WO2024084429A1 - Réseau d'ido agnostique de dispositif pour la prédiction et la prise en charge de l'épilepsie et la prévention du risque de mort subite inattendue en épilepsie - Google Patents

Réseau d'ido agnostique de dispositif pour la prédiction et la prise en charge de l'épilepsie et la prévention du risque de mort subite inattendue en épilepsie Download PDF

Info

Publication number
WO2024084429A1
WO2024084429A1 PCT/IB2023/060566 IB2023060566W WO2024084429A1 WO 2024084429 A1 WO2024084429 A1 WO 2024084429A1 IB 2023060566 W IB2023060566 W IB 2023060566W WO 2024084429 A1 WO2024084429 A1 WO 2024084429A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
individual
inputs
patterns
processing module
Prior art date
Application number
PCT/IB2023/060566
Other languages
English (en)
Inventor
Rajlakshmi Borthakur
Original Assignee
Rajlakshmi Borthakur
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Rajlakshmi Borthakur filed Critical Rajlakshmi Borthakur
Publication of WO2024084429A1 publication Critical patent/WO2024084429A1/fr

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25—Bioelectric electrodes therefor
    • A61B5/279—Bioelectric electrodes therefor specially adapted for particular uses
    • A61B5/28—Bioelectric electrodes therefor specially adapted for particular uses for electrocardiography [ECG]
    • A61B5/282—Holders for multiple electrodes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/40—Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
    • A61B5/4094—Diagnosing or monitoring seizure diseases, e.g. epilepsy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/683—Means for maintaining contact with the body
    • A61B5/6832—Means for maintaining contact with the body using adhesives
    • A61B5/6833—Adhesive patches
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00—Economic sectors
    • G16Y10/60—Healthcare; Welfare

Definitions

  • a DEVICE AGNOSTIC IoT NETWORK FOR PREDICTION AND MANAGEMENT OF EPILEPSY AND PREVENTION OF RISK OF SUDEP TECHNICAL FIELD The present subject matter relates, in general, to a wearable device, and, in particular, to an Internet of Things (IoT)-powered device agnostic platform for the prediction and management of chronic disorders, such as epilepsy, and for the prevention of catastrophic health events, including risk of SUDEP.
  • IoT Internet of Things
  • BACKGROUND Epilepsy is a neurological disorder that affects approximately 50 million people across the world. It has become a widespread phenomenon, and despite the newer anti-epilepsy drugs, People with Epilepsy (PWE) are at an increased risk of suffering a sudden and unexpected death (SUDEP).
  • SUDEP ranks second only to stroke and is ‘a sudden, unexpected, witnessed or unwitnessed, nontraumatic and non-drowning death in patients with epilepsy, with or without evidence for a seizure and excluding documented status epilepticus, in which post mortem examination does not reveal a toxicologic or anatomic cause of death.’
  • a seizure a patient is usually unable to get help, talk, think, or act.
  • help isn’t afforded patients may suffer related injuries, such as from falls, traffic accidents, and other events. There are some types of seizures, if not attended to, that can be fatal.
  • vEEG video-electroencephalography
  • vEEG video-electroencephalography
  • vEEG video-electroencephalography
  • Automated, machine learning-based seizure detection and prediction techniques combining various sensors increases the potential of predicting any upcoming seizures.
  • OBJECT OF INVENTION The intended goal behind the subject matter of this system and device is to create a low-variable epilepsy-prediction device for the management of epilepsy, including but not limited to prediction, detection, notification, journaling, tracking, and reporting.
  • the system is also intended to predict any sudden death due to epilepsy, and aims to prevent and eliminate the risk of such sudden death by the prior prediction of any epilepsy attacks.
  • the secondary goal of this subject matter is to create a versatile, scalable system and device whose capacities can be extended for the management of a multitude of diseases and disorders, not being restricted to epilepsy.
