NON-INVASIVE AI/ML-ASSISTED EEG SYSTEM FOR BINAURAL BEATS THERAPY FOR MENTAL HEALTH

Abstract

ABSTRACT NON-INVASIVE AI/ML-ASSISTED EEG SYSTEM FOR BINAURAL BEATS THERAPY FOR MENTAL HEALTH The present invention relates to a system for non-pharmacological cognitive enhancement by the use of audio stimuli called binaural beats. More particularly, the invention pertains to the use of binaural beats customised with “C Major scale” for induction of theta brainwaves for hemispheric synchronisation of the anxious human brain. The method includes combining auditory stimulation with advanced EEG data analysis techniques such as AI/ML models for prediction of mental states. A computational technique called Independent Component analysis (ICA) is used which enables enhanced feature extraction, improved data visualization, and noise reduction during audio signal processing. This EEG- Binaural beats-based neurofeedback therapy system helps to manage and regulate the brain functions and alleviate symptoms of various neurological disorders. The invention thus offers a promising tool for enhancing cognitive functions and prospective feasibility in curing anxiety and attention deficit hyperactivity disorder (ADHD) in a non-invasive manner.

Specification

Description:CROSS-REFERENCE TO RELATED APPLICATION AND PRIORITY
The present application claims no priority from any of the patent application(s).

FIELD OF THE INVENTION
The present invention relates to a system in the field of non-pharmacological cognitive enhancement and mental health interventions. Specifically, it pertains to the use of binaural beats customised with "C Major scale" for induction of theta brainwaves. Further, the EEG results were analysed with artificial intelligence (AI) and machine learning (ML) techniques for noise reduction and focused interpretation and prediction as a possible treatment of conditions such as anxiety and attention deficit hyperactivity disorder (ADHD). Thus, providing an insight regarding the feasibility of binaural beast to cure such conditions based on the induction of theta waves that calms the mind. This invention is particularly relevant to non-invasive method of therapeutic intervention and cognitive assessment. More particularly, for hemispheric synchronisation of the state of being of the anxious human brain by use of an audio stimuli inducing relaxing brain waves.

BACKGROUND OF THE INVENTION
Binaural beats are auditory stimuli created by presenting two slightly different frequencies to each ear, which the brain perceives as a third frequency (single, pulsating beat). The third frequency perceived by brain is the difference of the two-frequencies provided to the ears. As physiologically human can't hear sound frequencies lower than 20 Hz. This technique is hypothesized to induce brainwave synchronization, particularly in the theta frequency band, which have been proven to facilitate relaxation, enhance cognitive functions, and improve mental health conditions. Prior research has shown that binaural beats can affect brainwave patterns and potentially alter cognitive and emotional states. However, existing methods lack comprehensive data analysis to fully understand their impact on various demographics, especially concerning anxiety and ADHD. Various brainwave frequencies are associated to various states of human brain. A brain wave frequency band ranging from 4 to 7.5 hertz is known as "theta" that is associated with peaceful and relaxed/ meditative state or sleep state of brain. Researchers found that inducing brain wave frequencies via external stimuli are associated with normal or desired behavior. That has been induces by relaxation, either using music or other therapies too. In an anxious state the beta waves are predominate, binaural beats help in the equal distribution of these waves reducing the anxiety and producing a meditative state by inducing theta waves. Theta brainwaves have a frequency of 4 to 7.5 Hz, is classified as "slow" activity. It is not normal in adult brain but found in normal children and in sleep. Electroencephalograph is used in non-invasively observing the human brain activity. It records the electrical signals from the brain via electrodes on the subject's scalp. The electrical activity of brains is considered as a diagnostic marker for various brain functions and external neurostimulation stimuli used as therapeutic intervention. Much past research has shown the impact of binaural beats for inducing relaxed brain states. Heartbeat and brainwave tend to synchronize with the audio stimuli provided, helping reduce hypertensions and inducing sleep.
Binaural beats are somewhat like frequency oscillations produced by a heterodyne effect (however happen naturally in the brain itself). Thus, producing a frequency entrainment effect, when these beats are within the range of brainwave frequencies giving therapeutic intervention. Brainwave frequency can be induced in the brain by external signals of different frequencies bands to the two ears of a person (each ear receiving two different frequencies) inducing a sensation of "third frequency" and the human brain deciphering the frequency differences between them. Thus, Binaural beats are auditory responses of brain produced in the superior olivary nucleus of both cerebral hemispheres, as it lies out of human's normal hearing range of 20-20,000 Hz. It helps to entrain the brainwaves via frequency-following response (FFR) hence used as consciousness management technique. Table 1 depicts types of brainwaves.

