A physiological disorder alert system for alerting a subject and a method thereof

Abstract

The present disclosure discloses a physiological disorder alert system (114) for alerting a subject. The system (112) includes a plurality of biosensor electrodes (102), and a controller (104). The plurality of biosensor electrodes (102) measure at least one physiological parameter of the subject (112). The controller (104) detects a plurality of anomalies present in the EEG signal based on comparing the EEG signal with an updated output from a predetermined longitudinal EEG using at least one of Artificial Intelligence model and determines a presence of a plurality of biomarkers associated with at least one anomaly in the EEG signal, using a transformer-based machine learning model. The controller predicts an occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with at least one anomaly is present in the EEG signal and generate the physiological disorder alert for a user equipment (UE) (116).

Description

A PHYSIOLOGICAL DISORDER ALERT SYSTEM FOR ALERTING A SUBJECT AND A METHOD THEREOF TECHNICAL FIELD

[0001] The present disclosure relates to physiological disorders, and more particularly relates to a physiological disorder alert system for alerting a subject. BACKGROUND

[0002] In today’s world, human beings are suffering from several physiological disorders, for example, seizures and dementia, due to various factors. The recurrent seizures may cause epilepsy. Epilepsy is a brief episode of involuntary movement that involves a part of the body (partial) or the entire body (generalized) of human beings. Further, this also involves higher rates of psychological conditions, including anxiety and depression. Currently, the estimated proportion of the general population with active epilepsy (i.e., continuing seizures or with the need for treatment) at a given time is between 4 and 10 per 1000 people.

[0003] Further, especially during sleep, the seizures often occur unnoticed and represent the primary cause of Sudden Unexpected Death in Epilepsy (SUDEP) of a patient having uncontrolled seizures. Seizures affect the normal functioning of the body of human beings by affecting breathing and heart rates of human beings, potentially leading to life-threatening situations.

[0004] Additionally, the sudden and prolonged uncontrolled event of seizures in human beings may also result in death among individuals. Seizures can sometimes lead to injuries or falls, contributing to severe injuries requiring hospitalization, such as a traumatic head injury, and an increased risk of death.

[0005] Thus, to protect human beings from the impact of seizures, there is a need for medical services that can detect the occurrence of seizures in human beings and thus provide treatment to human beings suffering from seizures, accordingly. However, there still exists a gap between the need for medical / healthcare services and their availability. The shortage of healthcare professionals, lack of necessary health infrastructure, and costly diagnostic facilities impact the prolonged treatment of human beings suffering from seizures.

[0006] Particularly, the treatment of seizures is costly, as detecting seizures and epilepsy caused by the seizure, an electroencephalogram (EEG) test is done, which is not cost-effective. Conventional EEG requires a specific device that is bulky, thus reducing the accessibility of the device to individuals. Further, there is limited availability of healthcare professionals to analyze the data and compounds generated from the device. This may also lead to delay or inadequate documentation of the data, which can complicate treatment decisions and hinder monitoring of advancements. Moreover, the shortage of ambulatory devices and the lack of options for monitoring seizures while maintaining the comfort of the individual contribute to a significant gap in accessible and comprehensive medical / healthcare services related to seizures.

[0007] Thus, there is a need to provide a system that can detect, track, predict, warn and classify the occurrence of physiological disorders like Seizures and Dementia in the individual while overcoming one or more above mentioned problems. SUMMARY

[0008] This summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the invention. This summary is neither intended to identify key or essential inventive concepts of the invention and nor is it intended for determining the scope of the invention.

[0009] The present disclosure discloses a physiological disorder alert system for alerting a subject. The system includes a plurality of biosensor electrodes, and a controller. The plurality of biosensor electrodes is in contact with at least one of a scalp and an ear canal of the subject through at least one of elastic members or clip-based structures of at least one wearable device thereby establishing required skin contact with the subject. The plurality of biosensor electrodes is configured to measure at least one physiological parameter of the subject. The at least one physiological parameter is Electroencephalogram (EEG) signal. The controller is in communication with the plurality of biosensor electrodes. The controller is configured to detect a plurality of anomalies present in the EEG signal based on comparing the EEG signal with an updated output from a predetermined longitudinal EEG trend corresponding to daily activities and sleep of the subject using at least one of Artificial Intelligence model. The controller is configured to determine a presence of a plurality of biomarkers associated with at least one anomaly in the EEG signal, comprising Interictal Epileptiform Discharges (IEDs) and High-Frequency Oscillations (HFOs),using a transformer-based machine learning model, upon the detection. The controller is configured to predict an occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with at least one anomaly is present in the EEG signal. The controller is configured to generate the physiological disorder alert for a user equipment (UE). The physiological disorder alert indicates a notification for the subject, thereby predicting the occurrence of the physiological disorder.

[0010] In another embodiment, a method for generating a physiological alert is disclosed. The method includes detecting a plurality of anomalies present in EEG signal based on comparing the EEG signal with an updated output from a predetermined longitudinal EEG trend corresponding to daily activities and sleep of the subject using at least one of Artificial Intelligence model. The method includes determining a presence of a plurality of biomarkers associated with at least one anomaly in the EEG signal, comprising Interictal Epileptiform Discharges (IEDs) and High- Frequency Oscillations (HFOs), using a transformer-based machine learning model, upon the detection. The method includes predicting an occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with at least one anomaly is present in the EEG signal. The method includes generating the physiological disorder alert for a user equipment (UE), wherein the physiological disorder alert indicates a notification for the subject, thereby predicting the occurrence of the physiological disorder.

[0011] To further clarify the advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] These and other features, aspects, and advantages of the present invention will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:Figure 1 illustrates a block diagram for a physiological disorder alert system for alerting a subject, in accordance with an embodiment of the present disclosure; Figures 2A-2D illustrate a plurality of biosensor electrodes in different embodiments of a device for detecting, tracking, predicting and warning an occurrence of a physiological disorder, in accordance with an embodiment of the present disclosure; Figure 3 illustrates a block diagram of a controller for predicting the occurrence of the physiological disorder, in accordance with an embodiment of the present disclosure; Figure 4 illustrates a flowchart depicting a prediction of the occurrence of a physiological disorder in the subject, in accordance with an embodiment of the present disclosure; Figure 5 illustrates a flowchart depicting a process to reduce the impact of the physiological disorder in the subject, in accordance with an embodiment of the present disclosure; Figure 6 illustrates a flowchart depicting an assessment of the subject’s cognitive function, in accordance with another embodiment of the present disclosure; and Figure 7 illustrates a flowchart depicting a method performed by the system for generating a physiological disorder alert for alerting the subject, in accordance with an embodiment of the present disclosure. Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have necessarily been drawn to scale. Furthermore, in terms of the construction of the device, a plurality of components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present invention so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein. DETAILED DESCRIPTION OF FIGURES

[0013] For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates. Unless otherwise defined, all technical and scientific terms used herein have the same meaning ascommonly understood by one of ordinary skilled in the art to which invention belongs. The system and examples provided herein are illustrative only and not intended to be limiting.

[0014] For example, the term “some” as used herein may be understood as “none” or “one” or “more than one” or “all.” Therefore, the terms “none,” “one,” “more than one,” “more than one, but not all” or “all” would fall under the definition of “some.” It should be appreciated by a person skilled in the art that the terminology and structure employed herein is for describing, teaching, and illuminating some embodiments and their specific features and elements and therefore, should not be construed to limit, restrict, or reduce the spirit and scope of the present disclosure in any way.

[0015] For example, any terms used herein, such as “includes,” “comprises,” “has,” “consists,” and similar grammatical variants do not specify an exact limitation or restriction, and certainly do not exclude the possible addition of a plurality of features or elements, unless otherwise stated. Further, such terms must not be taken to exclude the possible removal of the plurality of the listed features and elements, unless otherwise stated, for example, by using the limiting language including, but not limited to, “must comprise” or “needs to include.”

[0016] Whether or not a certain feature or element was limited to being used only once, it may still be referred to as “plurality of features” or “plurality of elements” or “at least one feature” or “at least one element.” Furthermore, the use of the terms “plurality of” or “at least one” feature or element do not preclude there being none of that feature or element, unless otherwise specified by limiting language including, but not limited to, “there needs to be plurality of…” or “plurality of elements is required.”

[0017] Unless otherwise defined, all terms and especially any technical and / or scientific terms, used herein may be taken to have the same meaning as commonly understood by a person ordinarily skilled in the art.

[0018] Reference is made herein to some “embodiments.” It should be understood that an embodiment is an example of a possible implementation of any features and / or elements of the present disclosure. Some embodiments have been described for the purpose of explaining plurality of the potential ways in which the specific features and / or elements of the proposed disclosure fulfil the requirements of uniqueness, utility, and non-obviousness.