  • IoT Internet of Things
  • a set of inputs are received by a input module, the inputs include wearable device sensor data regarding an individual's physiological parameters, distributed IoT system’s data related to additional physiological data about the individual and his surroundings, enterprise system’s data related to medical information of the individual, sentiments expressed on social media by the individual, and direct user data entered by the individual.
  • multimodal, multisource, and multilingual processing is carried out using a processing module on the set of inputs, and a comprehensive profile of the individual is generated. Further, based on processing recurring patterns, discerns triggers, stressors, reactions, and recovery phases are identified. And, even the potential adverse events are forecasted. Optionally, alerts are generated in case of anticipation of adverse event.
  • the comprehensive report and other processed data are rendered on to an output module.
  • Fig.1 illustrates a base unit that can be worn by a user, in accordance with an implementation of the present subject matter.
  • Fig.2 shows an exemplary connector that connects the base unit and the sensing unit.
  • FIG. 3 shows an exemplary sensing unit that can be attached to the base unit or can be independently worn by the user, so as to be placed on to skin of the user to form a unit comprising of a single or two-lead ECG, an Electrodermal Activity (EDA) sensor, an Electromyography (EMG) sensor, temperature sensor, environment sensor, and motion sensors.
  • Fig.4 shows an exemplary extension of the sensing unit to form additional leads.
  • Fig.5 shows an exemplary assembly of independent units to form a complete, detachable wearable unit, with extended functionality to understand sympathetic function, muscle contractions, temperature profile and physical activity and orientation.
  • Fig.6 illustrates an exemplary implementation of a method for predicting seizure using EDA and BVP data points.
  • Fig.7 illustrates combined power spectrogram of EDA & BVP developed during implementation of methodology of Fig.6.
  • Fig. 8 illustrates graphically 5 top DC segments that may contain a seizure developed during implementation of methodology of Fig.6.
  • Fig. 9 illustrates graphically identification of seizure within a DC segment developed during implementation of methodology of Fig.6. DETAILED DESCRIPTION
  • IoT powered platform is a collection of managed and platform services across edge and cloud that connect, monitor and control billions of IoT assets.
  • the subject-matter herein relates to an Internet of Things (IoT)-powered wearable device and an IoT network that includes various other devices and sensors for the prediction and management of chronic disorders.
  • the wearable device and IoT network may be adapted to extract and process data related to an individual’s health to monitor physiological parameters non-invasively on a continuous basis for an individual with or without a chronic disorder and provides predictions prior to the occurrence of an adverse event. Based on the predictions, preventive care can be provided to the individual, and the adverse events either aborted, arrested, or managed in an appropriate manner, by either caregivers or medical professionals.
  • the subject-matter herein presents an innovative biosensor system and a wearable device tailored for individuals with epilepsy (PWE), empowering them to predict, detect, monitor, document, analyze, notify, and journal epileptic seizures or any abnormal patterns.
  • This wearable device contains a multitude of sensing capabilities to provide condensed, specific, and real-time information. It enables not only PWE but also medical practitioners, health-care professionals, and caregivers to remotely monitor, diagnose, and treat patients with not just epilepsy, but also suffering from disorders encompassing Central Nervous System (CNS), Autonomous Nervous System (ANS), cardiological, psychological, and other disorders.
  • CNS Central Nervous System
  • ANS Autonomous Nervous System
  • the subject matter herein tracks the activities of the human or animal physiology by monitoring bodily changes using a multitude of sensing capabilities, which includes bit not limited to temperature sensors, skin temperature, conductance and body temperature sensor, respiration sensor, blood pressure sensor, electrodermal activity (EDA) sensor, electromyography (EMG) sensor, piezoelectricity sensor, accelerometer, gyroscope, magnetometer, pressure sensor, vibration sensor, electrocardiogram, and pulse oximetry.
  • the wearable device enables predictive insights including both not limited to seizure onsets, abnormal events and their progress in real- time.
