Children hyperactivity degree evaluation system based on heartbeat evoked potential

Abstract

The invention discloses a child hyperactivity degree evaluation system based on heartbeat evoked potential, comprising: a signal acquisition module for synchronously acquiring an electroencephalogram signal and an electrocardiosignal of a testee in two resting states of eye-open rest and eye-closed rest; the signal preprocessing module is used for carrying out filtering and artifact removal on the collected electroencephalogram signals and carrying out R-wave detection and marking on the electrocardiosignals; the feature extraction module is used for extracting heartbeat evoked potential features based on the preprocessed signals; wherein the step of extracting the heartbeat evoked potential features comprises the following substeps: identifying a heartbeat evoked potential significant cluster with significant difference between the hyperactivity children and the typical development children by using a cluster-based arrangement test method, and extracting the average amplitude in the cluster; and the evaluation application module is used for constructing a multiple linear regression model based on the extracted heartbeat evoked potential characteristics so as to predict the severity of the hyperactivity symptom. According to the invention, accurate evaluation of the hyperactivity of children can be realized.

Description

Technical Field

[0001] This invention relates to the fields of medical diagnostics and neuroengineering, and in particular to a system for assessing the severity of ADHD in children based on cardiac evoked potentials. Background Technology

[0002] Attention-Deficit / Hyperactivity Disorder (ADHD) is one of the most common neurodevelopmental disorders in childhood. Its clinical diagnosis currently primarily follows the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria, relying on interviews with the child, parents, and teachers, as well as assessments using behavioral scales (such as SNAP-IV). This method is highly dependent on the attending physician's subjective experience and is time-consuming. Furthermore, parents or teachers may exhibit recall bias or subjective tendencies when completing the scales, leading to insufficient objectivity in the diagnostic process.

[0003] In search of objective biomarkers, researchers have extensively explored electrophysiological indicators. Electroencephalography (EEG) studies have shown that children with ADHD may exhibit increased theta wave power and decreased beta wave power; the theta / beta ratio (TBR) was once approved by the US FDA as an auxiliary diagnostic reference for ADHD. However, several recent meta-analyses have strongly questioned its specificity and reliability, arguing that TBR may be affected by factors such as individual alpha peak frequency, resulting in inconsistent discrimination effects and poor reproducibility across different studies (McVoy et al., 2019; Boxum et al., 2024), thus limiting its direct clinical diagnostic value.

[0004] Regarding the autonomic nervous system (ANS), studies have shown that children with ADHD may have abnormal heart rate variability (HRV), such as an elevated low-frequency to high-frequency power ratio (LF / HF), reflecting an imbalance in sympathetic and parasympathetic regulation (Griffithset et al., 2017). However, HRV indicators also exhibit significant individual heterogeneity and state dependence, with insufficient and stable inter-group differences at rest, and are easily affected by factors such as respiration and body movement, making them difficult to use as independent diagnostic criteria.

[0005] Heartbeat-evoked potentials (HEPs) are cortical potentials that reflect the brain's perception and processing of cardiac signals, serving as an important window into the brain-heart interaction mechanism. In recent years, HEPs have been shown to exhibit abnormalities in mental disorders such as depression and anxiety, revealing the importance of the brain-heart interaction mechanism in mental illness (Park & ​​Blanke, 2019). However, in the field of ADHD, HEP-based research is extremely scarce, and effective methods have not yet been developed. Rapp et al. (2023) found a positive correlation between ADHD symptoms and HEP amplitude in a community-based adolescent sample, but this study did not target clinically diagnosed ADHD children, and its paradigm was an emotional task rather than a resting state, which could not rule out the interference of task performance ability on the results.

[0006] Furthermore, a search revealed that existing patented technologies (such as CN118285795A and CN115462794A) mainly utilize traditional eye movement or EEG spectral features for ADHD assessment, and no schemes using resting-state HEP as a core feature for ADHD auxiliary diagnosis have been disclosed.

[0007] Therefore, the existing technology lacks a complete technical solution and system that can use resting-state brain-heart interaction signals (HEP) to objectively and efficiently assess clinically diagnosed ADHD children. Summary of the Invention

[0008] This invention provides a system for assessing the severity of ADHD symptoms in children based on heartbeat evoked potentials (HEPs). By analyzing the quantitative relationship between the neurophysiological characteristics of heartbeat evoked potentials (HEPs) and clinical symptoms, it is possible to achieve accurate assessment of ADHD symptoms.

