System and method for monitoring longitudinal shifts in neurobehavioral wellness due to therapeutic intervention

Abstract

The present disclosure provides a system (100) and method (300) that are utilized for assessing uni or multimodal treatment-based intervention in many neurological disorders, such as depression like mood disorders, etc. The system (100) is designed to forecast longitudinal changes in neuro-cognitive behavioral wellness over time, particularly following a few weeks of intervention, such as targeted neuromodulation. The system (100) operates by leveraging data from two initial points: day 0 and approximately 25% into the intervention course, or 1 week, whichever is earlier. By integrating information from cognition, behavior, and brain activity, the system (100) utilizes computational neuroscience and machine learning models to predict after 100% time course of intervention wellness associated with the neuro- cognitive profile, typically of a time length 4-6 weeks. Furthermore, the system (100) possesses the capability to identify abnormal profiles characteristic of neurological and mood disorders, and recommend behavioral and medical interventions suitable for improving wellness.

Description

SYSTEM AND METHOD FOR MONITORING LONGITUDINAL SHIFTS IN NEUROBEHAVIORAL WELLNESS DUE TO THERAPEUTIC INTERVENTIONTECHNICAL FIELD

[0001] The present disclosure pertains to advancements in neuroscience, mainly focusing on leveraging principles of computational cognitive neuroscience and artificial intelligence (Al) to assess the efficacy of uni or multimodal treatment interventions for various neurological disorders. The system and method entail the development of a digital invention designed to serve as a co-pilot for caretaking physicians, offering guidance in treatment decisions. By harnessing Al technology, the system aims to provide comprehensive evaluations of treatment outcomes, aiding physicians in optimizing patient care strategies.BACKGROUND

[0002] In the case of depression treatment, a field both active and extensively researched has attracted significant investments. Major players, such as Mayo Clinic (USA), utilize a combination of clinical health records, behavioral data, and genomic information to achieve a commendable efficiency rate of around 72% within a 4-week timeframe. Furthermore, companies specializing in neural stimulation prediction, including Neuroelectronics (USA), Brainsway (Israel), Neuroetics (USA), Cervel Neurotech (USA), Cerebral (USA), Pear Therapeutics (USA), Spring Health (USA), and Neurosigmal (USA), actively contribute to advancements in this domain. Concurrently, entities like Plvital (USA) focus exclusively on clinical and behavioral data, reporting an approximate 56% accuracy in predicting treatment outcomes over an 8-week period.

[0003] Depression treatment remains a highly researched domain, with institutions like the Mayo Clinic integrating clinical, behavioral, and genomic data to achieve notable response rates. Companies such as Neuroelectronics, Brainsway, Neuroetics, Cerebral, and others are advancing neural stimulation prediction, while entities like Plvital focus on behavioral data analytics.

[0004] Current treatment modalities include antidepressants (e.g., SSRIs), neurostimulation (e.g., rTMS), and psychological therapies (e.g., CBT). Despite progress, no existing solution can predict therapeutic outcomes within a two-week window, which is crucial for timely clinical decisions.

[0005] Traditional approaches often rely on trial-and-error over extended periods, leading to delays in treatment optimization. Recent research into EEG-based predictors andmachine learning models shows promise, but inconsistencies and limited sensitivity remain concerns.

[0006] Therefore, the present disclosure presents a system and method of treating depression that typically involves a trial-and-error process. In this traditional approach, patients are prescribed medications, but it can take 4 to 6 weeks before healthcare providers can accurately assess whether the prescribed treatment is effective for the individual. During this time, patients may experience side effects or find that the medication does not alleviate their symptoms, leading to a need for adjustments or changes in treatment.

[0007] Therefore, there is a need for an Al-driven system that evaluates treatment efficacy using a whole-body cognition framework, enabling rapid, personalized intervention while assisting physicians through actionable insights.OBJECTS OF THE PRESENT DISCLOSURE

[0008] An object of the present disclosure is to introduce an Al-based digital invention for assessing treatment interventions in various neurological disorders, acting as a co-pilot for clinical decisions.

[0009] Another object of the present disclosure is to emphasize the applicability of the invention in both unimodal and multimodal treatment approaches.

[0010] Another object of the present disclosure is to target cognitive and mood disorders like depression, neurological disorders like Parkinson’s disease, trauma, stroke, and epilepsy, any or a combination.

[0011] Another object of the present disclosure is to offer a solution that potentially enhances the effectiveness and precision of treatment interventions.

[0012] Another object of the present disclosure is to promote the development of personalized and tailored treatment strategies for individuals with neurological, mental health disorders.

[0013] Another object of the present disclosure is to contribute to the advancement of technology in healthcare, particularly in the field of neurology, psychiatry, and psychology.

[0014] Yet another object of the present disclosure is to facilitate the monitoring and evaluation of treatment outcomes using Al-driven analytics.SUMMARY

[0015] An aspect of the present disclosure pertain to a system for predicting longitudinal changes in neuro-cognitive behavioral wellness in a user undergoing aneuromodulation-based intervention comprises a server, the server including one or more processors coupled with a memory, where the memory may be configured to store instructions that, when executed by the one or more processors, cause the server to: receive, via a communication network, input data from a user device associated with a user, where the input data corresponds to a first time point (day 0) and a second time point (approximately 25% into the interventional course, or 1 week whichever is earlier); integrate the received input data into a unified user profile; apply a machine learning-based predictive model to the user profile to generate a predicted future neuro-cognitive behavioral wellness profile; classify the predicted future profile as treatment responsive or not responsive, where the nonresponse profile is associated with neurological or mood disorders; and transmit the predicted profile and classification results to a user device for display to a healthcare professional for monitoring and clinical decision support.