Table 1: Types of Brainwaves
Types Frequency (Hz) Function
Alpha 8-13 Relax &Awake
Beta 13-30 Logical thinking, conscious thought and concentration
Gamma More than 30 Peak concentration, problem solving
Delta Less than 4 During Sleep
Theta 4-7.5 deep relaxation and light sleep

Recent advances in AI and ML offer the potential to analyse complex EEG data more effectively, providing deeper insights into how binaural beats influence brain activity. This study aims to explore the therapeutic potential of binaural beats by leveraging AI-assisted EEG analysis to enhance hemispheric synchronization and improve mental health outcomes. There are some inventions made in past that has utilised binaural beats or some focuses on other tools for hemispheric synchronisation and mental health interventions. Following are a few patents that had emphasised on similar but not same non-invasive interventions.

Patent No. AU2018100093A4, entitled Variable Rhythmic Bilateral Stimulation and Brainwave Synchronization (VRBSBS), covers a system and method for synchronizing brainwave activity through variable rhythmic bilateral stimulation. Claiming the novel method for variable rhythmic stimulation, Synchronization of brainwaves and Bilateral stimulation techniques. It has a similar focus on brainwave synchronization but not specifically using binaural beats or targeting anxiety/ADHD.
Patent No. US2006116597A1, entitled Brain balancing by binaural beat describes a method for balancing brain activity using binaural beats. Claiming the use of binaural beats for brain balancing via specific frequency ranges for effective brain balancing, however its therapeutic aspects were not considered.
Patent No. KR102645578B1, entitled Cognitive Function Improving Service Artificial Intelligence Based Brainwave Entrainment and Light Therapy, focuses on improving cognitive function through AI-based brainwave entrainment and light therapy. AI-based brainwave entrainment is relevant, however the customisation of binaural beats and stimulation to produce theta state has no mention.
Patent No. RU2306852C1, entitled Method for Rehabilitation of Human Emotional-Effective Disorders, through specific therapeutic techniques and Emotional impact assessment. It addresses emotional disorders but not AI/EEG-specific or focusing on ADHD.
Patent No. US2011105938A1, entitled Binaural Beat Augmented Biofeedback System describes a biofeedback system enhanced by binaural beats to influence physiological responses. And a method for system calibration. Binaural beats are used, but the focus is on biofeedback rather than AI-assisted analysis.
Patent No. US2007282216A1, entitled Altering Brain Activity Through Binaural Beats covers methods to alter brain activity using Specific binaural beats. Frequencies for brain function modulation. Binaural beats were used to alter brain activity but lacks AI integration and focus on anxiety/ADHD.
Patent No. IN202411058220A, entitled Neuroresonance Relaxation System relates to a system designed and implementation to induce relaxation through neuroresonance techniques. The prime impetus of this patent is relaxation rather than therapeutic use for anxiety/ADHD with AI analysis.
Patent No. IN201641029383A, entitled Audio Aided Alpha State focuses on achieving an alpha brainwave state using audio aids. And methods for maintaining alpha brainwave activity for mental relaxation. It focuses on alpha state rather than theta waves or AI/EEG analysis for anxiety/ADHD.
Patent No. WO2018208167A1, entitled Brain Wave Regulator- BWR (BWI) involves a device for regulating brain waves, potentially using various stimulation methods. For brain wave modulation. Brain wave regulation is related, but not specific to binaural beats or AI/EEG.
Patent No. KR20100087902A, entitled Brain Waves Inducing Apparatus Using High Frequency Inside of Audio Frequency range, used for High-frequency brainwave induction. It mainly focuses on high frequencies and apparatus rather than binaural beats with AI/EEG.
Patent No. WO2013187544A1, entitled Brain Wave Inducing System Using Smartphone Application, and Method for Generating Brain Wave Signals. However, it uses smartphone applications for brain wave generation, but not AI or EEG analysis for anxiety.
Patent No. TW201707650A, entitled Method for Mixing Binaural Beats with Music to Modulate Various Neuronal Networks in the Brain Capable of Helping Sleep, Regulate Emotion, and Improving Concentration. This patent describes the binaural beats and modulation of brain function that are similar but lacks AI/EEG focus.
Patent No. US2013177883A1, entitled Systems and Methods for Directing Brain Activity describes systems designed to direct and control brain activity using specific methods and technologies. Further it describes the brain activity control methods, but not specific to binaural beats or AI/EEG analysis.
Patent No. RU2678546C1, entitled Method of Treatment Anxiety-Depressive Syndrome using specified therapeutic approaches and effectiveness assessment. It emphasises on the treatment for anxiety but not using AI/EEG or binaural beats specifically.
Patent No. CN114267364A, entitled Method and Device for Determining Audio of Self-Adaptive Binaural rhythms, for audio applications for Applications in audio and therapy. It describes binaural rhythm adaptation but not specifically AI-assisted or focused on ADHD.
Patent No. KR20050074947A, entitled Non-Audio Frequency Binaural-Beat for specific effects. And its primary focuses are on non-audio frequencies rather than theta waves or AI/EEG.
Patent No. KR20230033270A, entitled Composite Brain Function Improving Apparatus Using Binaural Beats for Prevention and Treatment Dementia and Brain Function Improvement Activity. However, it uses binaural beats for brain function but not AI-assisted EEG analysis for ADHD/anxiety.

