[0073] Scenario B: Classroom practice period;

[0074] Detected:

[0075] Parietal signal fluctuations <5% for 2 consecutive minutes (distracted attention); increased head movement frequency (accelerometer data >0.5g / s);

[0076] System response:

[0077] Pause the current exercise; start the "Math Maze" game (need to complete 3 basic questions to unlock subsequent content); the difficulty is reduced by 15% after resuming learning;

[0078] Comparison table of typical intervention strategies:

[0079]

[0080] Implementation effect verification:

[0081] In a four-week comparative experiment among the second grade students of a key middle school (n=45):

[0082] Efficiency improvement:

[0083] The effective learning time per unit time increased by 27 minutes per lesson (p<0.01); the average time spent solving geometry proof problems was shortened by 2.3 minutes;

[0084] Improvement of physiological indicators:

[0085] The proportion of high-load time decreased from 38% to 19%; attention concentration (parietal signal stability) increased by 41%;

[0086] Special scene adaptation:

[0087] Special programs for students with ADHD:

[0088] Adjust monitoring parameters (shorten the attention determination interval to 30 seconds); strengthen the feedback mechanism (distribute virtual rewards every 5 minutes); and add a special indicator on the teacher side (flashing reminder).

[0089] This embodiment is implemented through specific hardware configuration, data parameters and algorithms, fully demonstrating the entire process of the system from signal acquisition to educational intervention, and confirming its practicality and effectiveness in real teaching scenarios.

[0090] Example 2: Real-time classroom attention control system;

[0091] Application scenario: Junior high school English reading comprehension class (45 minutes), class size 30 students;

[0092] System Configuration:

[0093] NIRS equipment: An 8-channel headband device (covering the left prefrontal lobe and language area); sampling parameters: 10 Hz sampling rate, wavelength combination (730 nm / 850 nm); auxiliary sensor: eye tracking module (monitoring gaze point);

[0094] Workflow:

[0095] Phase 1: Preparatory monitoring (0-5 minutes);

[0096] The baseline was established when students read the text silently: the mean HbO in the language area (Broca's area) was 15.2±1.8μmol / L; the attention threshold was: the parietal lobe HbO / HbR ratio <1.1 was judged as distraction;

[0097] Stage 2: In-depth reading (15-25 minutes);

[0098] Typical intervention cases:

[0099] Student A: Insufficient activation of Broca's area was detected (HbO increased by only 12%);

[0100] System action:

[0101] Automatically highlight key sentences in the text; push vocabulary annotation pop-up window; the teacher's end prompts "language reinforcement required";

[0102] Student B: eye movement data + parietal signal showing saccades (fixation <100ms / line);

[0103] System action:

[0104] Enforce reading speed limits (scrolling speed reduced by 30%); insert comprehension test questions (one question after each paragraph);

[0105] Phase 3: Group discussion (30-40 minutes);

[0106] Group attention monitoring:

[0107] When >40% of students’ parietal lobe signals decrease, the system automatically triggers:

[0108] The discussion timer is shortened by 2 minutes; the projection screen switches to the mind map template;

[0109] Effect data:

[0110] The accuracy of reading comprehension tests increased by 22%; the time spent analyzing long and difficult sentences decreased by 35%; and the number of manual interventions by teachers decreased by 60%.

[0111] Example 3: Online programming education adaptive system;

[0112] System Architecture:

[0113] Hardware layer:

[0114] Lightweight fNIRS glasses (6 channels, focusing on monitoring the dorsolateral prefrontal cortex); keyboard tapping frequency sensor;

[0115] Software layer:

[0116] Python programming IDE integrated plug-in; 3D brain activity visualization interface;

[0117] Typical intervention scenarios:

[0118] Scenario A: Code debugging dilemma;

[0119] Characteristic signal:

[0120] Frontal HbO levels were persistently >50% of baseline (lasting 8 minutes); code-saving frequency increased dramatically (>5 times / minute);

[0121] System response:

[0122] Automatically call up the "Debug Assistant" panel; recommend similar solution cases; reduce the algorithm complexity of the next exercise;

[0123] Scenario B: Rigid thinking;

[0124] Characteristic signal:

[0125] The brain activation pattern was repeated (similarity > 85% for 10 minutes); the repetition rate of variable naming increased;

[0126] System response:

[0127] Pop up a "reboot your mind" mini-game (code refactoring challenge); inject unconventional programming paradigms (such as functional programming examples);

[0128] Data Validation:

[0129] index Before use After use Improvement rate Code pass rate 62％ 89％ +43％ Debugging takes time 47 minutes 28min -40％ Algorithm Diversity 3.2 types 5.7 types +78％

[0130] Example 4, special education emotion regulation system;

[0131] Target group:

[0132] Children on the autism spectrum (8-12 years old) have frequent mood swings;

[0133] Innovative configuration:

[0134] Enhanced Monitoring:

[0135] 16-channel high-density array (covering the limbic system projection area); simultaneous skin conductance monitoring;

[0136] Interactive devices:

[0137] Haptic feedback wristband; immersive VR environment;

[0138] Intervention model:

[0139] Emotional warning stage:

[0140] When detected:

[0141] HbR surge > 25% in the amygdala projection area; skin conductance > 5μS;

[0142] System startup:

[0143] Wristband vibration warning (graded intensity); VR scene switches to a safe space (such as the underwater world);

[0144] Cognitive reconstruction phase:

[0145] Through neurofeedback training:

[0146] Real-time display of "emotional volcano" animation (eruption intensity corresponds to activation level); deep breathing guidance to control animation effects; reward for reaching the target: virtual pet interaction;

[0147] Clinical Data:

[0148] Frequency of emotional outbursts: decreased from 7.2 times / day to 2.1 times / day;

[0149] Average focus time: increased from 3.5 minutes to 11.2 minutes;

[0150] The social interaction willingness score increased by 2.3 times (Vineland Scale).

[0151] Example 5: Sports training cognitive load management system;

[0152] Application scenarios:

[0153] Tactical training for basketball players (both physical and cognitive);

[0154] Hardware innovation:

[0155] Anti-sweat fNIRS headband (patented hydrophobic material); motion trajectory capture system (100Hz sampling);

[0156] Dynamic adjustment logic:

[0157] Load determination:

[0158] Frontal lobe HbO > 40% + tactical execution error rate > 30% → Overload

[0159] Intervention strategies:

[0160] Simplify the tactical board display (retain only the core movement); reduce the speed of voice commands by 50%; insert basic action exercises (dribbling / shooting);

[0161] Training effect:

[0162] The time to master complex tactics was shortened by 42%; decision-making accuracy increased by 28%; and blood lactate levels decreased by 35% after training (indirect effect of reduced cognitive load).

[0163] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. A student brain activity monitoring and learning progress adjustment system based on fNIRS, characterized by: include: fNIRS signal acquisition module: configured to collect blood oxygen signals from students' prefrontal cortex and parietal brain regions in real time through a multi-channel near-infrared light source and detector; Data processing and analysis module: pre-processes blood oxygen signals, extracts features, and classifies cognitive states, outputting cognitive load levels and attention states; Adaptive learning engine: Dynamically adjusts learning content difficulty and pace based on cognitive state. This includes adjusting task difficulty based on cognitive load level and controlling learning pace based on attention state. User interaction module: provides a visual monitoring interface for teachers and real-time feedback for students.

2. The system according to claim 1, wherein: The fNIRS signal acquisition module includes: Dual-wavelength near-infrared light source, sampling frequency ≥10Hz; The flexible headband integrates an optode array with an optode spacing of 3 cm, covering the prefrontal and parietal cortex; Motion artifact suppression unit with built-in accelerometer.

3. The system according to claim 1, wherein: The pre-processing of the blood oxygen signal in the data processing and analysis module includes: Remove high-frequency noise and low-frequency drift through a band-pass filter; The wavelet transform algorithm combined with accelerometer data eliminates motion artifacts; Changes in HbO and HbR concentrations were calculated based on the modified Beer-Lambert law.

4. The system according to claim 1, wherein: The feature extraction includes: Extract the time domain features and frequency domain power spectral density of the prefrontal HbO signal; Attention concentration was assessed by changes in the HbO / HbR ratio in the parietal area.

5. The system according to claim 1, wherein: The adaptive learning engine adopts a model that integrates support vector machines and long short-term memory networks. The input parameters include the dynamic curve of blood oxygen signal, task type and individual historical data, and the output is cognitive load level and attention state.

6. The system according to claim 5, characterized in that: The dynamic adjustment strategy includes: When high cognitive load is detected, reduce the difficulty of the learning content or insert a knowledge point review; When attention is distracted, shorten the duration of a single study session and trigger interactive sessions; Dynamically adjust intervention thresholds based on individual resting-state baseline data.

7. The system according to claim 1, wherein: The user interaction module includes: The teacher's terminal displays the whole brain blood oxygen thermogram and the list of students who need intervention in real time; The student side embeds an attention value progress bar and a neural feedback reward mechanism.

8. A learning method based on the system according to any one of claims 1 to 7, characterized in that: The following steps are involved: Initial baseline calibration: Establishing an individual signal-cognitive state mapping model through resting-state measurements and standardized tasks; Real-time monitoring and dynamic adjustment: triggering learning content switching, rhythm control or interactive intervention based on changes in blood oxygen signals; Long-term learning path optimization: Generate neural data reports and recommend personalized learning plans.

9. The method according to claim 8, characterized in that The dynamic adjustment includes: If the prefrontal HbO concentration exceeded 150% of the baseline value and the variance increased, it was determined to be cognitive overload and the difficulty of the questions was reduced; If the parietal HbO / HbR ratio was continuously below the threshold of 80%, the attention activation task was initiated.

10. The system according to claim 1, wherein: The system can be integrated into smart classrooms or online education platforms, supporting simultaneous monitoring of multiple students and adaptation to cross-disciplinary tasks.

Images