[0016] In an aspect, the input data can include any or a combination of cognitive parameters, behavioral indicators, and brain activity signals, including but not limited to PHQ9, GAD7, MMSE, and EEG-based brain maps, particularly for conditions such as depression, anxiety, or cognitive decline. The machine learning -based predictive model is selected from the group consisting of convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory networks (LSTMs), support vector machines (SVMs), and linear or non-linear functional approximate models such as regression models.

[0017] In an aspect, the unified user profile may be generated by applying normalization, feature extraction, and temporal alignment techniques to the input data from the first and second time points.

[0018] In an aspect, the classification of the predicted profile as normal or abnormal based on the response levels may be performed using threshold criteria derived from a labeled training dataset associated with diagnosed neurological or mood disorders. Furthermore, the system may identify and inform about salient features contributing to nonresponse, and recommend behavioral routines or lifestyle modifications to positively intervene in the progression of depressive features.

[0019] Another aspect of the present disclosure pertains to a method for generating cognitive brain maps and extracting features from electroencephalography (EEG) data, where the method can include the steps of receiving, by processor(s) over a communication network, input data from a user device associated with a user, where the input data corresponds to a first time point (day 0) and a second time point (approximately 25% into the interventional course or 1 week whichever is earlier); integrating, by processor(s), thereceived input data into a unified user profile; processing, by processor(s), the unified user profile using a machine learning-based predictive model to generate a predicted future neuro- cognitive behavioral wellness profile; classifying, by processor(s), the predicted profile as either normal or abnormal, where the abnormal classification is indicative of a potential neurological or mood disorder; and transmitting, by processor(s), the predicted profile and the classification result to a user device for use by a healthcare professional in monitoring the user and adjusting the interventional strategy.

[0020] In an aspect, the method for acquiring, preprocessing, and analyzing electroencephalography (EEG) and electrogastrogram (EGG) data — together referred to as EG in the present disclosure — for generating cognitive brain maps and performing feature extraction for identifying neurocognitive characteristics associated with mental health conditions, specifically includes application to depression. The method includes acquiring EG data by recording the electrical activity of the brain and the gut via a plurality of electrodes placed on a user’s scalp and stomach / intestine area.

[0021] In an aspect, the method further involves receiving auxiliary data comprising any or a combination of demographic information, cognitive and behavioral profiles, and self-reported wellness measures such as PHQ9 and GAD7 — commonly used in depression screening — along with MMSE scores and EEG-based brain maps. The EG data is then preprocessed, followed by the generation of a cognitive brain map that represents spatial patterns of brain activity associated with cognitive functions, including attention, memory, and decision-making, which are often impaired in individuals with depression. Features are extracted from this brain map and used to generate a prediction or report tailored for diagnosing or evaluating depression and associated neurocognitive or psychological conditions.

[0022] In an aspect, the method can further include the demographic information that includes any or a combination of age, gender, and educational background of the user.

[0023] In an aspect, the preprocessing of EEG data can include the steps of identifying and removing artifacts including any or a combination of eye blinks, muscle activity, and environmental noise using independent component analysis (ICA) or automated artifact detection techniques; applying filters including a bandpass filter to eliminate high- frequency noise and low-frequency drift; segmenting the EEG data into epochs based on experimental conditions or task events; performing baseline correction by adjusting signal amplitude per epoch; normalizing the EEG signal data across channels and users; applying spatial filtering techniques including spatial averaging or spatial Laplacian; estimating neuralsources using inverse modeling techniques comprising dipole fitting or distributed source localization; and performing quality checks to ensure the integrity and reliability of the preprocessed data. Whereas the EGG data can include steps of spectral filtering to understand the motility of the gut.

[0024] In an aspect, the feature extraction can include the steps of identifying discriminative features indicative of different cognitive states or mental conditions, where the features include spectral power in defined frequency bands, inter-regional connectivity metrics, and spatial distributions of neural activity; and detecting event-related potentials (ERPs) and other synchronization or coherence patterns between brain regions during cognitive tasks.

[0025] Various objects, features, aspects, and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which numerals represent like components.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The following drawings form part of the present specification and are included to further illustrate aspects of the present disclosure. The disclosure may be better understood by reference to the drawings in combination with the detailed description of the specific embodiments presented herein.

[0027] FIG. 1 illustrates an exemplary block diagram architecture for monitoring longitudinal changes in neuro-cognitive behavioral wellness in a user undergoing a neuromodulation-based intervention, in accordance with an embodiment of the present disclosure.

[0028] FIG. 2 illustrates an exemplary module diagram architecture for monitoring longitudinal changes in neuro-cognitive behavioral wellness in a user undergoing a neuromodulation-based intervention, in accordance with an embodiment of the present disclosure.