[0019] Use of the phrases and / or terms including, but not limited to, “a first embodiment,” “a further embodiment,” “an alternate embodiment,” “one embodiment,” “an embodiment,” “multiple embodiments,” “some embodiments,” “other embodiments,” “further embodiment”, “furthermore embodiment”, “additional embodiment” or other variants thereof do not necessarily refer to the same embodiments. Unless otherwise specified, plurality of particular features and / or elements described in connection with plurality of embodiments may be found in one embodiment, or may be found in more than one embodiment, or may be found in all embodiments, or may be found in no embodiments. Although plurality of features and / or elements may be described herein in the context of only a single embodiment, or in the context of more than one embodiment, or in the context of all embodiments, the features and / or elements may instead be provided separately or in any appropriate combination or not at all. Conversely, any features and / or elements described in the context of separate embodiments may alternatively be realized as existing together in the context of a single embodiment.

[0020] Any particular and all details set forth herein are used in the context of some embodiments and therefore should not necessarily be taken as limiting factors to the proposed disclosure.

[0021] Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

[0022] Figure 1 illustrates a block diagram for a physiological disorder alert system 114 (referred to herein as a system 114) for alerting a subject 112, in accordance with an embodiment of the present disclosure.

[0023] In an embodiment, the system 114 may include a plurality of biosensor electrodes 102, an adaptive protective member 106, a plurality of sensors 108, and a controller 104. In such an embodiment, the plurality of biosensor electrodes 102 (referred to herein as an electrode 102), the adaptive protective member 106, and the plurality of sensors 108 may be integrated in at least one wearable device 100 (referred to herein as a device 100), without departing from the scope of the present disclosure. Further, the controller 104 may be one of in communication or disposed in the device 100. In an embodiment, the system 114 may be communicatively coupled with the device 100, without departing from the scope of the present disclosure. In another embodiment, the system 114 may be deployed in the device 100, without departing from the scope of the present disclosure.

[0024] In an embodiment, the system 114 may be configured to detect a plurality of anomalies present in Electroencephalogram (EEG) signal of the subject 112. The system 114 may be configured to detect the anomalies based on comparing the EEG signal with an updated output from a predetermined longitudinal EEG trend corresponding to daily activities and sleep of the subject 112. The system 114 may be configured to detect anomalies using at least one of Artificial Intelligence model. The system 114 may be configured to determine a presence of a plurality of biomarkers associated with at least one anomaly in the EEG signal using a transformer-based machine learning model, upon the detection. The plurality of biomarkers may include, but is not limited to, Interictal Epileptiform Discharges (IEDs) and High-Frequency Oscillations (HFOs). The system 114 may be configured to predict an occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with at least one anomaly is present in the EEG signal. The system 114 may be configured to generate the physiological disorder alert for a user equipment (UE) 116. The physiological disorder alert indicates a notification for the subject 112, thereby predicting the occurrence of the physiological disorder.

[0025] In an advantageous aspect, the system 114 generates the alert for alerting the subject by predicting the occurrence of the physiological disorder, which results in improved safety of the patient, ensures timely medical intervention, and enhances quality of life of the subject, etc.

[0026] The constructional and functional details of the system 114, along with the device 100 are explained in the subsequent paragraphs.

[0027] Figures 2A-2D illustrate the electrode 102 in different embodiments of the device 100 which may be adapted to detect, track, predict, and warn about the occurrence of the physiological disorder in the subject 112, in accordance with an embodiment of the present disclosure.

[0028] In an embodiment, the device 100 may be adapted to predict the occurrence of the physiological disorder in the subject 112, without departing from the scope of the present disclosure. In one example, the physiological disorder may be a seizure(s), dementia, A lzheimer’s, etc.

[0029] In an embodiment, the device 100 may be at least one of an ear-wearable device and a scalp wearing device, without departing from the scope of the present disclosure. The constructional detail of the device 100 for different embodiments may be explained in subsequent paragraphs.

[0030] In an embodiment, referring to Figure 2A, the device 100 may include a housing 104. The housing 104 may be adapted to support a hub 202, a plurality of elastic members 107 on the hub 202, and the plurality of biosensor electrodes 102. In an embodiment, the plurality of elastic members 107 may be a spring, without departing from the scope of the present disclosure.

[0031] In an embodiment, the electrode 102 may in contact with the at least one of the scalp and the ear canal of the subject 112 through at least one of elastic members 107 or clip-based structures (110) (explained with reference to Figure 2B), thereby establishing skin contact with the subject 112 during dynamic motion and sleep. The electrode 102 may be positioned in each spring, without departing from the scope of the present disclosure. The electrode 102 may be in contact with at least one of a scalp and an ear canal of the subject with the help of each spring, when the subject, suffering from the physiological disorder, wears the device 100. The plurality of springs 107 may be adapted to expand and retract to efficiently engage and disengage each electrode with the ear canal. Each electrode may be configured to measure at least one physiological parameter of the subject 112. The at least one physiological parameter may be the EEG signal. Further, the at least one physiological parameter may also include, but is not limited to, Electromyography (EMG) signal, and autonomic signals.

[0032] Referring to Figure 2B, in another embodiment, the device 100 may include the housing 104. The housing 104 may be adapted to support a plurality of clip structures 110 and the electrode 102. In another embodiment, the plurality of clip structures 110 may be based on the plurality of elastic members (not shown). Further, each electrode may be adapted to be positioned on each leg of the plurality of clip structures 110. The plurality of clip structures 110 may be configured to expand and retract to efficiently engage and disengage each electrode with the ear canal. Each electrode may be adapted to measure at least one physiological parameter of the subject 112.

[0033] The device 100, as explained in above mentioned paragraphs, ensures a stabilized configuration of each electrode. The configuration ensures efficient contact of each electrode with the ear canal even while moving the head, thus maintaining the working efficiency of each electrode. The device 100 ensures the optimal positioning of each electrode, even in a dynamic environment, by dynamically adjusting based on contact evaluation parameters. The optimal positioning of each electrode may be ensured by the plurality of springs as provided in the device 100, which significantly enhances measurement reliability and accuracy, marking a significant advancement in sensor-based measurements.

[0034] In yet another embodiment, referring to Figure 2C, the device 100 may be a band having a flexible structure that may be worn on the head of the subject 112. The band may be adapted to monitor the brain wave of the subject 112 and execute real-time insights and alerts. This configuration ensures seamless integration into the subject's daily routine, ensuring uninterrupted monitoring of brain activity throughout various activities, including sleep. In an embodiment, the band may be made of a breathable and stretchable fabric, thereby ensuring a secure fit, optimizing the accuracy and reliability of the data collected.

[0035] In one embodiment, the band may include the adaptive protective member 106. The adaptive protective member 106 may include at least one of a plurality of inflation members 106a and expandable shields. The at least one of the plurality of inflation members 106a and the expandable shields may be adapted to receive the electrode 102. Further, the at least one of the plurality of inflation members 106a and expandable shields inflates to distribute impact forces, upon predicting the occurrence of the physiological disorder, thereby ensuring adaptive safeguarding of the subject 112.

[0036] When the band detects an impending fall, accident, or seizure, each inflation member or the expandable shield rapidly inflates to protect the individual. This uniform inflation around the head distributes the impact force and minimizes localized pressure points. Additionally, the band may sense the direction and magnitude of specific forces involved by utilizing real-time data provided by Micro-Electro-Mechanical Systems (MEMS) based sensors, for example, gyroscopes and accelerometers. Therefore, by using the real-time data, each inflation member automatically adjusts the level of inflation to provide optimal shock absorption for the detected impact.

[0037] In an embodiment, the at least one of the plurality of inflation member 106a and expandable shields may operate in conjunction with structures that are made up of an anisotropic materials, non-Newtonian fluids, materials with negative Poisson’s ratio. Additionally, the at least one of the plurality of inflation member 106a and expandable shields may include multi- chambered silicone bladders, anisotropic stiffness polymers for directional force resistance, and shear-thickening non-Newtonian matrices (e.g., silica nanoparticle suspensions). The at least one of the plurality of inflation member 106a and expandable shields may be adapted to execute response which are as follows: impact force distribution: anisotropic materials channel kinetic energy away from critical cranial regions (e.g., temporal lobe), reducing localized pressure by 40%–60% and inflatable membrane: by deploying air cushions in case of emergency to absorb thefall impact effectively reduces lethal injuries, thus, often saving the subject 112. This may be performed either through small explosive charges that are triggered in such a condition or through a controlled bladder inflation.

[0038] Further, the expandable shields may be activated based on the insights generated by the device 100 upon detection and prediction of the occurrence of the physiological disorder in the subject 112.

[0039] In yet another embodiment, referring to Figure 2D, the device 100 may include a head band 204 and a plurality of headphones 206. The plurality of headphones 206 may be integrated with the head band 204 and enclose entire ear for better audio isolation. Each headphone may include the electrode 102, without departing from the scope of the present disclosure.

[0040] In an embodiment, the plurality of sensors 108 may be integrated with the device 100. The plurality of sensors 108 may be coupled to at least one of the electrode and a biosensing insight module of the device 100 to monitor at least one of the plurality of anomalies and incoming motion data to dynamically modulate the adaptive protective member to mitigate injury. In an embodiment, the plurality of sensors may include, but is not limited to, gyroscopes, and accelerometers.