  • the subject matter herein can be device agnostic which can be applied on various wearable devices such as a glove, a sock, an arm band, a ring, a badge, a headband, an earplug, a vest, strap, or have any form factor that allows it to obtain the required information from the individual wearing the wearable device.
  • the present subject matter has been described using a novel, detachable system that can be assembled to form multiple shapes to form a wearable best suited for the disease or disorder that need to be addressed.
  • the present subject matter enables the creation of unique and personalized profiles using biomedical signal data collected from device sensors along with data from other sources.
  • Data from sensors is automatically infused with input from third party IoT and other static and internet-enabled electronic systems, large enterprise systems like electronic medical reports (EMR), hospital management systems, social media systems, global positioning system (GPS), environment-related data, and other smart systems to generate personal health profiles and predictive health insights about individuals.
  • EMR electronic medical reports
  • GPS global positioning system
  • user-input data may also be added along with the input from various sensors.
  • the unique profiles thus created are updated and adjusted automatically by the wearable system as it collects more data over a period of time.
  • lack of trained doctors and lack of availability of diagnostic devices locally pose as a huge challenge in the detection and management of disorders like epilepsy. Sudden Unexpected Death in Epilepsy (SUDEP) is a rare but serious complication associated with epilepsy.
  • SUDEP cardiac arrhythmias
  • One significant aspect is the dysfunction of the autonomic nervous system, which controls involuntary bodily functions like heart rate and breathing. This dysregulation may lead to potentially fatal changes in heart rhythm and respiratory patterns. Additionally, individuals with epilepsy, particularly those with poorly controlled seizures, may be at a higher risk of experiencing cardiac arrhythmias. Seizures, especially generalized tonic-clonic seizures, can induce changes in heart rate and blood pressure, potentially triggering cardiac arrhythmias or other cardiac events. Furthermore, certain antiepileptic medications can influence cardiac function, with effects varying depending on the specific medication and dosage.
  • ECG electrocardiography
  • ECG monitoring is a crucial component of SUDEP prevention, it can be part of a comprehensive approach that involves multimodal data acquisition.
  • event monitoring including for SUDEP monitoring, continuous, ambulatory ECG recordings can be performed. But this is mainly done using the non-invasive Holter-method, connected to traditional ECG electrodes.
  • a portable recorder is carried by the patient, while the patient conducts normal daily activities.
  • two electrodes these are placed diagonally across the chest, one on the fifth rib, along the left anterior, axillar line, and the other on the right clavicula (collar bone).
  • Using three electrodes these are placed on the right and left collar bones, and on the fifth rib, along the left anterior, axillar line. Also with these positions, there will be multiple attachment points, and cables crisscrossing the chest of the patient.
  • the device is wireless and detachable and is based on Internet of Things (IoT) and Artificial Intelligence/Machine Learning (AI/ML) technology.
  • the wireless components involved are low-cost and detachable, transforming instantly from a one-lead ECG to a multi-modal, multiple-lead ECG system.
  • the wearable device has a base unit Fig.1.
  • the base unit serves as the main processing and data transmission hub 1. It contains the analog front-end, ADC, signal processing, storage, sensors, AI/ML module, communication modules, and power management units 1.
  • the base unit is designed to be compact and lightweight so that it can be comfortably worn by the user.
  • the base unit has multiple, detachable lead ports 2.
  • the ports connect to sensing units that are designed to securely attach and hold the ECG leads and other required sensors.
  • the base unit features multiple mechanical connectors strategically placed to align with different lead configurations and to accommodate the placement requirement of other sensors. These connectors can be physical slots, ports, or magnetic attachment points 2. The number and arrangement of these connectors depend on the types of leads that’s required by doctors (e.g., 3- lead, 5-lead, 12-lead configurations).
  • lead modules can come in different shapes and sizes, each equipped with the necessary electrodes, sensors and wiring according to the lead configuration it supports (e.g., limb leads, precordial leads) 7, 9, 11.