[0029] FIG.3A illustrates an exemplary flow diagram generating cognitive brain maps and extracting features from electroencephalography and electrogastrogram (EEG and EGG) data, in accordance with an embodiment of the present disclosure.

[0030] FIG. 3B illustrates an exemplary detailed flow diagram for utilizing computational neuroscience and machine learning techniques to reliably predict changes inthe neurocognitive behavioral wellness profile, in accordance with an embodiment of the present disclosure.DETAILED DESCRIPTION

[0031] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. If the specification states a component or feature “may”, “can”, “could”, or “might” be included or have a characteristic, that particular component or feature is not required to be included or have the characteristic.

[0032] As used in the description herein and throughout the claims that follow, the meaning of “a,” “an,” and “the” includes plural reference unless the context clearly dictates otherwise. Also, as used in the description herein, the meaning of “in” includes “in” and “on” unless the context clearly dictates otherwise.

[0033] While embodiments of the present disclosure have been illustrated and described, it will be clear that the invention is not limited to these embodiments only. Numerous modifications, changes, variations, substitutions, and equivalents will be apparent to those skilled in the art, without departing from the spirit and scope of the invention, as described in the claim.

[0034] In an embodiment, the present disclosure enables a system and method for generating cognitive brain maps and extracting features from EG data that may be a combination of Electroencephalogram (EEG) and Electrogastrogram (EGG) to monitor neurocognitive wellness. EEG signals are acquired via scalp electrodes and combined with auxiliary data such as demographics and wellness scores (e.g., PHQ9, GAD7, and MMSE). The data is preprocessed to remove artifacts, filtered, segmented, normalized, and analyzed to estimate neural sources. Cognitive brain maps are generated, and features such as spectral power, connectivity metrics, and ERPs are extracted. Furthermore, the EGG signals are acquired via gut electrodes. These EG whole person cognition features are processed using a machine learning model to predict future neurocognitive profiles, which are classified as normal or abnormal after estimating the extent of response, and sent to a user device to assist healthcare professionals in treatment planning.

[0035] The manner in which the proposed system works is described in further detail in conjunction with FIGs. 1 to 3B. It may be noted that these figures are only illustrative, and should not be construed to limit the scope of the subject matter in any manner.

[0036] FIG. 1 illustrates an exemplary block diagram architecture for monitoring longitudinal changes in neuro-cognitive behavioral wellness in a user undergoing a neuromodulation-based intervention, in accordance with an embodiment of the present disclosure.

[0037] Referring to FIG. 1, a system 100 for predicting longitudinal changes in neuro- cognitive behavioral wellness in a user 112 undergoing a neuromodulation-based intervention includes a server 102 of the system 100, described is a central processing unit that coordinates the collection, processing, and analysis of data related to neuro-cognitive wellness. The server 102 can include one or more processors 104 (interchangeably referred to as a processor 104, hereinafter) coupled with a memory 106, which can store the necessary instructions and algorithms for data handling.

[0038] Furthermore, the server 102 is responsible for receiving input data from a user device 110 through a communication network 108, integrating the data into a unified user profile, and applying machine learning models to predict future neuro-cognitive behavioral wellness profiles. The server 102 can also classify these profiles as normal or abnormal, based on predefined criteria, and transmits the results to the user device 110. The server 102 can enable healthcare professionals to monitor and adjust interventions in real-time, assisting in decision-making and improving patient outcomes.

[0039] In an embodiment, the processor 104 manages the operations related to receiving, analyzing, and predicting the neuro-cognitive behavioral wellness profile based on the input data. The processor 104 can process the data received from the user device 110, apply machine learning models, perform classifications, and generate outputs such as predictions or reports.

[0040] Furthermore, the memory 106 is a storage component that retains the necessary instructions, algorithms, and data structures that the processor 104 uses during these operations. The memory 106 can include both volatile memory (e.g., RAM) for temporary data storage during processing and non-volatile memory (e.g., flash or hard disk storage) for storing permanent data, such as user profiles, processed data, and machine learning models. The memory 106 can ensure that the system 100 can store and retrieve the data efficiently, allowing the processor 104 to execute the tasks in a timely and accurate manner, supporting the overall functionality of the server 102.

[0041] FIG. 2 illustrates an exemplary module diagram architecture for monitoring longitudinal changes in neuro-cognitive behavioral wellness in a user undergoing aneuromodulation-based intervention, in accordance with an embodiment of the present disclosure.

[0042] In an exemplary embodiment, referring to FIG. 2, a system 100 may include one or more processor(s) 104 (interchangeably referred to as a processor 104, hereinafter). The processor 104 may be implemented as one or more microprocessors, microcomputers, microcontrollers, edge or fog microcontrollers, digital signal processors, central processing units, logic circuitries, and / or any devices that process data based on operational instructions. Among other capabilities, the processor 104 may be configured to fetch and execute computer-readable instructions stored in a memory 106 of the system 100.

[0043] Furthermore, the memory 106 may be configured to store one or more computer-readable instructions or routines in a non-transitory computer-readable storage medium, which may be fetched and executed to create or share data packets over a network service. The memory 106 may comprise any non-transitory storage device, including, for example, volatile memory such as Random Access Memory (RAM), or non-volatile memory such as Erasable Programmable Read-Only Memory (EPROM), flash memory, and the like.