[0041] In an embodiment, the system 114 may include a Near-field communication (NFC) tag which may be disposed in the device 100. Thus, when the device 100 may be in proximity to another NFC tag, the system 114 may provide swift access to essential subject details to caregivers, first responders, or medical personnel. This enables prompt and appropriate care during emergencies, especially when the subject is unable to communicate effectively during the occurrence of the physiological disorder.

[0042] Further, the system 114 may be adapted to detect and predict the occurrence of the physiological disorder through the controller 104. The controller 104 receives and monitors the subject’s neural dynamics, encompassing intricate brainwave oscillations and neuromuscular activity, to discern subtle precursory signatures indicative of an impending physiological order occurrence. The constructional and operational details of the controller 104 may be explained in subsequent paragraphs.

[0043] Figure 3 illustrates a block diagram of the controller 104, in accordance with an embodiment of the present disclosure.

[0044] The controller 104 may be in communication with the electrode 102, without departing from the scope of the present disclosure. The controller 104 receives data related to the physiological parameters detected by each electrode and predicts the occurrence of the physiological disorder in the subjects.

[0045] In an embodiment, the controller 104 may operate based on a plurality of processes, for example, an embedded edge computing, diffractive deep neural network, machine learning model, insights generation, etc. The edge computing may be used to mitigate challenges associated with latency, energy consumption, and privacy as in the conventional cloud-based approaches. The embedded edge computing neural network is configured for low-latency processing of multi- channel biosignals. The embedded edge computing neural network utilizes either electronic circuitry or a diffractive optical neural network architecture to enable real-time detection of the plurality of anomalies.

[0046] The controller 104 includes a processor 304, a memory 306, module(s) 308. The memory 306, in one example, may store the instructions to carry out the operations of the modules 308. The modules 308 and the memory 306 may be coupled to the processor 304.

[0047] The processor 304 can be a single processing unit or several units, all of which could include multiple computing units. The processor 304 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and / or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor 304 is configured to fetch and execute computer-readable instructions and data stored in the memory 306. The processor 304 may include one or a plurality of processors. At this time, one or a plurality of processors may be a general-purpose processor, such as a central processing unit (CPU), an application processor (AP), or the like, a graphics-only processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU), and / or an AI-dedicated processor such as a neural processing unit (NPU). The one or a plurality of processors control the processing of the input data in accordance with a predefined operating rule or artificial intelligence (AI) model stored in the non-volatilememory and the volatile memory. The predefined operating rule or machine learning model is provided through training or learning.

[0048] The memory 306 may include any non-transitory computer-readable medium known in the art including, for example, volatile memory, such as static random-access memory (SRAM) and dynamic random-access memory (DRAM), and / or non-volatile memory, such as read-only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.

[0049] The modules 308, amongst other things, include routines, programs, objects, components, data structures, etc., which perform particular tasks or implement data types. The modules 308 may also be implemented as, signal processor(s), state machine(s), logic circuitries, and / or any other device or component that manipulates signals based on operational instructions.

[0050] Further, the modules 308 can be implemented in hardware, instructions executed by a processing unit, or by a combination thereof. The processing unit can comprise a computer, a processor, such as the processor 304, a state machine, a logic array, or any other suitable devices capable of processing instructions. The processing unit can be a general-purpose processor that executes instructions to cause the general-purpose processor to perform the required tasks or, the processing unit can be dedicated to performing the required functions. In another embodiment of the present disclosure, the modules 308 may be machine-readable instructions (software) which, when executed by the processor 304 / processing unit, perform any of the described functionalities. Further, the data serves, amongst other things, as a repository for storing data processed, received, and generated by one or more of the modules 308.

[0051] The modules 308 may perform different functionalities which may include , but may not be limited to predict the occurrence of the physiological disorder in the subjects and subsequently alerting the subject 112. In an embodiment, the modules 308 may include a receiving module 310, a detecting module 312, a storing module 313, a determining module 314, a segregating module 316, a converting module 318, a supplementing module 320, a transmitting module 322, a prepending module 324, an aggregating module 326, a predicting module 328, an extracting module 330, a generating module 332, a correlating module 334, an adjusting module 336, and a delivering module 338.. Each module 310-338 may be in communication with each other. Eachmodule 310-338 may be adapted to perform operations to predict the occurrence of the physiological disorder as explained in subsequent paragraphs.

[0052] Figure 4 illustrates a flowchart depicting the prediction of the occurrence of the physiological disorder in the subject 112, in accordance with an embodiment of the present disclosure.

[0053] Further, in an embodiment, as shown at step 402, the electrode 102 measures the at least one physiological parameter (referred to herein as EEG signals / biosignals) of the subject 112. In an embodiment, each electrode may be placed to capture brain wave data from various regions of the brain. In another embodiment, other physiological parameters beyond heart rate, EMG, and galvanic skin response may also be measured to determine and predict seizures. The other physiological parameters may include, but are not limited to, heart rate variability (HRV), electrocardiogram (ECG) signals, blood pressure changes, and a combination of ECG and EMG signals. This may also be sourced from other wearables, for example, smart watches, glasses, or rings. Adopting multimodal approaches that integrate multiple physiological signals from different modalities could potentially enhance the accuracy and reliability of the system 114 by providing a more holistic view of the complex physiological changes associated with seizure events.

[0054] Further, at step 404, a biosignal acquisition unit having a plurality of sensors may be adapted to receive the EEG signals from each electrode. In an embodiment, EEG signals are received instantaneously, without any delay from each electrode, which may be used for further processing to detect & predict the occurrence of the physiological disorder. In an embodiment, the biosignal acquisition unit may include a decentralized data storage system including multiple nodes distributed across a network. Each node stores encrypted portions of the EEG signals. Additionally, the system 114 incorporates an artificial intelligence (AI) module trained to analyze the stored data of the EEG signals and extract insights regarding the subject's health status, physiological trends, and anomalies. The AI module utilizes federated learning techniques to collaboratively analyze the distributed data across the decentralized storage system while ensuring subject privacy and data security. Thus, by this configuration, the biosignal acquisition unit enables seamless acquisition, secure storage, and insightful analysis of the data associated with the EEG signals, fostering personalized healthcare monitoring and management for subjects / individuals.

[0055] In an embodiment, the plurality of sensors may include an EEG sensor, an Inertial Measurement Unit sensor, and other physiological sensors, without departing from the scope of the present disclosure.

[0056] Further, at step 406, the receiving module 310 may be configured to receive EEG signals as an input from the biosignal acquisition unit. Thereafter, the determining module 314, along with a local AI model and a Foundational AI model as shown at step 418 and step 420, respectively, may be configured to determine and analyse the EEG signals to detect the health of the brain.

[0057] After determining and analysing the EEG signals, the EEG signals further undergo anomaly detection at step 408. In such an embodiment, the detecting module 312 may be configured to detect the plurality of anomalies present in the EEG signal. The detecting module 312 may be configured to detect each anomaly based on comparing the EEG signal with the updated output from the predetermined longitudinal EEG trend. The predetermined longitudinal EEG trend may correspond to the daily activities and sleep of the subject 112 using at least one of the Artificial Intelligence model.

[0058] In an embodiment, to detect the plurality of anomalies, the detecting module 312 may be configured to assess photosensitive physiological disorder via controlled visual exposure and audio sensitive physiological disorder via controlled audio exposure while monitoring Visual Evoked Potential (VEP) and Auditory Evoked Potential (AEP) biomarkers respectively. Thereafter, the adjusting module 336 may be configured to integrate subliminal frequency modulations into audiovisual stimuli to modulate neural oscillations, thereby inducing adaptive neuroplasticity through neurotherapeutic conditioning to detect the plurality of anomalies.

[0059] Prior to detecting the plurality of anomalies, the storing module 313 may be configured to passively record and store longitudinal EEG data via the device 100 during daily activities and sleep. Further, the generating module 332 may be configured to generate the predetermined longitudinal EEG trend based on preprocessing the longitudinal EEG data based on predetermined techniques such as wavelet decomposition, phase-space reconstruction.

[0060] In an embodiment, upon detecting the plurality of anomalies, at step 410, the determining module 314 may be configured to determine the presence of the plurality of biomarkers associated with at least one anomaly, for example, seizure, in the EEG signal. The plurality of biomarkers may include, but is not limited to, the Interictal Epileptiform Discharges (IEDs) and the High-Frequency Oscillations (HFOs). In another embodiment, the plurality of biomarkers may be any biomarker, without departing from the scope of the present disclosure. The determining module 314 may be configured to determine the presence of the plurality of biomarkers using the machine learning model. In an embodiment, the machine learning model may be, but is not limited to, a transformer-based model.