  • the modules also have mechanical connectors on one end that securely attach to the base unit's and extension connectors on the other end 6, 8, 10, and 12.
  • Fig.4 shows the additional extension unit.
  • the base unit Fig.1 detects the type of sensing unit Fig.3 connected to it. This detection can be achieved in various ways, such as using identification chips on the lead modules, specific electrical signatures, or even visual markers recognized by the device's sensors. Once the type of lead module is identified, the base unit interprets the signals received from each electrode.
  • the ECG leads have connectors 3, 6, 8, 10, 12, 13, 16 that can easily snap into the lead ports on the base unit.
  • Each lead could represent a different limb lead (e.g., RA, LA, RL, LL) or additional leads for specific purposes.
  • the sensing unit can be easily converted from a one-lead sensing unit to16-lead ECG Fig. 5 depending on the diagnostic need.
  • the sensing unit can be assembled effortlessly to increase the number of leads.
  • the unique design allows the ECG leads to be placed in an appropriate location on the chest or back of a patient to derive optimal signals for predictive processing.
  • the detachable joints of the wearable device encompass strategically placed male and female connectors to optimize user comfort, signal integrity, and usability.
  • the male connector featuring protruding pins or plugs, is ideally positioned on the base unit 2, facilitating direct attachment to the female connector on the receiving end, which is the sensing unit 3, or the connect units 4.
  • the female connector is equipped with corresponding receptacles for the male pins and situated on the receiving device.
  • the processing unit (or processing module) 1 of the base unit is a critical component responsible for various functions, including signal acquisition, signal processing, data analysis, and communication. It plays a central role in capturing the electrical signals generated by the heart, converting them into digital data, and making it usable for real-time display and analysis.
  • the processing unit 1 of the base unit includes analog front-end circuitry that interfaces with the sensing unit consisting of sensors and electrodes 4, 5, 7, 9, 14, 15 that are attached to the body to capture the raw electrical signals produced by the heart.
  • the analog front-end is designed to amplify and filter the signals to improve the signal-to-noise ratio and remove unwanted noise.
  • the Analog-to-Digital Converter converts these continuous analog signals into discrete digital samples. This conversion is necessary for further digital signal processing and analysis.
  • the digitized ECG and sensor data undergoes various signal processing techniques to enhance the quality of the waveforms and extract relevant features. DSP algorithms can include filtering (e.g., low-pass, high-pass, and notch filters), baseline wandering removal, and noise reduction.
  • DSP algorithms can include filtering (e.g., low-pass, high-pass, and notch filters), baseline wandering removal, and noise reduction.
  • One of the critical features in ECG analysis is the detection of R-waves, which represent the depolarization of the ventricles.
  • the processing unit 1 uses algorithms to accurately identify R- waves, enabling the determination of heart rate and other cardiac parameters.
  • the processing unit 1 is equipped with communication modules such as Bluetooth or Wi-Fi, allowing seamless data transmission to external systems and devices like cloud systems, smartwatches, smartphones, tablets, or computers. This enables real-time monitoring, data visualization, and remote analysis by healthcare professionals.
  • the processing unit 1 may use machine learning algorithms capable of analyzing the data for signs of arrhythmias, ischemia, or other cardiac abnormalities that may indicate a seizure. This analysis can provide real-time alerts to users and healthcare providers for timely intervention.
  • the processing unit of a wearable unit/device is a combination of hardware and software designed to handle the analog signals from the body, process them into digital data, and enable efficient communication and analysis to facilitate continuous cardiac monitoring and diagnostics.
  • the base unit Fig.1 and the sensing unit Fig.3 with the required number of leads can be attached to the chest or an appropriate location as determined by the experts.
  • the wearable device can be designed with built-in elastic straps or bands that can be comfortably wrapped around the chest. These straps are adjustable to fit various chest sizes while keeping the electrodes in proper contact with the skin.