[0044] In an exemplary embodiment, the system 100 may include an interface(s) 208. The interface(s) 208 may include a variety of interfaces, for example, interfaces for data input and output devices, referred to as I / O devices, storage devices, and the like. The interface(s) 208 may facilitate communication to / from the system 100. The interface(s) 208 may also provide a communication pathway for one or more components of the system 100. Examples of such components include, but are not limited to, processing unit / engine(s) 210 and a database 202.

[0045] In an embodiment, the processing unitZengine(s) 210 may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine(s) 210. In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing engine(s) 210 may be processor-executable instructions stored on a non-transitory machine-readable storage medium, and the hardware for the processing engine(s) 210 may include a processing resource (for example, one or more processors), to execute such instructions.

[0046] In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine(s) 210. In such examples, the system 100 may include the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or themachine -readable storage medium may be separate but accessible to the system 100 and the processing resource. In other examples, the processing engine(s) 210 may be implemented by electronic circuitry.

[0047] In an embodiment, the database 202 may include data that may be either stored or generated as a result of functionalities implemented by any of the components of the processor 104 or the processing engine 210. In an embodiment, the database 202 may be separate from the system 100.

[0048] In an exemplary embodiment, the processing engine 210 may include one or more engines selected from any of a data acquisition module 212, a preprocessing module 214, a feature extraction module 216, and a prediction and reporting module 218.

[0049] In an embodiment, the data acquisition module 212 of the system 100 is configured to design to collect and process one or more types of data for monitoring and predicting neuro-cognitive wellness. It is also capable of recommending behavioral or medical interventions to precisely improve wellness that is not limited to appetite, sleep, anxiety, cognitive training like management. The module 212 has two primary functions. First, it records the electrical activity of the brain through a plurality of electrodes placed on the user’s scalp and gut. These electrodes capture electroencephalography (EEG) and Electrogastrogram (EGG) data, which reflects the brain’s neural dynamics, providing valuable insights into cognitive states such as attention, memory, and decision-making.

[0050] Second, the module 212 acquires supplementary input, which can include demographic information (such as age, gender, and educational background), cognitive and behavioral profdes, and self-reported wellness measures. The wellness measures can encompass tools like the PHQ-9 (Patient Health Questionnaire-9) for depression assessment, GAD-7 (Generalized Anxiety Disorder-7) for anxiety evaluation, and MMSE (Mini-Mental State Examination) along with any ecological momentary assessments for general cognitive function. Furthermore, the module 212 gathers brain map-related data, which refers to visual representations of brain activity derived from the EEG data. This combination of EEG and supplementary data can enable a comprehensive assessment of the user’s cognitive state and overall wellness, laying the groundwork for subsequent analysis and predictions about neuro- cognitive behavioral wellness overtime.

[0051] In an embodiment, the preprocessing module 214 can be designed to enhance the quality of the EEG data collected by the data acquisition module 212 before it is used for further analysis. The module 214 can perform a series of steps to ensure that the data is clean, accurate, and suitable for generating reliable brain maps and feature extraction. First, themodule 214 can identify and remove artifacts from the EEG data, which may arise from various sources such as eye blinks, muscle movements, and environmental noise. Techniques like independent component analysis (ICA) and automated artifact detection algorithms are applied to eliminate these unwanted signals.

[0052] Second, the module 214 can apply bandpass filtering to remove high- frequency and low-frequency noise components, ensuring that only the relevant brain and gut signals are retained. The EG data is then segmented into epochs, based on task events or experimental conditions, allowing for more focused analysis of specific periods of brain activity. The module 214 can also perform baseline correction and normalization, adjusting amplitude variations across epochs and making it easier to compare the data across different channels or users.

[0053] In an embodiment, spatial filtering techniques such as spatial averaging and spatial Laplacian are applied to improve spatial resolution and reduce blurring of the EG signals. Furthermore, the module 214 uses inverse modeling methods like dipole fitting or distributed source localization to estimate the sources of the recorded EG signals, helping map the brain activity to specific regions or networks, and its coupling with specific gut activity. Finally, quality checks are performed through both visual inspection and automated diagnostics to validate the integrity of the preprocessed data, ensuring that it is accurate and ready for further analysis. The preprocessing module 214 ensures the EG data is optimized for predicting and monitoring neuro-cognitive wellness.

[0054] In an embodiment, the feature extraction module 216 is responsible for deriving meaningful and discriminative information from the preprocessed EG data, which serves as the foundation for subsequent analysis and interpretation of neuro-cognitive states. The module 216 may be configured to extract features that capture the underlying neural dynamics by analyzing the spectral power in predefined frequency bands (such as delta, theta, alpha, beta, and gamma, and its coupling with the bradygastric, normogastric and tachygastric frequency band), which are known to correlate with various mental and cognitive states. The module 216 can also identify spatial distribution patterns of brain activity and connectivity measures that reflect the interaction and communication between different brain regions.

[0055] In an embodiment, the module 216 can focus on extracting features specifically associated with cognitive brain maps, including event-related potentials (ERPs), which are time-locked responses to specific cognitive or sensory events. The module 216 also evaluates synchronization or coherence between brain regions, offering insights into functional connectivity and network-level interactions. Furthermore, the module 216identifies topographical activation patterns that are linked to particular cognitive tasks or mental conditions, allowing the system 100 to map brain function in a task-specific and patient-specific manner. Together, these extracted features provide a comprehensive representation of the user’s neurobehavioral state, enabling accurate assessment and monitoring of cognitive wellness.