OBJECTIVES OF THE INVENTION
The main objective of the invention is to Design and Customize Binaural Beats with theta (4-7.5 Hz) frequencies with refined audio tracks to optimize the effectiveness of these binaural beats for inducing hemispheric synchronization. Other objectives of the invention are:
1. To design and customise the binaural beats audio pattern
a. To determine the optimal frequency band to target specific cognitive and emotional states
b. To develop the audio tracks incorporating the optimized frequencies
c. To create the binaural beats pattern
2. To configure the EEG setup & optimise the EEG paradigm for the study
a. To position the headgear with suitable number of EEG electrodes
b. To identify cohorts suitable for this study considering various gender, age groups and mental conditions
c. To perform the studies with designed binaural beats audio pattern with exposure and silent phases
3. To evaluate the acquired EEG data via AI/ML
a. To Apply AI/ML to the acquired EEG data
b. To perform the predictive analysis to binaural beats based on physiological and cognitive data.
c. To demonstrate the feasibility of binaural beats

SUMMARY OF THE INVENTION:
The present invention involves a non-invasive theranostic tool that integrates binaural beats with AI-assisted EEG interpretation to address anxiety and ADHD. The customised binaural beats were used to target specific brainwave frequencies and uses advanced AI and ML algorithms to analyse EEG data. A system and method are disclosed to achieve the cerebral synchronization via inducing theta state in brain. The binaural beat was generated by applying two different frequencies to two ears. "C Major scale" is a commonly used note in music for imparting happiness. Hence, the binaural beats were customised in "C Major scale's frequency of 261.5 Hz for base frequency to one ear and 267 Hz for another ear. Further, the physiological capacity of brain allows these two frequencies to be perceived as the difference of the provided stimuli (5.5 Hz lying in theta brain wave frequencies). Hence, the customised tracks have the benefits of both C major scale and theta frequencies as well. Thus, provide cognitive enhancement via a non-pharmacological intervention, i.e., Theta binaural beats. A computer is used in the present invention for controlling the equipment and to provide feedback between the brain wave measurement and the record generated signals. The goals are to enhance hemispheric synchronization to reduce ADHD & Anxiety, assess the impact of binaural beats on mental states, and develop personalized therapeutic interventions. The system consists of EEG apparatus to record the brain wave frequency spectrum through electrodes. The invention offers a novel approach to cognitive enhancement and mental health treatment by combining auditory stimuli with sophisticated data analysis techniques. The AI/ML techniques aid in noise reduction and focused interpretation. Noise reduction in EEG data improved the interpretation of the brainwave state, clearly indicating the dissipation of particular brainwaves, hence physiological state of brain.

BRIEF DESCRIPTION OF THE DRAWINGS:
Figure 1 EEG apparatus with computer system (100) disseminating Binaural beats via
earphones (300)

Figure 2 Representative Binaural Beats Pattern (102) customized using Logic Pro
workstation (101; Tone generator software/ Digital Audio Workstation)