[0009] A system for assessing the severity of ADHD in children based on cardiac evoked potentials includes: The signal acquisition module is used to simultaneously acquire the electroencephalogram (EEG) and electrocardiogram (ECG) signals of the subjects in two resting states: with eyes open and with eyes closed. The signal preprocessing module is used to filter and remove artifacts from the acquired EEG signals, and to detect and label the R-wave of the ECG signals. The feature extraction module extracts heartbeat evoked potential features based on the signal processed by the signal preprocessing module. The extraction of heartbeat evoked potential features includes: using a cluster-based permutation test method to identify significant clusters of heartbeat evoked potentials that show significant differences between children with ADHD and children with typical development, and extracting the average amplitude within the cluster. The evaluation application module constructs a multiple linear regression model based on the extracted heartbeat evoked potential features to predict the severity of ADHD symptoms.

[0010] In the signal acquisition module, multi-lead EEG equipment and high-precision ECG electrodes were used for synchronous acquisition. The EEG electrodes were placed according to the international 10-20 system standard, and the ECG electrodes were placed under the left clavicle and the right abdomen. The sampling rate was set to 512Hz. The subjects were required to remain relaxed and avoid large movements during the acquisition process. The environment was also required to be quiet and the lighting was soft.

[0011] In the signal preprocessing module, the acquired EEG signals are filtered and artifacts are removed, specifically as follows: The EEG signal was bandpass filtered from 0.5 to 35 Hz, and a large amplitude artifact was removed using an artifact space reconstruction algorithm. Independent component analysis was used to remove EEG and ECG artifacts. The BurstCriterion parameter of the artifact space reconstruction algorithm was set to 20. The independent component analysis used the extended Infomax algorithm combined with the ICLabel plugin to automatically identify and remove artifact components with a probability threshold greater than 80%.

[0012] In the signal preprocessing module, R-wave detection and labeling are performed on the electrocardiogram signal, specifically as follows: The electrocardiogram (ECG) signal was filtered, and the Pan-Tompkins algorithm was used to automatically detect the R-wave peak value. Visual inspection and manual correction were also used to ensure the accuracy of the marker points. The ECG signal was obtained by calculating the difference between the EXG3 and EXG4 channels.

[0013] The feature extraction module includes a clustering-based permutation test method, specifically: The EEG signal was segmented with a time window of 200ms before and 800ms after the R wave, and corrected with a baseline of -200ms to -100ms. The t-value of the difference in the amplitude of the heartbeat evoked potential between the two groups of subjects at each time point and on each electrode was calculated. Then, adjacent significant points were clustered together, and the statistical significance of these clusters was evaluated by permutation test. Only the clusters that passed the significance test were retained as the feature extraction region.

[0014] In the assessment application module, the multiple linear regression model uses the heartbeat evoked potential characteristics extracted from both open and closed eyes conditions as independent variables and the symptom scores assessed by the clinical standardized scale as dependent variables; the clinical standardized scale is the SNAP-IV scale.

[0015] Furthermore, when constructing the multiple linear regression model, a stepwise regression method is used for feature selection to optimize the model and prevent overfitting.

[0016] Furthermore, the symptom scores assessed by the clinically standardized scales include inattention scores, hyperactivity scores, and impulsivity scores.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention innovatively employs a data-driven method based on clustering permutation tests, which can unbiasedly identify the neurophysiological markers that best distinguish the severity of ADHD symptoms. This avoids the subjectivity of traditional fixed-time-window analysis, significantly improving the accuracy and reliability of the assessment. Compared to existing classification diagnostic methods, this invention provides a continuous, dimensional assessment of symptom severity, more nuancedly reflecting the mild, moderate, or severe differences in children's clinical symptoms. This provides direct and objective data support for clinicians to develop individualized intervention plans and quantify efficacy assessments.

[0018] 2. This invention is based on the resting-state paradigm, eliminating the need for children to complete complex cognitive tasks, significantly reducing testing difficulty, improving children's cooperation, and enhancing data quality. It is particularly suitable for children with ADHD who have poor attention and cooperation abilities. The entire assessment process takes only about 10 minutes, demonstrating high clinical feasibility and efficiency, and can serve as an effective supplement to routine clinical assessments.

[0019] 3. By combining the innovative biomarker of brain-heart interaction mechanism with multivariate statistical methods, this invention not only provides a new method for symptom assessment, but also offers a new perspective for understanding the neurophysiological mechanism of ADHD, which has important theoretical value and clinical significance. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the process of a childhood ADHD severity assessment system based on heartbeat evoked potentials, as described in this invention.