[0056] In an embodiment, the prediction and reporting module 218 may be configured to translate the extracted neural features into meaningful insights by generating spatiotemporal whole person cognition brain maps that visually represent physiological activity linked to various cognitive processes. These brain maps are constructed using the extracted features such as spectral power, connectivity, and topographical patterns through time, providing spatially and temporally resolved indicators of brain function.

[0057] In addition to visualization, the module 218 performs analytical integration of the extracted EG features with supplementary data such as demographic profiles, behavioral assessments, and self-reported wellness scores. By correlating neural data with cognitive and psychological markers, the module 218 produces prediction outcomes or diagnostic reports that reflect the user’s neurobehavioral wellness. These outcomes may include early signs or risk levels associated with cognitive or psychological conditions — such as depression, anxiety, or cognitive decline, allowing for longitudinal monitoring, clinical decision support, or personalized wellness feedback.

[0058] FIG.3A illustrates an exemplary flow diagram generating cognitive brain maps and extracting features from electroencephalography and electrogastrography (EG) data, in accordance with an embodiment of the present disclosure.

[0059] As illustrated, a method 300A may be configured to predict longitudinal neuro-cognitive behavioral wellness in a user 112 undergoing neuromodulation-based intervention, and compliance. At step 302A, the method 300A may involve receiving, by processors 104 (interchangeably referred to as a processor 104) over a communication network 108, an input data from a user device 110 associated with a user 112, where the input data is collected at two distinct time points — specifically, a first time point corresponding to day 0, representing the baseline or pre-intervention state of the user 112, and a second time point corresponding to approximately 25% progression or 1 week whichever is earlier into an interventional course.

[0060] In an embodiment, the input data received at these time points may include electrophysiological signals such as EEG recordings, as well as supplementary data comprising demographic details, behavioral assessments, cognitive performance metrics, andself-reported wellness indicators. This time-stamped data enables temporal comparison and longitudinal tracking of changes in neurobehavioral parameters throughout the intervention.

[0061] Continuing further, at step 304A, the method 300A may involve integrating the received input data into a unified user profile, wherein the integration involves aggregating heterogeneous data types including electrophysiological signals, demographic attributes, cognitive and behavioral assessments, and self-reported wellness metrics. The processor 104 systematically associates the collected data with the corresponding user identity and temporal markers to construct a cohesive and structured representation of the user’s neurobehavioral status. The unified profile facilitates subsequent analysis by maintaining data continuity, enabling personalized interpretation, and supporting longitudinal comparisons across multiple time points during the interventional course.

[0062] Continuing further, at step 306A, the method 300A may involve processing the unified user profile using a machine learning-based predictive model to generate a predicted future neuro-cognitive behavioral wellness profile, wherein the predictive model is trained on historical datasets comprising electrophysiological features, demographic parameters, and behavioral wellness indicators. The model applies statistical learning techniques to identify patterns, correlations, and trends within the user’s data and extrapolates these to estimate future cognitive and behavioral wellness outcomes. The processing includes feature vector generation, model inference, and prediction scoring, thereby enabling anticipatory insights into the user’s prospective mental health trajectory.

[0063] Continuing further, at step 308A, the method 300A may involve classifying the predicted neuro-cognitive behavioral wellness profde as either normal or abnormal based on the treatment outcome, where the classification is performed using a trained classification algorithm configured to evaluate the predicted features against predefined thresholds or patterns indicative of clinical relevance. An abnormal classification is indicative of a potential neurological or mood disorder, such as depression, anxiety, or cognitive impairment. The classification step utilizes statistical or machine learning-based discriminative models to assign the appropriate category, thereby enabling early identification and flagging of deviations from normative cognitive and behavioral baselines.

[0064] Continuing further, at step 310A, the method 300A may involve transmitting the predicted neuro-cognitive behavioral wellness profile along with the corresponding classification result to a user device 110 associated with a healthcare professional, over a communication network 108. The transmission can enable the healthcare professional to review the predicted outcome and classification in real-time or near real-time, therebyfacilitating informed decision-making for monitoring the user’s condition and implementing or adjusting an interventional strategy as needed. The transmission may include generating a visual or textual report for clinical interpretation and integration into an electronic health record (EHR) or related monitoring system.

[0065] FIG. 3B illustrates an exemplary detailed flow diagram for utilizing computational neuroscience and machine learning techniques to reliably predict changes in the neurocognitive behavioral wellness profile, in accordance with an embodiment of the present disclosure.

[0066] Referring to FIG. 3B, a flow diagram for longitudinally tracking the changes in neurobehavioral wellness and responsiveness to interventions 300B (hereinafter referred to as method 300B). The method 300B may be configured to refer to a technology developed using artificial intelligence (Al) techniques, any or a combination of machine learning techniques, neural networks, and natural language processing, which is capable of processing and analyzing data to generate insights and make predictions. The use of the Al -based digital invention to evaluate the effectiveness and impact of different treatment approaches or interventions for neurological disorders.