[0061] Further, to determine the presence of the plurality of biomarkers, the segregating module 316 may be configured to segregate the EEG signal into fixed-length patches. The patches indicate overlapping time windows. The converting module 318 may be configured to convert, linearly, each patch into a token embedding. The supplementing module 320 may be configured to supplement, converted patch with predefined positional encodings to enable a large context window spanning multiple seconds of neural activity. The transmitting module 322 may be configured to transmit, a complete sequence of token embeddings through a stack of transformer- encoder layers. Each transformer-encoder layer may include multi-head self-attention and feed- forward sublayers to capture long-range spatiotemporal dependencies, indicating correlations across sub-band features such as ripples (80–200 Hz) and fast ripples (200–500 Hz). The prepending module 324 may be configured to prepend a classification token to the complete transmitted sequence. Further, the aggregating module 326 may be configured to aggregate the encoded information via the classification token. Lastly, the determining module 314 may be configured to determine the presence of biomarkers based on the aggregated encoded information through layer normalization and a softmax output layer, pretrained on labelled datasets.

[0062] The IEDs are substantial intermittent electrophysiological occurrences observed between seizures in subjects diagnosed with epilepsy. Interictal epileptiform discharges (IEDs), which may not be detected by epileptologists during visual interpretation, can be identified by the device 100 and hence are more efficient than existing techniques as used. Further, HFOs are used for early identification of epileptic seizures, supplementing the Interictal Epileptiform Discharges (IEDs) data and ictal activity data. The brain activities observed in the EEG recorded data (EEG) within the frequency ranges of 80–500 Hz are categorized into ripples (80–200 Hz) and fast ripples (200– 500 Hz) based on their distinct characteristics. The HFO activities are used to identify the onset of seizures, as well as to record them during seizure activity. Further, the data may also be analysed to detect other anomalies, for example, sleep anomalies , sleep apnea traits, etc.

[0063] Additionally, upon the detection that the plurality of anomalies is present in the EEG signals, at step 422, the storing module 313 may be configured to store the EEG signals in a repository communicatively coupled with the controller 104. The EEG signals may be further used by the Foundational AI model at step 420 for future operation to generate a desired result. Further, the desired result may be further communicated to the subject 112 through a stimulus generator of the controller 104 and a stimulus interface, for example, display unit, as shown at step 414.

[0064] In an embodiment, utilizing the repository (cloud-based storage) for the data of the EEG signals creates an accessible platform for healthcare professionals and researchers worldwide to retrieve patient data remotely and prescribe appropriate medication. This data serves as a foundation for collaborative study, analysis, and identification of hidden biomarkers crucial for diagnosing, tracking, and predicting various neurological disorders. This is especially vital for running Large Language Models (LLMs) trained on biosignal data to find transformer-mode- based insights with much more speed and accuracy.

[0065] The centralized processing and learning in cloud infrastructure enable the processing and analysis of large volumes of data. This increased access to broader datasets leads to improved accuracy and more robust insights compared to processing smaller, localized datasets. By thoroughly examining data points of the subject over time, intricate seizure patterns and individual parameters may be identified, thus enhancing accuracy and prediction rates. Additionally, this data and insights are provided under the supervision of medical professionals, potentially located in a central hub, in real-time, ensuring initial system guidance and subsequent minimal oversight. This configuration empowers unskilled caregivers or subjects to administer care.

[0066] Additionally, graphical representations are generated to visually depict the state of brain health over extended periods, allowing for the identification of trends and patterns. These representations provide a comprehensive overview of brain activity, offering insights into long- term neurological well-being. By analyzing these graphs, healthcare professionals can discern fluctuations and changes in brain health, facilitating the detection of potential issues and the monitoring of treatment efficacy over time.

[0067] Moreover, continuous monitoring of electroencephalography (EEG) signals over extended periods and storing the measured EEG signal in the repository may reveal correlations between specific EEG patterns and a range of physiological parameters that have a direct impact on brainactivity and cognitive function. This increases the possibility to estimate and track levels of parameters such as blood glucose, cortisol, sleep, and neurotransmitters, for example, dopamine, serotonin, and norepinephrine.

[0068] Further, upon determining that the plurality of biomarkers associated with the at least one anomaly is present in the EEG signal, at step 412, the predicting module 328 may be configured to predict the occurrence of the physiological disorder, for example, seizure, sleep anomalies, or cognitive decline and thus, procedures to overcome impact of the physiological disorder may be initiated as explained with reference to Figure 4.

[0069] Further, upon determining that the plurality of biomarkers associated with the at least one anomaly is absent in the EEG signal, then at step 424, the detecting module 312 may be configured to detect the plurality of other anomalies in the EEG signal, and the cycle persists.

[0070] Additionally, at step 408, if the detecting module 312 detects the absence of the anomaly in the EEG signal, the process returns to the step 404, where the biological acquisition unit receives the EEG signal from each electrode.

[0071] Particularly, the recorded EEG signals may be segmented into multiple sub-band components through various mechanisms, for example, Fourier series transform or wavelet decomposition. Further, each sub-band undergoes analysis to identify distinct neural patterns and spike occurrences which may be important for detecting deviations in an activity of the brain indicative of abnormal neurological conditions. Further, each sub-band may be analysed through a plurality of processes, for example, linear method process utilizing frequency domain analysis of the EEG signals and parametric models based on multivariate spectrum estimation.

[0072] Additionally, nonlinear signal theory techniques, for example, correlation density, largest Lyapunov exponent, and dynamic similarity index, may be utilized to address changes in dynamics of the EEG signal during a preictal period of physiological disorder. These techniques may be leveraged to identify features present only during or immediately before seizure onset. By scrutinizing sub-bands for these specific features, the system 114 enhances seizure detection and prediction accuracy and reliability.

[0073] The determination, and prediction of the physiological disorder, as well as the classification of the physiological disorder into different types based on intensity, frequency ofoccurrence, and time duration, may be achieved by the controller 104. The controller 104 may achieve the objective by using the at least one AI model, i.e., a local AI model, the foundational AI model being the transformer-based machine learning model and other models applied to data generated from the EEG signals as discussed below.

[0074] In an embodiment, the foundational AI model having a transformer architecture may be used to decrease challenges of scarcity and heterogeneity in the data of EEG signals for determining physiological disorder, for example, seizure, predicting the occurrence of the seizures, and classifying seizures, and other related disorders, for example, epilepsy, alzheimer's, parkinson's, and traumatic brain injuries. The foundational AI model may classify raw data, for example, raw ear data of the EEG signal into different categories by eliminating the need for extensive feature extraction thereby reducing computational complexities.

[0075] Temporal and spatial features may be effectively extracted from the EEG signals that may be crucial for accurate classification, allowing the local AI model and the foundational AI model to learn the relevant patterns and characteristics from the raw data from the EEG signals. The system 114 captures long-range dependencies and models complex relationships and is well-suited to determine seizures, predict the occurrence of seizures, and classify seizures based on individualized data patterns and generalized data analysis.

[0076] Further, the training and evaluation of the foundational AI model may be explained as below:

[0077] The foundational AI model may be trained on labelled datasets of EEG signals, employing cross-entropy loss to minimize the cross-entropy loss function. Optimization processes like Adam (adaptive moment estimation) or Radam (type of stochastic optimiser) are employed to iteratively update the model parameters, aiming to minimize the loss and enhance prediction accuracy. The foundational AI model's performance is comprehensively evaluated on an independent test set using metrics such as accuracy, sensitivity, specificity, and positive predictive value, which quantify its ability to correctly classify seizure and non-seizure events. Regularization techniques, for example, dropout are also incorporated during training to prevent overfitting and improve generalization capabilities.

[0078] The foundational AI model may be then fine-tuned for specific tasks, such as classification or prediction of physiological disorder. During fine-tuning, the foundational AI model adapts one or more parameters to generate effective features for classification based on individual neural patterns. This involves adjusting the weights of the foundational AI model based on labelled datasets of EEG signals, where each signal is labelled as either seizure or non-seizure.

[0079] In another embodiment, diffractive neural networks or optical neural network may be used for determining seizures, predicting the occurrence of seizures, and classifying seizures by recognizing specific patterns of the biosignal with minimal latency compared to traditional neural networks. The inherent wavelength-scale parallelism facilitated by optical diffraction allows for the simultaneous processing of massive datasets contained within multi-channel biosignals. Thereby providing an energy-efficient solution for handling the computational complexity of extracting relevant spatio-temporal features from such high-dimensional data.

[0080] In yet another embodiment, the machine learning (ML) models may be configured to process electroencephalogram (EEG) signals through a multi-stage analytical pipeline to detect recurrent patterns, anomalies, and transient biomarkers associated with physiological and neurological disorders. The system 114 employs supervised and unsupervised learning techniques, including but not limited to convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and transformer models. The system 114 employs supervised and unsupervised learning techniques based on attention mechanisms and anomaly detection algorithms such as isolation forests or autoencoders, to analyze both time-domain and frequency-domain representations of EEG data. Feature extraction techniques, such as wavelet transforms, power spectral density (PSD) analysis, and temporal-spatial decomposition, are applied to isolate discriminative patterns indicative of epileptiform discharges, seizure precursors, or other aberrant neural activity.