Figure 3 Study Paradigm (103) designed for exposure of binaural beats with sound phase and
silent phases

Figure 4: Illustrates the Proposed AI/ML based method (402) for analysis and classification
of impact of binaural beats with EEG signals (401)

Figure 5 has the Independent Component analysis using AI/ML model (501)

Figure 6 illustrates the Random forest predictions (502) of impact of Binaural beats

Figure 7 Impact of Binaural beats in inducing theta brainwave patterns via Confusion matrix
and classification report in Long Short-Term Memory (LSTM) (503)

Figure 8 illustrates the AI/ML analysis (504) of Binaural beats showing the highest impact in
second track

DETAILED DESCRIPTION OF THE INVENTION:
Cerebral or hemispheric synchronisation techniques utilises induction of various rhythms in the brain that has been proven to elicit a calming effect, via theta brain waves. The left cerebrum is analytical & logical, while the right one is artistic. Normal brain has a spontaneous balance between left and right hemispheres. Thus, the current invention discloses a system and method to achieve the cerebral synchronisation via exposure to external binaural patterns measured using EEG electrodes to treat an anxious brain using itself as an analyser of the binaural signals provided.
As shown in FIGURE 1 the present invention has a system, comprising a computer 100 for controlling the equipment, an EEG system 200 to measure the brain wave spectrum, and a binaural beat 102 and an earphone 300. The EEG system comprises of plurality of electrodes 201 with optimised positioning, attached to the scalp of the user. The number of electrodes 32 is even, half for each half of the brain. The electrodes 201 can communicate with the computer 100 through 200. The Binaural beat 102 were generated using Digital Audio workstation software 101. The binaural beat comprises produces a first audio signal at a first frequency on a first channel 301 and a second audio signal at a second frequency on a second channel 302. The frequency difference between the first and second audio signals creates the binaural beat corresponding to a chosen imbalance brain wave frequency. First channel sends the first audio signal to one ear of the user through an earphone 300, and second channel send the second audio signal to the other ear of the user through an earphone 300. The binaural beats comprise a step of alternate stimuli and silent phases repeating thrice in a paradigm. The binaural beats comprise of fixed intervals of rest frequency between two stimuli frequencies or binaural beats. Further, a data analysis unit 400 having AI/ML tools 402 for EEG analysis 401 where the 503 provides the confusion matrix of the data and LSTM analysis as further explained below. Alongside, 501 Consists of ICA components, 502 provides Random-forest predictions and 504 for the AI analysis of the EEG signals, as described below in figure 5 to 8. There are optional devices accessory of the computer including keyboard, mouse and display for conventional input and output devices, and volume, waveform, and balance controls for adjusting to the individual user and the purpose of the use. EEG data is collected using a 32-channel system to monitor brain activity during binaural beat exposure. The setup includes channels focused on temporal, parietal, and occipital lobes. Filters are applied to isolate theta band frequencies (3-9 Hz). EEG signals are recorded during binaural beat exposure and silent phases.