[0067] Furthermore, the method 300B can encompass both single-mode treatments (uni-modal) and combination therapies involving multiple modes of intervention (multimodal). This suggests that the Al-based invention can assess the effectiveness of various treatment strategies, whether they involve a single approach or a combination of approaches. The method 300B can refer to a range of medical conditions affecting the nervous system, including any or a combination of Parkinson’s disease, mood disorders (such as depression or bipolar disorder), trauma-related conditions (like traumatic brain injury), stroke, epilepsy, and others. These disorders often involve complex neurological mechanisms and require tailored treatment approaches.

[0068] In an exemplary embodiment, the method 300B can specify some examples of neurological disorders where the Al-based digital invention can be applied. This list is not exhaustive but rather illustrative of the diverse range of conditions that could benefit from such technology. The ability of the digital tool to forecast or anticipate changes in the person’s well-being and cognitive-behavioral state over an extended period. This suggests that the tool can provide insights into how a person’s mental and behavioral health might evolve over time. The Al-based digital tool may be configured to indicate that the predictive capability is enabled through the use of Al techniques. This can include machine learningtechniques trained on large datasets of neurobehavioral information to make accurate predictions about future states.

[0069] Therefore, the key innovation can lie in the ability to longitudinally predict wellness using Al technology. This implies that such a capability represents a novel advancement in the field, as it goes beyond the existing method 300B for assessing the mental and behavioral health. The practical application of the method 300B is to act like a copilot, which can involve continuously observing and recording changes in the person’s neurobehavioral state to assess their overall wellness. The monitoring capability enables healthcare professionals to intervene or adjust treatment plans as needed based on the predictive insights provided by the tool.

[0070] The method 300B may be configured to involve non-invasive electrophysiology (EG), which is described as an effective and economical tool for examining whole person physiological dynamics. This method can include recording electrical activity in the brain and the gut through electrodes placed on the scalp and stomach area. The EG is known for its reliability and affordability compared to one or more neuroimaging techniques. A Data Acquisition Point 1 may be configured to include step 302B, and 304B where the raw data is collected and recorded.

[0071] In case of depression, a step 302- IB, 302-2B, 302-NB, the method 300B may involve a specific scenario or application context where EEG data 302- IB may be configured to combined with demographic information 302-2B, cognition and behavioral profile 302-3B (hereinafter referred to as the step 302B) and self-reported wellness measures including any or a combination of brain maps as required through EEG, PHQ9 (Patient Health Questionnaire-9), GAD7 (Generalized Anxiety Disorder-7), and MMSE (Mini-Mental State Examination).

[0072] At step 304B, the method 300B may involve a series of preparatory steps to clean, enhance, and refine the EEG data before creating spatial representations of brain activity. In this context, preprocessing ensures that the EEG data is optimized for generating accurate and reliable brain maps tailored to specific research or clinical objectives. The preprocessing steps for brain maps through EEG typically include combination of: identifying and removing artifacts any or a combination of eye blinks, muscle movements, and environmental noise that can distort the EEG signals. Techniques including combination of independent component analysis (ICA) and automated artifact detection techniques commonly used for the artifact removal purpose.

[0073] Furthermore, at step 304B applying filters to remove high-frequency noise and low-frequency drifts from the EEG signals. Bandpass filters can often be used to isolate frequency bands relevant to the brain activity of interest while removing unwanted noise. Dividing the EEG data into smaller segments or epochs based on experimental conditions or task events. This allows for the analysis of brain activity during specific tasks or stimuli. Adjusting the EEG data to establish a baseline or reference point, typically by subtracting the mean or median signal amplitude from each epoch. Baseline correction can help in comparing brain activity across different experimental conditions. Scaling the EEG data to a consistent range to facilitate comparison between different channels or users. Normalization can ensure that variations in signal amplitude do not confound the analysis of brain activity.

[0074] Applying spatial filtering techniques including any or a combination of spatial averaging or spatial Laplacian to enhance the spatial resolution of the EG signals and reduce spatial blurring. Estimating the neural sources of EG signals using inverse modeling techniques such as dipole fitting or distributed source localization. Source localization helps to map brain activity onto specific anatomical regions or functional brain networks, and performs quality checks to ensure the integrity and reliability of the preprocessed EG data. The data may be involved in visually inspecting the data for artifacts and abnormalities and making adjustments as necessary.

[0075] A Data Acquisition Point 1 and Prediction Report at Point 2 may be configured to include step 306B, where the raw data is collected, processed, and analyzed to generate a report or prediction outcome.

[0076] At step 306B, a whole-person cognitive map acquired through electroencephalography and electrogastrogram (EG) can refer to a visual representation of neural activity in the brain associated with cognitive processes. An EEG can include a non- invasive technique, records the brain’s electrical activity through electrodes placed on the scalp. These electrodes detect fluctuations in voltage resulting from ionic current flows within neurons, providing real-time insight into cognitive functions including any or a combination of attention, memory, and decision-making. The cognitive brain map may be configured to derive from EEG data, illustrates spatial patterns of brain activity corresponding to specific cognitive tasks or mental states.