[0081] The ML model is trained on labelled datasets comprising EEG recordings from both healthy subjects and patients with diagnosed neurological conditions, enabling the classification of signal deviations with high specificity and sensitivity. Real-time processing is facilitated through sliding-window analysis and adaptive thresholding, allowing for continuous monitoring and early detection of pathological events. Additionally, the system 114 may integrate reinforcement learning to refine predictive accuracy over time by incorporating feedback from clinical outcomes. By correlating extracted features with known neurological biomarkers, themodel not only predicts the onset of disorders such as epilepsy but also stratifies risk levels and identifies subclinical manifestations that may precede overt symptomatology. This approach enhances diagnostic precision, supports proactive intervention, and enables longitudinal tracking of disease progression.

[0082] Further, long-term monitoring of brain activity offers valuable insights into brain health, mental well-being, and patterns of physiological disorders, surpassing the limitations of short-term monitoring and analysis. Moreover, the longitudinal nature of the monitoring process allows for the identification of trends, patterns, and correlations that may not be apparent in short-term observations, paving the way for advancements in the field of neurology and personalized healthcare.

[0083] Over time, the device 100 continues to periodically record the subject’s EEG at regular intervals and compares the current data to the previous measurements to identify any changes or deviations. Based on these insights, the system 114 may then dynamically adjust the frequency of EEG monitoring, i.e., increase the frequency of the EEG monitoring, if the data shows significant variations. Further, the frequency may decrease if the data remains stable, to provide personalized and informed assessments of the subject’s neurological health and enable early detection of potential issues.

[0084] Figure 5 illustrates a flowchart depicting a process to overcome the impact of the physiological disorder in the subject 112 upon the prediction of the occurrence of the physiological disorder, in accordance with an embodiment of the present disclosure.

[0085] Referring to Figures 4 and 5, if the determining module 314 determines that the plurality of biomarkers associated with the at least one anomaly is present in the EEG signals, then at step 502, the storing module 313 may be configured to analyze the seizure / physiological disorder and thereafter record / store the analysis. Further, at step 504, generating module 332 may be configured to generate the physiological disorder alert for a user equipment (UE) 116 based on the prediction and the recorded analysis. The physiological alert may indicate a notification for the subject thereby predicting the occurrence of the physiological disorder. The physiological alert may include, but is not limited to, an audio notification, haptic feedback and mobile alerts to notify the subject or caregivers of the predicted occurrence.

[0086] . In one embodiment, when the device 100 may be worn in the ear by the subjects, the generating module 332 may be configured to warn the subjects by providing the alert through audio messages. The generating module 332 may provide real-time alerts to the subjects through earphones about an impending occurrence of the seizure and offer guidance on appropriate actions to take. Additionally, the generating module 332 may play calming music to help calm the subject 112 during the occurrence of the seizure.

[0087] When the device 100 may be the band, then, at step 506, the generating module 332 may be configured to inflate each inflating member of the band around the subject's head to help prevent injury during the seizure. Simultaneously, the generating module 332 may deploy the expandable shield or covering to further safeguard the head of the individual and upper body during the seizure.

[0088] Further, the generating module 332 may also be configured to perform different operations as provided below:

[0089] Inform Caregivers: The generating module 332 may promptly notify the subject’s designated caregivers, such as family members or medical professionals, about the impending seizure. The mesh network may be set up to instantly transmit the physiological disorder, i.e, seizure warning to any nearby designated individuals who have the compatible app or are within range of the mesh network. The decentralized nature of the mesh enhances reliability, bypassing traditional cellular or Wi-Fi infrastructure to deliver alerts directly to nearby caregivers.

[0090] Provide Guidance to the subject: The generating module 332 may assist the subject 112 in remaining calm in a safe and secure location, and to take any necessary precautions until the seizure has passed.

[0091] Authorize Access to Records: In the event of an emergency, the generating module 332 may alert and allow authorized / designated individuals, such as first responders, to access the subject’s personal and medical records to ensure appropriate care.

[0092] At step 508, the determining module 314 may be configured to classify a type of the physiological disorder, for example, the seizure, where at step 510, each electrode may be adapted to be utilized in tandem to triangulate and locate the epicentre of the seizure within the brain. This facilitates the accurate classification of the physiological disorder / seizures as either focal orgeneralized, enabling more precise interventions and personalized treatment approaches. EMG data obtained alongside EEG data, enables the accurate characterization of seizure into various type, for example, tonic-clonic, absence seizure, etc.

[0093] At step 512, after classifying the seizure, the determining module 314 may be configured to determine the intensity of the physiological disorder, for example, seizure.

[0094] The determining module 314 may be configured to determine the intensity of the physiological disorder using a multidimensional Seizure Intensity Score (S). This standardized metric integrates electrophysiological, autonomic, clinical, and recovery-based parameters to quantify severity, enabling data-driven treatment protocols. The SIS may be computed by equation 1 as provided below: Seizure Intensity Score (SIS) = (EEG_Score × W1) + (ANS_Score × W2) + (Symptom_Score × W3) + (Recovery_Score × W4) + Bias………..(1) weighting factors derived from multivariate analysis of clinical datasets, and Bias accounts for population-specific baselines.

[0095] The EEG_Score quantifies electrographic severity through amplitude, duration, frequency, spatiotemporal spread, and the presence of high-frequency oscillations (HFOs) such as fast ripples (200–500 Hz).

[0096] The ANS_Score evaluates autonomic disruption using heart rate variability (HRV) decline, sustained tachycardia / bradycardia, respiratory pauses, and fluctuations in skin conductance.

[0097] The Symptom_Score classifies clinical manifestations such as motor symptoms (, and injury risk

[0098] The Recovery_Score assesses postictal burden through recovery time. Weighting factors (W1–W4 ) may be calibrated using regression models trained on clinically validated datasets to align with established epilepsy severity scales.

[0099] The total SIS may be able to stratify seizures into tiers such as Mild, triggering caregiver alerts and self-management strategies; Moderate, initiating rescue medication, auto-injection andGPS-based emergency alerts; or Severe, activating protective mechanisms (e.g., wearable air cushion inflation) and notifying emergency services.

[0100] For onset tracking, the system 114 may leverage prodromal biomarkers, such as HRV decline exceeding 20% combined with elevated beta power (16–24 Hz), to issue pre-seizure alerts 10–30 minutes pre-ictal. A Dynamic Risk Score, may be computed by equation 2: Risk Score=(HFO Density×EEG Entropy) / HRV………(2) This may further stratify risk into thresholds (for example >8.0 = high risk, 4.0–8.0 = moderate, <4.0 = low), guiding preemptive interventions.

[0101] At step 514, the system 114 incorporates multimodal feedback and stimulation capabilities, for example, haptic, auditory, and thermal modalities to deliver interventional stimulations to potentially reduce seizure intensity or prevent seizure onset upon detection or prediction of an impending seizure event.

[0102] Particularly, upon recording the person's brain activity and predicting the occurrence of the physiological disorder, at step 514, the determining module 314 may be configured to transmit a signal to an interventional stimulation unit. The determining module 314 may be configured to determine, dynamically, at least one pre-stored task for the subject 112 based on the EEG signal. Thereafter, the transmitting module 322 may be configured to transmit the activation signal to the interventional stimulation unit to activate one of the electrical stimulation and magnetic stimulation, based on the determined at least one task.

[0103] In such an embodiment, the transmitting module 322 may implement a proactive therapeutic approach by transmitting the activation signal to the interventional stimulation unit for delivering targeted, time-bound pulses of electrical or magnetic stimulation to modulate abnormal neural activity upon early detection, thereby aiming to prevent the progression into a full seizure episode. Stimulation modalities that may be employed include Transcranial Magnetic Stimulation (TMS), Transcranial Direct Current Stimulation (tDCS), Transcranial Alternating Current Stimulation (tACS), and Vagus Nerve Stimulation (VNS). In some embodiments, TMS delivers focused magnetic fields to specific cortical regions, inducing localized electric currents to disrupt pathological neuronal synchronization associated with seizure initiation. tDCS may apply low- intensity direct currents across the scalp to modulate cortical excitability by either enhancinginhibitory pathways or attenuating hyperexcitable networks. VNS may be administered through implanted or external devices to stimulate the vagus nerve, thereby exerting modulatory effects across broad brain areas implicated in seizure propagation.

[0104] The stimulation parameters, including but not limited to frequency (e.g., approximately 1 Hz for inhibitory TMS), current intensity, pulse width, and stimulation site, may be dynamically adjusted in real time based on the characterization of the detected abnormal activity. Safety mechanisms such as continuous impedance monitoring, stimulation envelope tracking, and automated shutoff thresholds may be incorporated to ensure that stimulation remains within predefined clinical safety margins.

[0105] In addition to delivering stimulation, the system 114 may continuously acquire and analyze neurophysiological and autonomic signals during and after stimulation intervention. Biomarkers such as interictal spike frequency, high-frequency oscillation (HFO) density, spectral band shifts, and heart rate variability may be monitored to assess the brain's response to the stimulation intervention. Further, analytical algorithms, including machine learning models, may process these multimodal datasets to evaluate the therapeutic efficacy of the delivered stimulation. Based on this analysis, the system 114 may automatically refine stimulation parameters, such as adjusting the current intensity of tDCS, modifying the pulse frequency of TMS, or altering the duty cycle of VNS, to enhance treatment outcomes.