[0022] Figure 2 Comparison of heart evoked potential (HEP) waveforms between a typical developing child and a child with ADHD under eye closure conditions.

[0023] Figure 3 Comparison of heart evoked potential (HEP) waveforms between a typical developing child and a child with ADHD under open-eye conditions.

[0024] Figure 4 Scatter plot showing the correlation between the predicted values ​​of the HEP feature from a multiple linear regression model and the clinical attention score.

[0025] Figure 5Scatter plot showing the correlation between the predicted values ​​of HEP features from a multiple linear regression model and the clinical hyperactivity score. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] It should be noted that, unless otherwise specified, the features in the following embodiments and implementation methods can be combined with each other.

[0028] like Figure 1 As shown, a system for assessing the severity of ADHD symptoms in children based on heartbeat evoked potentials mainly includes four core modules: signal acquisition, signal preprocessing, feature extraction, and assessment application. The system first simultaneously acquires raw EEG and ECG signals from subjects in both open-eye and closed-eye resting states using a multi-lead EEG device and ECG electrodes. The acquired signals are then preprocessed, including filtering and artifact removal of the EEG signals, and R-wave detection and labeling of the ECG signals. Next, heartbeat evoked potentials are calculated using the R-wave as the event lock point, and a cluster-based permutation test method is used to identify temporal-spatial clusters with significant inter-group differences. The average amplitude of these clusters is extracted as the core feature. Finally, the extracted features are input into a pre-trained multiple linear regression model, which outputs a predicted value for the severity of ADHD symptoms and generates an assessment report.

[0029] (1) Signal acquisition module: In the specific implementation process, more than 30 clinically diagnosed children with ADHD and over 30 age-matched children with typical developmental characteristics were selected as subjects. All subjects had an IQ above 70 and no other neurodevelopmental disorders. Signal acquisition was performed using the BioSemi ActiveTwo system, equipped with 32-lead EEG electrodes and 4 external electrodes, placed according to the international 10-20 system standard. The ECG acquisition electrodes were placed in the left subclavian and right abdominal regions, respectively, with a signal sampling rate set to 512 Hz. During data acquisition, subjects were required to sit comfortably in a soundproof, softly lit room. Data acquisition was first performed for 3 minutes in an open-eye resting state, during which they focused on the crosshair fixation point in the center of the screen, followed by 3 minutes of closed-eye resting state data acquisition.

[0030] (2) Signal preprocessing module: Signal preprocessing was performed in the MATLAB environment using the EEGLAB and FieldTrip toolboxes. EEG signal preprocessing included bandpass filtering from 0.5 to 35 Hz, removal of large amplitude artifacts using an artifact space reconstruction algorithm (BurstCriterion=20), and automatic identification and removal of artifact components (probability threshold >80%) using extended Infomax independent component analysis and the ICLabel plugin. ECG signal preprocessing obtained the ECG signal by calculating the difference between the EXG3 and EXG4 channels, automatically detecting the R-wave peak using the Pan-Tompkins algorithm, and supplementing this with visual inspection and manual correction to ensure accurate marker placement.

[0031] (3) Feature extraction module: In the feature extraction stage, the preprocessed EEG signal was segmented within a time window of 200 ms before to 800 ms after each R-wave marker, and correction was performed using a baseline of -200 ms to -100 ms. A cluster-based permutation test was used to identify significantly different HEP components at the group level. A two-tailed t-test was performed, with a clustering threshold of p < 0.05. 10,000 permutation tests were conducted, and the final significant clusters were required to satisfy p < 0.05.

[0032] like Figure 2 and Figure 3 As shown in the example, in this implementation case, under both eye-closed and eye-open conditions, a significant cluster of negative components was identified in the central parietal lobe region within a time window approximately 200 ms after the R wave. The average amplitude of each subject within this spatiotemporal cluster was extracted as the core HEP feature. The HEP amplitude of children with ADHD during this period was significantly lower than that of children with typical development, indicating an abnormality in the early attentional allocation stage of heartbeat perception in children with ADHD.

[0033] (4) Evaluation of application modules: In the model building and symptom prediction phase, HEP features extracted under both EO and EC conditions were used as independent variables, and the hyperactivity / impulsivity subscale score of the SNAP-IV scale was used as the dependent variable to construct a multiple linear regression model. Stepwise regression was employed for feature selection to optimize the model and prevent overfitting.