Figure 2 represents Representative Binaural Beats Pattern (102) customized using Logic Pro workstation (101; Tone generator software/ Digital Audio Workstation). The Designing and Customization of Binaural beat pattern in Tone generator of Digital Audio Workstation (101). The invention involves creating binaural beats (102) with specific frequencies in the C Major scale with base frequency at 261 Hz and 267 Hz for another ear. Brain perceives the difference of different auditory signal as third frequency called theta frequency. Theta varies from that band from 4 Hz to 7.5 Hz, targeting theta waves. Custom audio tracks are designed with varying frequency patterns to study their effects on hemispheric synchronization.

Figure 3 shows the Study Paradigm (103) designed for exposure of binaural beats with sound phase and silent phases. Each binaural beat track is structured into one-minute to one-minute fifty-one-second epochs epochs, preceded and succeeded by one minute of silence phase to control for external auditory influences. Sound phases comprise of the signals 261 Hz frequency (301) to right ear and 267 Hz frequency (302) to left ear.

Figure 4 illustrates the Proposed AI/ML based method (402) for analysis and classification of impact of binaural beats with EEG signals (401); this is the final data analysis unit (400) where 4(a) describes the steps of electrodes selection for input channel to sort the data for pre-processing and focus on only most relevant channels that have maximum impact of binaural beats such as EOGL, Pz, P3, P7, O1, Oz, O2, P4, P8, EOGR. Where, EOGL and EOGR Stands for Electro Oculo Gram (Left and Right), P Stands for the electrode at parietal lobes of brain and O at occipital lobes locations. After this, the raw filter was adjusted to 3 to 9 hertz to prevent the data loss, and 10 ICA components were taken, and data was pre-processed for further analysis. Figure 4(b) illustrates the input as pre-processed EEG data and various tools such as LSTM deep learning model and its features with Random Forest machine learning module to take the final output.

Figure 5 illustrates the Independent Component analysis using AI/ML model (501).
Independent Component analysis (ICA) using AI/ML model; ICA is a computational technique used for signal processing and EEG data analysis converting multivariate signal into independent components showing exclusive results and the area of interest via enhanced feature extraction, improved data visualization, and also assist in tasks like noise reduction and blind source separation, thus used in audio/signal processing, and image analysis. Here, our results show how the various brain parts are involved in binaural beats annihilation.

Figure 6 illustrates the Random Forest predictions (502) of impact of Binaural beats. Random Forest predictions of impact of Binaural beats. It is an ensemble machine learning technique used to predict the impact of binaural beats on various outcomes such as mood, cognition, and relaxation. It analyses features like frequency, duration, and individual differences to identify patterns and correlations. This approach enhances understanding of how binaural beats influence brain activity and psychological states, providing insights for therapeutic interventions.

Figure 7 illustrates the Impact of Binaural beats in inducing theta brainwave patterns via Confusion matrix and classification report in Long Short-Term Memory (LSTM) (503). Impact of Binaural beats in inducing theta brainwave patterns via Confusion matrix and classification report in Long Short-Term Memory (LSTM). Machine learning model i.e. Long Short-Term Memory (LSTM) was employed to classify and analyses EEG data.

Further, Figure 8 described the AI/ML analysis (504) of Binaural beats showing the highest impact in second track. Where X axis represents the sound phases I, 2 and 3 as Audio 1,2 and 3, respectively. While Y-axis is for the impact of binaural beats with respect to the sound phases. It represents the Audio analysis showing the highest impact in second track of binaural beats. The effects of binaural beats are evaluated by comparing EEG data from different phases of binaural beat exposure, showing that beats had maximum impact in second exposure of sound phase. Time-domain and frequency-domain analyses, along with statistical methods, are used to assess the impact of binaural beats on brain activity and cognitive states. AI/ML algorithms correlate EEG markers with changes in mental states, such as relaxation and focus. Physiological markers, including heart rate and skin conductance, are also analysed to validate the findings and assess the overall efficacy of the binaural beats.