[0106] The system 114 thus forms a closed-loop framework that adapts therapy based on real- time physiological feedback, enabling personalized, responsive intervention strategies tailored to the individual’s evolving neurophysiological state. Over time, the continuous adaptation of stimulation parameters may improve seizure control, particularly in individuals with pharmacoresistant epilepsy or those inadequately managed by conventional therapies.

[0107] Additionally, the device 100 and user application may function as an epilepsy screening tool, including a photoparoxysmal response (PPR) assessment feature. This configuration generates visual stimuli like flashing lights or patterns through the stimulus interlacer, gradually changing in frequency and intensity within safe limits to assess the subject's sensitivity for PPR induction. This information, alongside quantified response characteristics, aids in diagnosing and evaluating photosensitive epilepsy or related conditions. Similar techniques may be employed to assess risk to audio or startle evoked seizure episodes by studying the evoked responses as well.

[0108] In another embodiment, the system 114 utilizes the EEG data to tailor multimedia content for subjects with or at risk of seizures. The system 114 carefully regulates visual parameters such as flickering rates, brightness variations, contrast, geometric patterns, and colors present in the content, ensuring that the visual stimuli do not induce or exacerbate seizure episodes.

[0109] After determining the intensity of the seizure, at step 515, the determining module 314 may be configured to determine whether the occurrence of the seizure in the subject has come to an end. If the determining module 314 determines that the occurrence of the seizure in the subject 112 has come to an end, then at step 516, the analysed data of the seizure may be logged / stored and the generating module 332 may stop providing the warning / alert messages to the subjects. Further, at step 518, the logged analyzed data may be provided to the local AI model for future operations. Further, if the determining module 212 determines that the occurrence of the seizure in the individual has not ended, then again the procedure returns to step 502.

[0110] Figure 6 illustrates a flowchart depicting an assessment of the subject’s cognitive function, upon the prediction of the occurrence of the physiological disorder, in accordance with an embodiment of the present disclosure. At step 602, the Biosignal Acquisition Module / Biosignal Acquisition Unit measures and records the person's brain activity. This data is then transmitted to the controller 104 for further processing as shown at step 604, upon the prediction of the occurrence of the physiological disorder. The controller 104, using the edge computing technique, performs advanced computations on the acquired data. Additionally, the personalized profile setup module generates customized assessment tasks tailored to each individual based on their unique brain wave patterns. These personalized tasks are then forwarded to the Stimulus Interlacer as shown at step 606, which merges the content to be presented with imperceptible frequencies, flickers or gratings and provides to a cognitive assessment and stimulus delivery unit as shown at step 618, and finally to the subject 112.

[0111] Further, the system 114 leverages AI-enabled insights generated from the acquired data. These insights are utilized by three distinct modules: Cognitive Function Analysis as shown at step 608, Brain Activity Analysis as shown at step 610, and Diagnostic Stimulus Response Analysis as shown at step 612. Each module performs specialized analyses to evaluate different aspects of the subject’s cognitive function and brain activity. The outputs obtained from these analyses are then fed into the Local AI Model, as shown at step 614, that is in communication with the Foundational AI Model, as shown at step 616, where advanced algorithms evaluate the dataand generate scores. These scores are subsequently displayed through the Cognitive Function Analyzer as shown at step 620, providing insights into the likelihood of cognitive function deterioration and quantifying the extent of any cognitive decline.

[0112] In an embodiment, particularly, imperceptible frequency modulations of varying frequencies and patterns are embedded across different durations into auditory or visual stimuli to elicit neural oscillations that are designed to modulate and normalize abnormal neuronal activity upon anomaly detection. The visual stimuli employed may take several forms, including presenting patterns featuring varied spatial frequencies, orientations, and contrasts, as well as employing flickering or temporally modulated visual content. Additionally, specific temporal or spatial frequencies may be embedded within visual content. It is promptly presented to the subject through a screen on their mobile device, personal computer, or wearable device. These visual cues are designed to appear instantly, aiming to capture the user's attention during the critical anomaly period. The neurophysiological effects decrease pathological neural activity associated with anomalies, for example, seizures through the entrainment or desynchronization of abnormal oscillatory patterns across distributed neuronal populations.

[0113] In an embodiment, presentation of content embedded with these crafted frequencies may modulate neurophysiological mechanisms and neural connectivity across multiple scales, spanning molecular, cellular, and network levels within the brain. This process thereby promotes long-term neural adaptations that fortify the brain's resilience against seizure susceptibility.

[0114] Now, referring to step 608, cognitive decline analysis:

[0115] In an embodiment, the extracting module 330 may be configured to extract response features based on generating targeted brainwave oscillations. The target brainwave oscillations may be slow gamma rhythms of the subject 112 via at least one of a Rapid Invisible Frequency Tagging (RIFT) and counter-phased visual gratings. Further, the response feature indicates spectral power.

[0116] The determining module 314 may be configured to determine, simultaneously, deviations in brain wave power, especially gamma-band power of the subject 112. The determining module 314 may determine deviations in the brain wave power and evoke and record P300 event-related potentials and visual evoked potentials (VEPs) during structured cognitive tasks.

[0117] The generating module 332 may be configured to generate, simultaneously, a cognitive capacity score based on task performance metrics, including reaction time and memory recall accuracy.

[0118] Particularly, brain waves may be used to assess cognitive decline in the subjects while also fostering resilience. This may be achieved by examining disruptions in brainwave oscillations, including but not limited to gamma oscillations, particularly in brain regions crucial for memory formation and retrieval, as gamma brain rhythms are closely associated with higher cognitive functions such as attention, memory, and learning.

[0119] The disclosed system 114 uses audiovisual stimulation protocols, including Rapid Invisible Frequency Tagging (RIFT), static or counter-phasing gratings, and auditory inputs such as frequency-modulated tones or binaural beats and speech stimuli including natural speech, to evoke and measure gamma oscillations. Visual stimuli, such as full-screen cartesian gratings on a display unit of the UE with specified spatial frequencies (1–4 cycles / degree) and orientations (0°, 45°, 90°, 135° ,etc. ), may be designed to induce slow (20–34 Hz) and fast (36–66 Hz) gamma activity, while counter-phasing gratings at 16 Hz may be employed to elicit 32 Hz Visual Evoked Potentials (VEPs), i.e., Steady-State Visual Evoked Potentials (SSVEPs) for enhanced sensitivity to neural decline. Further, other evoked potentials such as transient visual evoked potential, pattern reversal visual evoked potential, pattern onset visual evoked potential and flash visual evoked potential may be used alternatively in place of steady state visual evoked potentials to get more precision.

[0120] Neural signals may be non-invasively acquired using the device 100, including ear-EEG earphones, headbands, or scalp electrodes, enabling real-world and longitudinal monitoring. Advanced artifact rejection techniques such as bipolar referencing, multi-taper spectral analysis, and adaptive filtering may be implemented to suppress noise from microsaccades, ocular movements, and muscular activity, ensuring signal fidelity.

[0121] Quantitative analysis modules track gamma-band power, shifts in center frequency, and SSVEP amplitude and coherence. Declines in fast gamma power, reductions in center frequency, or SSVEP instability serve as potential early markers of cognitive impairment. Behavioral performance metrics, including reaction time, task accuracy, memory recall, and engagementduring real-world tasks (e.g., games, quizzes, puzzles, mathematical operations, recollection tests, interviews), are integrated with neurophysiological data to strengthen diagnostic precision.

[0122] Further, generating module 332 along with a cognitive capacity scoring engine consolidates these neural and behavioral metrics into composite indices, generating the cognitive capacity score, facilitating personalized baselines and longitudinal tracking of cognitive health. Significant deviations, such as greater than 20% reductions in fast gamma power relative to baseline, are flagged as indicators of potential cognitive decline. Early detection enables the timely implementation of interventions, including personalized gamma-frequency stimulation therapies aimed at enhancing cognitive resilience and slowing the progression of conditions such as Alzheimer’s, dementia, and related brain ailments.

[0123] Furthermore, in an exemplary implementation, the system may be configured for early diagnosis and monitoring of a broader range of neuropsychiatric and neurodegenerative conditions. By incorporating structured audiovisual stimuli, passive biosignal tracking, and machine learning-based pattern analysis, the system 114 supports the identification and tracking of disorders such as depression, anxiety, schizophrenia, bipolar disorder, and Parkinson’s disease at earlier stages than conventional diagnostic methodologies.

[0124] Further, in an embodiment, upon predicting the occurrence of the physiological disorder, the controller 104 may perform additionally below mentioned operations:

[0125] In an embodiment, the determining module 314 may be configured to trigger at least one of protective measures or neurostimulation based on a quantification derived from EEG amplitude, autonomic disruption, and symptom severity. Further, the determining module 314 may be configured to synchronize data, securely, with encrypted cloud-based platforms or decentralized storage networks, including blockchain-enabled architectures utilizing smart contracts. The determining module 314 may synchronize data for tokenised access control that enables enhanced insights, remote clinician access, or collaborative research.