[0077] Furthermore, by analyzing EEG signals, researchers can identify regions of the brain that are active during cognitive processes, map functional connectivity between brain regions, and investigate how cognitive functions are represented and integrated within the brain’s neural networks. Functional connectivity between brain and gut is estimated usingsimultaneous EEG and electrogastrogram (EGG) signal connectivity analysis during various cognitive tasks, and with various bands representing slow, normal, fast motility in the gut. Cognitive brain maps acquired through EG play a crucial role in understanding the neural basis of cognition, informing the development of therapies for cognitive disorders, and guiding interventions aimed at enhancing cognitive performance.

[0078] At step 308B, the method 300B may be configured to extraction feature in the context of preprocessed brain maps acquired through EG involves identifying and extracting relevant characteristics or patterns from the EG data that are indicative of specific brain and body states or cognitive processes. The process may be fundamental for reducing the dimensionality of the data and capturing essential information for further analysis or interpretation.

[0079] In the case of feature extraction of preprocessed cognition maps acquired through EG may be configured to identify discriminative features that differentiate between different conditions. These features can include any or a combination of spectral power in specific frequency bands, connectivity measures between brain regions, and spatial distribution patterns of neural activity. Feature extraction techniques may involve any or a combination of signal processing techniques, statistical methods, and machine learning approaches tailored to EEG data analysis.

[0080] Similarly, in the context of cognitive brain maps acquired through EEG, feature extraction can involve identifying salient features that capture the neural correlates of cognitive processes. The method 300B can include features related to event-related potentials (ERPs), which can characteristic patterns of brain activity elicited by cognitive stimuli or tasks. Other features may include measures of synchronization or coherence between brains regions during cognitive tasks, or topographical patterns of brain activation associated with specific cognitive functions.

[0081] Overall, at step 308B, the method 300B may include feature extraction of preprocessed brain maps acquired through EEG is essential for transforming raw EEG data into meaningful representations that capture the underlying neural dynamics of cognitive processes. The extracted features serve as input for subsequent analysis comprises any or a combination of classification, clustering, or prediction tasks, enabling people to gain insights into brain function and cognition.

[0082] At step 310B, the method 300B may include a component that is designed to acquire knowledge and expertise over time. The component utilizes various machine learningtechniques to analyze data, extract patterns, and develop models that can make predictions or provide insights based on past experiences or input data.

[0083] Furthermore, a comparator Between Point 1 and Point 2 involve a comparison mechanism implemented to assess differences or similarities between data acquired at two distinct points in the data acquisition process. The comparison can involve analyzing changes include any or a combination of in data characteristics, trends, and patterns between Point 1 (an initial data acquisition stage) and Point 2 (a subsequent data acquisition or processing stage).

[0084] Additionally, early Predict the Wellness during Point 3 involves a predictive capability to forecast or anticipate wellness-related outcomes at an early stage of data acquisition or processing, referred to as Point 3. The prediction can involve estimating changes in wellness indicators, including any or a combination of physiological or behavioral measures, based on available data or input features.

[0085] At step 312B, the method 300B may include forecasting the well-being of a person’s neurocognitive state following an intervention at a specified time in the future, typically occurring 4 to 6 weeks after the intervention’s initiation. The predictive task may involve leveraging data collected prior to or during the intervention to estimate the person’s neurocognitive state wellness at the designated future time point.

[0086] To achieve the prediction, one or more factors may be considered, including any or a combination of demographic information, baseline neurocognitive assessments, intervention protocols, and possibly additional monitoring data gathered during the intervention period. Machine learning technique or statistical models may be employed to analyze these data and generate predictions based on patterns and trends observed in the dataset.

[0087] Therefore, the objective of predicting neurocognitive state wellness can provide insights into the effectiveness of the intervention and the likelihood of positive outcomes, include any or a combination of improvements in cognitive function, mood, or overall well-being, at the specified time in the future. The predictive capability can inform clinical decision-making, treatment planning, and resource allocation, allowing for proactive interventions or adjustments to optimize patient outcomes.

[0088] In summary, the present disclosure provides a system and method for monitoring longitudinal changes in neurobehavioral wellness using computational neuroscience and machine learning techniques. The system comprises a server including one or more processors and a memory, operatively coupled with modules configured to acquire,preprocess, and analyze electrophysiology (EG) data along with supplementary inputs such as demographic, cognitive, and wellness-related information. The system performs artifact removal, filtering, segmentation, normalization, and feature extraction from EG data to identify cognitive brain activity patterns. A predictive model is employed to forecast a user’s future neuro-cognitive behavioral wellness profile based on integrated data collected at multiple time points during an intervention course.ADVANTAGES OF THE PRESENT DISCLOSURE

[0089] The present disclosure provides a learning expert built using computational neuroscience and machine learning techniques, predicting the wellness / outcome associated with a treatment protocol well ahead in about 25% of the standard time taken or 1 week whichever is earlier for the case of depression that usually takes 4-6 weeks in standards to assess the outcome of medication.

[0090] The present disclosure acts as a guide for a wide range of neurological disorders, providing monitoring and improvement strategies tailored to each disorder’s unique characteristics

[0091] The present disclosure provides by leverages computational neuroscience principles and advanced machine learning techniques, the method achieves superior predictive capabilities.

[0092] The present disclosure provides an integration that enables sophisticated analysis of complex neurocognitive data, leading to more accurate predictions and personalized interventions.

[0093] The present disclosure also provides recommendations on behavioral interventions sensitive to improving neurobehavioral wellness.