[0034] Figure 4 and Figure 5 The predictive performance of the multiple linear regression model is demonstrated, among which... Figure 5 There was a significant positive correlation between the predicted values ​​of the ADHD assessment scores and the actual clinical scores (R0). 2 = 0.42, p < 0.001). This indicates that the assessment model based on HEP ​​characteristics can effectively predict the severity of ADHD symptoms, providing an objective and quantitative reference for clinical diagnosis.

[0035] This embodiment demonstrates the effectiveness and practicality of the method described in this invention through specific experimental data and charts. The system adopts a resting-state paradigm, eliminating the need for children to complete complex tasks, thus greatly improving the feasibility of clinical application. Through a cluster-based permutation test method, it can identify the most discriminative neurophysiological markers in a data-driven manner, improving the accuracy and reliability of the assessment. Finally, by establishing a quantitative relationship between HEP ​​characteristics and clinical symptoms, it achieves an objective assessment of the severity of ADHD symptoms.

[0036] The embodiments described above provide a detailed explanation of the technical solutions and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, additions, and equivalent substitutions made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A system for assessing the severity of ADHD in children based on cardiac evoked potentials, characterized in that, include: The signal acquisition module is used to simultaneously acquire the electroencephalogram (EEG) and electrocardiogram (ECG) signals of the subjects in two resting states: with eyes open and with eyes closed. The signal preprocessing module is used to filter and remove artifacts from the acquired EEG signals, and to detect and label the R-wave of the ECG signals. The feature extraction module extracts heartbeat evoked potential features based on the signal processed by the signal preprocessing module. The extraction of heartbeat evoked potential features includes: using a cluster-based permutation test method to identify significant clusters of heartbeat evoked potentials that show significant differences between children with ADHD and children with typical development, and extracting the average amplitude within the cluster. The evaluation application module constructs a multiple linear regression model based on the extracted heartbeat evoked potential features to predict the severity of ADHD symptoms.

2. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 1, characterized in that, In the signal acquisition module, multi-lead EEG equipment and high-precision ECG electrodes were used for synchronous acquisition. The EEG electrodes were placed according to the international 10-20 system standard, and the ECG electrodes were placed under the left clavicle and the right abdomen. The sampling rate was set to 512Hz. The subjects were required to remain relaxed and avoid large movements during the acquisition process. The environment was also required to be quiet and the lighting was soft.

3. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 1, characterized in that, In the signal preprocessing module, the acquired EEG signals are filtered and artifacts are removed, specifically as follows: The EEG signal was bandpass filtered from 0.5 to 35 Hz, and a large amplitude artifact was removed using an artifact space reconstruction algorithm. Independent component analysis was used to remove EEG and ECG artifacts. The BurstCriterion parameter of the artifact space reconstruction algorithm was set to 20. The independent component analysis used the extended Infomax algorithm combined with the ICLabel plugin to automatically identify and remove artifact components with a probability threshold greater than 80%.

4. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 1, characterized in that, In the signal preprocessing module, R-wave detection and labeling are performed on the electrocardiogram signal, specifically as follows: The electrocardiogram (ECG) signal was filtered, and the Pan-Tompkins algorithm was used to automatically detect the R-wave peak value. Visual inspection and manual correction were also used to ensure the accuracy of the marker points. The ECG signal was obtained by calculating the difference between the EXG3 and EXG4 channels.

5. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 1, characterized in that, The feature extraction module includes a clustering-based permutation test method, specifically: The EEG signal was segmented with a time window of 200ms before and 800ms after the R wave, and corrected with a baseline of -200ms to -100ms. The t-value of the difference in the amplitude of the heartbeat evoked potential between the two groups of subjects at each time point and on each electrode was calculated. Then, adjacent significant points were clustered together, and the statistical significance of these clusters was evaluated by permutation test. Only the clusters that passed the significance test were retained as the feature extraction region.

6. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 1, characterized in that, In the assessment application module, the multiple linear regression model uses the heartbeat evoked potential characteristics extracted from both open and closed eyes conditions as independent variables and the symptom scores assessed by the clinical standardized scale as dependent variables; the clinical standardized scale is the SNAP-IV scale.

7. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 6, characterized in that, When constructing a multiple linear regression model, stepwise regression is used for feature selection to optimize the model and prevent overfitting.

8. The ADHD severity assessment system for children based on heartbeat evoked potentials according to claim 6, characterized in that, Symptom scores assessed by standardized clinical scales include inattention score, hyperactivity score, and impulsivity score.

Images