[0126] Spike-and-wave discharge detection: The detecting module 310 employs advanced computational techniques to detect the abrupt onset and termination of 3-Hertz spike-and-wave discharges on electroencephalogram (EEG), a characteristic pattern associated with absence seizures. Furthermore, the determining module 314 further determines whether the subject 112 isexperiencing absence seizures or if any detected EEG anomalies may indicate a likelihood of developing them.

[0127] Personalized Dose-Response Modeling: In an embodiment, upon predicting the occurrence of the physiological disorder, the correlating module 334 may be configured to correlate the dosage of neuromodulatory drug with the plurality of biomarkers based on a plurality of neurophysiological metrics. The correlating module 334 may be configured to correlate the dosage based on the machine learning model. The plurality of neurophysiological metrics may include, but is not limited to, the plurality of neurophysiological metrics comprises P300 amplitude attenuation value, theta-to-alpha power ratio shifts value, or reaction time variability value. Further, the plurality of biomarkers may include, but is not limited to, HFO density and spectral band shifts. Thereafter, the generating module 332 may be configured to generate longitudinal dose-response profiles visualized via probabilistic heatmaps having overlay of pharmacokinetics for clinician review. Particularly, this integration links physiological signals with antiepileptic drug (AED) pharmacokinetics to build personalized dose-response profiles. Techniques, for example, Gaussian Process Regression quantify high-frequency oscillation (HFO) density changes relative to dosage adjustments, while convolutional neural networks classify EEG patterns into drug-responsive or drug-refractory states, such as identifying temporal delta surges indicative of oversedation. The generating module 332 then produces longitudinal dose-response profiles visualized through probabilistic heatmaps overlaid with pharmacokinetic data. These insights, paired with a dynamic dosing interface, enable clinicians to simulate and evaluate regimen adjustments prior to clinical implementation.

[0128] Additionally, in an embodiment, the adjusting module 336 may be configured to adjust the dosage or intensity of a neurostimulation technique, based on determining an effect of neurostimulation techniques on the subject. The adjusting module 336 may be configured to adjust the dosage by creating a personalized dose response profile (DRP). The neurostimulation techniques may include the transcranial direct current stimulation (tDCS), the transcranial Alternating current stimulation (tACS) or the vagus nerve stimulation (VNS). Additionally, the delivering module 338 may be configured to deliver, simultaneously, interventional stimulations, dynamically, to the subject (as shown at step 514) upon detecting abnormal neural activity, heart rate variability (HRV) and EEG coherence shifts of the subject.

[0129] In an embodiment, the correlating module 334 may be configured to correlate the occurrence of the physiological disorder with environmental or stress triggers using federated learning. The generating module 332 may be configured to generate a signal corresponding to a personalized lifestyle recommendation via the device 100 and the UE 116 based on real-time biofeedback loops.

[0130] The method 700 includes a series of operations shown at step 702 through step 708 of Figure 7. The method 700 may be performed by the system 114 in conjunction with modules 308, the details of which are explained in conjunction with Figures 1 to 6, and the same are not repeated here for the sake of brevity in the present disclosure. The method 700 begins at step 702.

[0131] At step 702, the method 700 includes detecting the plurality of anomalies present in the EEG signal based on comparing the EEG signal with the updated output from the predetermined longitudinal EEG trend corresponding to daily activities and sleep of the subject 112 using at least one of the Artificial Intelligence model.

[0132] For detecting the plurality of anomalies, the method 700 includes assessing photosensitive physiological disorder via the controlled visual exposure and audio sensitive physiological disorder via controlled audio exposure while monitoring Visual Evoked Potential (VEP) biomarkers as well as Auditory Evoked potentials (AEP). The method 700 includes detecting the plurality of anomalies based on integrating subliminal frequency modulations into the audiovisual stimuli to modulate neural oscillations, thereby inducing adaptive neuroplasticity through neurotherapeutic conditioning.

[0133] Prior to detecting the plurality of anomalies, the method 700 includes passively recording and storing longitudinal EEG data via the at least one wearable device 100 during daily activities and sleep. The method 700 includes generating the predetermined longitudinal EEG trend based on preprocessing the longitudinal EEG data based on the predetermined techniques such as wavelet decomposition, and phase-space reconstruction.

[0134] At step 704, the method 700 includes determining the presence of the plurality of biomarkers associated with at least one anomaly in the EEG signal, including Interictal Epileptiform Discharges (IEDs) and High-Frequency Oscillations (HFOs), using the transformer- based machine learning model, upon the detection.

[0135] At step 706, the method 700 includes predicting the occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with at least one anomaly is present in the EEG signal.

[0136] At step 708, the method 700 includes generating the physiological disorder alert for a user equipment (UE) 116, where the physiological disorder alert indicates a notification for the subject thereby predicting the occurrence of the physiological disorder. The physiological disorder alert may include, but is not limited to, the audio notifications, haptic feedback, and mobile alerts to notify the subject (112) or caregivers of the predicted occurrence.

[0137] The method 700 includes triggering at least one of protective measures or neurostimulation based on a quantification derived from EEG amplitude, autonomic disruption, and symptom severity. The method 700 includes synchronizing data, securely, with encrypted cloud-based platforms or decentralized storage networks, including blockchain-enabled architectures utilizing smart contracts for tokenised access control that enable enhanced insights, remote clinician access or collaborative research.

[0138] Further, in an embodiment, the use cases of the system 114 are provided as below:

[0139] Engagement-Driven Validation: Cognitive tasks such as n-back working memory tests and visuomotor coordination games are integrated into the system 114 to assess drug-modulated brain responses. Key features such as P300 amplitude attenuation and shifts in theta-to-alpha power ratios are tracked alongside task performance metrics as reaction time and accuracy. Composite adherence scores are generated from electrophysiological and behavioral data, with gamified rewards, such as milestone unlocks, encouraging consistent engagement and compliance.

[0140] Artificial Intelligence-Powered Dose Optimization: A reinforcement learning agent dynamically recommends dose adjustments within safe operating limits (typically ±10% of baseline). Persistent pathological biomarkers, such as high-frequency oscillations above fifteen events per minute or significant increases in delta- the optimization loop. Federated learning architectures enable continuous refinement of models across multiple institutions while preserving patient privacy.

[0141] Clinical Deployment: Quarterly 24-hour ambulatory EEG recordings are used to validate model predictions against expert clinician annotations, with semi-supervised retraining deployedto correct any discrepancies. Integration with electronic health record (EHR) systems streamlines data capture and prescription workflows, replacing traditional trial-and-error dose titration with biomarker-driven precision optimization. The system 114 aims to significantly reduce the time to therapeutic stabilization while improving clinical outcomes in epilepsy management.

[0142] Closed-Loop Monitoring System: The system 114 integrates a closed-loop neurofeedback platform that continuously monitors the efficacy of antiepileptic drug (AED) therapy by analyzing brainwave biomarkers such as interictal spikes, high-frequency oscillations (80–500 Hz), and gamma-band power (30–80 Hz), alongside autonomic signals including heart rate variability and electrodermal activity. These multimodal signals are time-synchronized with medication dosing events and seizure logs through precise time-locked fusion. Adaptive thresholding isolates pathological activity, for example, identifying high-frequency oscillations exceeding two standard deviations above baseline, while wavelet transforms extract spectral features sensitive to pharmacological intervention.

[0143] The system 114 thus forms a closed-loop framework that adapts therapy based on real- time physiological feedback, enabling personalized, responsive intervention strategies tailored to the individual’s evolving neurophysiological state. Over time, the continuous adaptation of stimulation parameters may improve seizure control, particularly in individuals with pharmacoresistant epilepsy or those inadequately managed by conventional therapies.

[0144] Lifestyle and Environmental Factors: Potential triggers for physiological disorders for subjects are identified by correlating occurrences of the physiological disorder with lifestyle and environmental data, including factors like stress, sleep patterns, or specific activities. The system 114 offers behavioural recommendations, leveraging insights from data analysis to suggest lifestyle adjustments for effective epilepsy management. Combining data-driven insights with tailored lifestyle guidance delivered through earphones, aims to enhance seizure control, improve quality of life, and promote overall health outcomes.

[0145] As would be gathered, the system 114 provides an efficient solution for predicting and managing physiological disorder, for example, seizures resulting in epilepsy. By seamlessly integrating advanced biosensing, computational techniques, and intelligent algorithms, the system 114 offers comprehensive real-time brain activity tracking, anomaly detection, and personalized seizure intervention. The unobtrusive form factor enables long-term data collection and analysis,unveiling insights into neurological health previously unattainable. Sophisticated machine learning models accurately determine, predict and classify seizures, enabling timely interventions. Decentralized data storage enhances privacy and security, while edge computing optimizes efficiency. The system 114 improves the quality of life for subjects with neurological conditions by providing real-time alerts, location tracking, protective mechanisms, and emergency integration. The system 114 quantifies seizure intensity, offers neurofeedback on medication effectiveness, and paves the way for future advancements in personalized healthcare, precision medicine, and novel therapeutic interventions.