[0094] The present disclosure provides timely insights into potential changes in neurocognitive well-being; the method acts like a guiding co-pilot and empowers healthcare professionals to intervene proactively, adjusting treatment strategies to optimize patient outcomes.

Claims

We Claim:

1. A system (100) for predicting longitudinal changes in neuro-cognitive behavioral wellness in a user (112) undergoing a neuromodulation-based intervention comprises: a server (102) comprising: processor(s) (104) coupled with a memory (106), wherein said memory (106) configured to store instructions that, when executed by the one or more processors (104), cause the server (102) to: receive, via a communication network (108), input data from a user device (110) associated with a user (112), wherein the input data corresponds to a first time point (day 0) and a second time point (approximately 25% into the interventional course); integrate the received input data into a unified user profile; apply a machine learning-based predictive model to the user profile to generate a predicted future neuro-cognitive behavioral wellness profile; classify the predicted future profile as normal or abnormal, wherein the abnormal profile is associated with neurological or mood disorders; and transmit the predicted profile and classification results to a user device (110) for display to a healthcare professional for monitoring and clinical decision support.

2. The system (100) as claimed in claim 1, wherein the input data comprising any or a combination of cognitive parameters, behavioral indicators, self-reports PHQ9, GAD7, MMSE, and EEG-EGG based whole person cognition map.

3. The system (100) as claimed in claim 1, wherein the machine learning -based predictive model is selected from the group consisting of: convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory networks (LSTMs), support vector machines (SVMs), regression models or any linear or non-linear functional approximate models.

4. The system (100) as claimed in claim 1, wherein the classification of the predicted profile as normal or abnormal is based on threshold criteria derived from a labeled training dataset associated with diagnosed neurological or mood disorders.

5. The system (100) as claimed in claim 4, wherein the system (100) is further configured to recommend precise and personalized behavioral and medical interventions comprising at least one of sleep regulation, appetite management, anxiety reduction techniques, social inclusivity practices, and cognitive training, the recommendations being generated based on response data from predictive models trained on similarity data corresponding to the test sample.

6. A method (300A) for generating cognitive brain maps and extracting features from electroencephalography (EEG) data, wherein the method (300A) comprises the steps of: receiving (302A), by processor(s) (104), over a communication network (108), an input data from a user device (110) associated with a user (112), wherein the input data corresponds to a first time point (day 0) and a second time point (approximately 25% into the interventional course or 1 week whoever is earlier); integrating (304A), by processor(s) (104), the received input data into a unified user profile; processing (306A), by processor(s) (104), the unified user profile using a machine learning-based predictive model to generate a predicted future neuro- cognitive behavioral wellness profile; classifying (308A), by processor(s) (104), the predicted profile as either normal or abnormal, wherein the abnormal classification is indicative of a potential neurological or mood disorder; and transmitting (310A), by processor(s) (104), the predicted profile and the classification result to a user device (110) for use by a healthcare professional in monitoring the user (112) and adjusting the interventional strategy.

7. The method (300A) as claimed in claim 6, wherein for acquiring, preprocessing, and analyzing electroencephalography and electrogastrography (EG) data to generate cognitive brain maps and perform feature extraction for identifying neurocognitive characteristics associated with mental health conditions, wherein the method (300A) comprises the steps of:acquiring EEG data by recording electrical activity of the brain via a plurality of electrodes placed on a user’s scalp and gut area; receiving auxiliary data comprising any or a combination of demographic information, cognitive and behavioral profiles, and self-reported wellness measures, comprises PHQ9, GAD7, MMSE, and EG-based cognition maps; preprocessing the EG data; generating a cognitive brain map from the preprocessed EG data, wherein the whole person cognitive map represents patterns of activity associated with cognitive functions comprising any or a combination of attention, memory, and decision-making; extracting features from the whole person cognitive map; and generating a prediction or report based on the extracted features for use in diagnosing or evaluating a neurocognitive or psychological condition.

8. The method (300A) as claimed in claim 7, further comprises demographic information that comprises any or a combination of age, gender, and educational background of the user, wherein the demographic information is utilized to enhance prediction precision and to personalize recommended behavioral or therapeutic interventions.

9. The method (300A) as claimed in claim 7, wherein the pre-processing of EG data comprises the steps of: identifying and removing artifacts comprising any or a combination of eye blinks, muscle activity, and environmental noise using independent component analysis (ICA) or automated artifact detection techniques; applying filters comprising a bandpass filter to eliminate high- frequency noise and low-frequency drift; segmenting the EG data into epochs based on experimental conditions or task events; performing baseline correction by adjusting signal amplitude per epoch; normalizing the EG signal data across channels and users; applying spatial filtering techniques comprising spatial averaging or spatial Laplacian;estimating neural sources using inverse modeling techniques comprising dipole fitting or distributed source localization; and performing quality checks to ensure the integrity and reliability of the preprocessed data.

10. The method (300A) as claimed in claim 7, wherein the feature extraction comprises the steps of: identifying discriminative features indicative of different cognitive states or mental conditions, wherein the features comprise spectral power in defined frequency bands, inter-regional connectivity metrics, and spatial distributions of neural activity; and detecting event-related potentials (ERPs) and other synchronization or coherence patterns between brain regions during one or more cognitive tasks.

Images