Conclusively, the AI/ML analysis of the EEG pattern shows the culminated results of electrodes placed for ocular (eye) movement, heartbeat, skin conductance and other electrodes placed on scalp. Figure 8 is the result analysis that shows how the theta waves increases when the binaural beats stimulus is provided to the subject and also it decreases the heartbeat, eye movement and the skin conductance too. All these signals pertain to calm brain activity, thus leading to hemispheric synchronisation and reduced anxiety. The table 2 below shows the inferences of the increase of theta brainwaves before and after the stimulus of binaural beats.

Table 2: Impact of binaural beats before and after theta brainwave induction

Sr. No. Electrode Positioned/ Site Theta brainwave intensity (Before Binaural beats stimulus) Theta brainwave intensity (After Binaural beats stimulus) Inference
1 EOGL/R
Eye muscles relax with binaural beats stimulus
2 ECG
Heart rate decreases
3 Skin Conductance/ EKG
Sweat production releases

The results, further indicates the feasibility of using binaural beast for the treatment of conditions like anxiety and ADHD as it helps in induction of theta waves. Hence calms the brain down. , Claims:WE CLAIM:

1. A system for cerebral synchronization via customised binaural beat track, the system comprising:
a) An electro-encephalography set-up (200) essentially consisting of plurality of electrodes (201);
b) A digital audio workstation (101) for generating the binaural beats, wherein the binaural beats are sent as signals to the left and right ear through earphones (300);
Characterized in that
i) the customised binaural beats have specific frequencies that band from 4 Hz to 7.5 Hz that signifies theta brain wave;
and
ii) each binaural beat track is programmed into one-minute to one-minute fifty-one-
second epochs, preceded and succeeded by one minute of silence phase to control for
external auditory influences.

2. The system for cerebral synchronization as claimed in claim 1, wherein the said electro-encephalography setup (200) includes channels focused on temporal, parietal, and occipital lobes of the brain, ECG, EKG.

3. The system for cerebral synchronization as claimed in claim 1, wherein the electro-encephalography data is collected using a 32-channel system to monitor neurofeedback / brain activity during binaural beat exposure.

4. The system for cerebral synchronization as claimed in claim 1, wherein the electro-encephalography set-up (200) captures brain activity during binaural beat exposure and silent phases

5. A method for cerebral synchronization comprising the steps of:
a) applying a constant frequency of 261Hz (301) audio signal to one ear and another
audio signal having a frequency of 267Hz (302) to the other ear simultaneously;
b) applying the binaural beat to the left and right ear of the user;
c) recording the signals by electro-encephalography set-up (200) to generate electro-
encephalography data;
d) analysis (400) of the electro-encephalography data (401) using AI and ML models
(402) to assess brainwave synchronization;
wherein, the resulting beats acts to entrain brain waves into unique waveform
characteristic of identified states of consciousness.

6. The method for cerebral synchronization as claimed in claim 5, wherein the first audio signal (301) and second audio signal (302) differ in magnitude, phase or both.

7. The method for cerebral synchronization as claimed in claim 5, wherein the AI/ML models for predicting the impact of binaural beats on mental states, including but not limited to anxiety reduction; improved attention span; and enhanced cognitive function.

8. The method for cerebral synchronization as claimed in claim 5, wherein the binaural beats are customised to induce theta brain waves.

9. The method for cerebral synchronization as claimed in claim 5, wherein the binaural beats comprise a step of alternate stimuli and silent phases repeating thrice in a paradigm.

10. The method for cerebral synchronization as claimed in claim 5, wherein the AI/ML models utilise Independent Component analysis (ICA) (501), Random forest (502) and Confusion matrix (503) for computations.

Documents