[0146] The system 114 also provides the following advantages:

[0147] Physiological Disorder Logging and Documentation: The system 114 enables subjects to manually log physiological disorder events through a mobile app or device interface, capturing details like duration, severity, aura, triggers, and postictal symptoms via voice or text input. This maintains a comprehensive physiological disorder diary, allowing a review of past events with associated biosignal / EEG signal data. Detailed reports summarizing physiological disorder frequency, duration, characteristics over time, and subject well-being data are generated to facilitate discussions with healthcare providers and support medication dosage adjustments.

[0148] Communicate with Nearby Devices: The system 114 may employ low-power radio waves within the 2.400 GHz to 2.483.5 GHz frequency band to establish communication with nearby devices, like Bluetooth-enabled tags or trackers. These devices can subsequently notify authorized individuals in the subject's proximity about an ongoing seizure event, facilitating potential assistance and support. This interconnected network among devices can facilitate notifications and alerts even in regions with limited network connectivity.

[0149] While specific language has been used to describe the present disclosure, any limitations arising on account thereto, are not intended. As would be apparent to a person in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein. The drawings and the foregoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment.

Claims

We Claim:

1. A physiological disorder alert system (114) for alerting a subject (112), the system (114) comprising: a plurality of biosensor electrodes (102) in contact with at least one of a scalp and an ear canal of the subject (112) through at least one of elastic members (107) or clip-based structures (110) of at least one wearable device (100) thereby establishing required skin contact with the subject (112) and configured to measure at least one physiological parameter of the subject (112), wherein the at least one physiological parameter is Electroencephalogram (EEG) signal; and a controller (104) in communication with the plurality of biosensor electrodes (102), the controller (104) is configured to: detect a plurality of anomalies present in the EEG signal based on comparing the EEG signal with an updated output from a predetermined longitudinal EEG trend corresponding to daily activities and sleep of the subject (112) using at least one of Artificial Intelligence model; determine a presence of a plurality of biomarkers associated with at least one anomaly in the EEG signal, comprising Interictal Epileptiform Discharges (IEDs) and High-Frequency Oscillations (HFOs), using a machine learning model, upon the detection; predict an occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with the at least one anomaly is present in the EEG signal; and generate the physiological disorder alert for a user equipment (UE) (116), wherein the physiological disorder alert indicates a notification for the subject (112) thereby predicting the occurrence of the physiological disorder and comprises an audio notifications, haptic feedback, and mobile alerts to notify the subject (112) or caregivers of the predicted occurrence.

2. The system (114) as claimed in claim 1, wherein to determine the presence of the plurality of biomarkers using the machine learning model, the controller (104) is configured to: segregate the EEG signal into fixed-length patches, wherein the patches indicate overlapping time windows; convert, linearly, each patch into a token embedding; supplement, converted patch, with predefined positional encodings to enable a large context window spanning multiple seconds of neural activity; transmits, a complete sequence of token embeddings through a stack of transformer- encoder layers, wherein each transformer-encoder layer comprises multi-head self-attention and feed-forward sublayers to capture long-range spatiotemporal dependencies, indicating correlations across sub-band features such as ripples (80–200 Hz) and fast ripples (200–500 Hz); prepends a classification token to the complete transmitted sequence; aggregates the encoded information via the classification token; and determine the presence of biomarkers based on the aggregated encoded information through layer normalization and a softmax output layer, pretrained on labelled datasets.

3. The system (114) as claimed in claim 1, comprising an adaptive protective member (106), wherein the adapted protective member comprises: at least one of a plurality of inflation members (106a) and expandable shields adapted to receive the plurality of biosensor electrodes (102) and inflate to distribute impact forces, upon predicting the occurrence of the physiological disorder, wherein the at least one of the plurality of inflation members (106a) and the expandable shields operates in conjunction with structures that are made up of anisotropic materials, non-Newtonian fluids, materials with negative Poisson’s Ratio.

4. The system (114) as claimed in claim 1, comprising a plurality of sensors (108) coupled to at least one of the plurality of biosensor electrodes (102) and a biosensing insight module of at least one wearable device (100) to monitor at least one of the plurality of anomalies and incoming motion data to dynamically modulate the adaptive protective member to mitigate injury.

5. The system (114) as claimed in claim 1, wherein upon predicting the occurrence of the physiological disorder, the controller (104) is configured to: determine, dynamically, at least one pre-stored task for the subject (112) based on the EEG signal; and transmit, to an interventional stimulation unit, an activation signal to activate one of an electrical stimulation and magnetic stimulation, based on the determined at least one task.

6. The system (114) as claimed in claim 1, wherein upon predicting the occurrence of the physiological disorder, the controller (104) is configured to: extract response features based on generating targeted brainwave oscillations such as slow gamma rhythms of the subject (112) via at least one of a Rapid Frequency Invisible Tagging (RIFT) and counter-phased visual gratings, wherein the response feature indicates spectral power; determine, simultaneously, deviations in brain wave power especially gamma-band power of the subject (112) as well as evoke and record P300 event-related potentials and visual evoked potentials (VEPs) during structured cognitive tasks; and generate, simultaneously, a cognitive capacity score based on task performance metrics, including reaction time and memory recall accuracy.

7. The system (114) as claimed in claim 1, wherein upon predicting the occurrence of the physiological disorder, the controller (104) is configured to: correlate dosage of neuromodulatory drug with the plurality of biomarkers based on a plurality of neurophysiological metrics , wherein the plurality of neurophysiological metrics comprises P300 amplitude attenuation value, theta-to-alpha power ratio shifts value, or reaction time variability value, and wherein the plurality of biomarkers comprises HFO density and spectral band shifts; and generate longitudinal dose-response profiles visualized via probabilistic heatmaps having overlay of pharmacokinetics for clinician review.

8. The system (114) as claimed in claim 11, wherein the controller (104) is configured to: adjust dosage or intensity of a neurostimulation technique, based on determining an effect of neurostimulation techniques on the subject (112), by creating a personalized dose response profile (DRP), wherein the neurostimulation techniques comprises a transcranialdirect current stimulation (tDCS), transcranial Alternating current stimulation (tACS) or vagus nerve stimulation (VNS); and deliver, simultaneously, interventional stimulations, dynamically, to the subject (112) upon detecting abnormal neural activity, heart rate variability (HRV) and EEG coherence shifts of the subject (112).

9. The system (114) as claimed in claim 1, wherein upon predicting the occurrence of the physiological disorder, the controller (104) is configured to: correlate the occurrence of the physiological disorder with environmental or stress triggers using federated learning; and generate a signal corresponding to a personalized lifestyle recommendations via at least one wearable device (100) and the UE (116) based on real-time biofeedback loops.

10. The system (114) as claimed in claim 1, wherein the controller (104) comprises an embedded edge computing neural network configured for low-latency processing of multi- channel biosignals, utilizing either electronic circuitry or a diffractive optical neural network architecture to enable real-time detection of the plurality of anomalies.

11. A method (700) for generating a physiological disorder alert for a subject (112), the method (700) comprising: detecting (702) a plurality of anomalies present in EEG signal based on comparing the EEG signal with an updated output from a predetermined longitudinal EEG trend corresponding to daily activities and sleep of the subject (112) using at least one of Artificial Intelligence model; determining (704) a presence of a plurality of biomarkers associated with at least one anomaly in the EEG signal, comprising Interictal Epileptiform Discharges (IEDs) and High- Frequency Oscillations (HFOs), using a machine learning model, upon the detection; predicting (706) an occurrence of the physiological disorder, upon determining that the plurality of biomarkers associated with the at least one anomaly is present in the EEG signal; and generating (708) the physiological disorder alert for a user equipment (UE) (116), wherein the physiological disorder alert indicates a notification for the subject (112) thereby predicting the occurrence of the physiological disorder and comprises an audio notifications,haptic feedback, and mobile alerts to notify the subject (112) or caregivers of the predicted occurrence.

12. The method (700) as claimed in claim 11, wherein prior to detecting the plurality of anomalies, the method (700) comprises: passively recording and storing longitudinal EEG data via at least one wearable device (100) during daily activities and sleep; and generating a predetermined longitudinal EEG trend based on preprocessing the longitudinal EEG data based on a predetermined techniques such as wavelet decomposition, phase-space reconstruction.

13. The method (700) as claimed in claim 11, wherein for detecting the plurality of anomalies present in EEG signal , the method (700) comprises: assessing photosensitive physiological disorder via controlled visual exposure and audio sensitive physiological disorder via controlled audio exposure while monitoring Visual and Auditory Evoked Potential biomarkers; and detecting the plurality of anomalies based on integrating subliminal frequency modulations into audiovisual stimuli to modulate neural oscillations, thereby inducing adaptive neuroplasticity through neurotherapeutic conditioning.

14. The method (700) as claimed in claim 11, wherein upon predicting the occurrence of the physiological disorder, the method (700) comprises: triggering at least one of protective measures or neurostimulation based on a quantification derived from EEG amplitude, autonomic disruption, and symptom severity; synchronizing data, securely, with encrypted cloud-based platforms or decentralized storage networks , including blockchain-enabled architectures utilizing smart contracts for tokenised access control that enable enhanced insights, remote clinician access or collaborative research.

Images