  • the device can utilize snap buttons or fasteners, similar to those used in clothing, to attach the module to a fabric chest belt.
  • the belt can then be worn snugly around the chest to ensure good electrode contact.
  • the wearable device and the chest strap can be equipped with magnets that allow for easy attachment and removal. The magnets are strong enough to hold the device securely in place during movement.
  • silicone adhesive pads can be attached to the back of the wearable. These pads are hypoallergenic, reusable, and provide a gentle adhesive force to stick to the skin without causing irritation.
  • the device can have Velcro strips on its back, and a matching Velcro patch can be attached to the chest strap or directly on the chest. This method allows for easy and adjustable attachment of the module.
  • the wearable can also be designed as part of compression garments (e.g., sports bras, shirts). These garments provide a snug fit, maintain good skin contact, and reduce motion artifacts during physical activities. These alternative attachment methods aim to ensure a secure and comfortable connection between the electrodes and the skin, providing accurate measurements while minimizing discomfort or skin irritation.
  • the received signals from the patient can be transmitted over a network to anywhere in the world.
  • the system can be used for home as well as for hospital monitoring, giving users affordability and doctors relevant insights to provide timely care.
  • the system comprises AI and machine learning- based algorithms that provide predictions of any untoward events and also show doctors if the current treatment regime, if any, is working.
  • the system of the present subject matter can be life- saving and can contribute to bringing down the incidence of unnecessary cardiac deaths.
  • the wearable offers various lead types Fig.2, Fig 3. And Fig. 4 that users can interchange.
  • standard limb leads, augmented limb leads, or precordial leads could be available that can be mix and matched or assembled using connector units Fig.2 and Fig.4.
  • each lead and its corresponding lead port are color- coded, simplifying the process of attaching the leads in the right configuration.
  • the design considers maintaining good signal integrity and minimizing interference. Proper shielding and grounding techniques helps to reduce noise and ensure accurate measurements.
  • the wearable device has a user-friendly interface, such as an LCD screen or a smartphone app, to display real-time ECG data and guide users in lead placement.
  • the detachable leads has a secure locking 3, 6, 8, 10, 12, 13, 16 mechanism to prevent accidental disconnections during use. This ensures reliable and continuous ECG monitoring. Since the wearable can have multiple leads attached, efficient power management is crucial to extend battery life and hence, the entire system is designed to conserve power when fewer leads are in use.
  • a communication module is integrated into the base unit 1 for seamless data transfer to external devices like smartphones or computers.
  • the sensing unit also comprises a communication interface. The communication interface communicates the received signals from the plurality of leads and other sensors to the base unit that stores, analyzes, and interprets the received cardiac signals.
  • the communicating interface can also transmit the recorded cardiac signal, either in realtime or in an offline mode to an external unit, not limited to a microprocessor, a smartphone, a smartwatch, a central computer, and a cloud computing device.
  • the received signals can be accessed by a user or a medical professional over a network to monitor and diagnose the user in real-time or when a situation arises.
  • the base unit Fig. 1 and the sensing system Fig. 3, which includes the plurality of leads is provided on a flexible substrate and the flexible substrate is contacted with the user.
  • the plurality of leads is detachably attached to the flexible substrate. Therefore, the number of leads can be changed from 1-16 during monitoring easily.
  • the plurality of leads are individual units Fig. 3 that can be attached to the user, for example, by using medical adhesives.
  • the plurality of leads can be placed on an appropriate position, not limited to chest or back of the patient, thereby, ensuring further flexibility in monitoring.
  • Each of the plurality of leads is associated with an electrode, a communication interface, and a plurality of sensors Fig.3.
  • the plurality of sensors includes, but are not limited to motion sensors, respiration sensor, temperature sensor, electrodermal activity sensor, electromyography (EMG), environment sensors and integrated audio and video.
  • EMG electromyography
  • the various components are powered by a rechargeable battery.
  • the electrodes receive cardiac signals from the user.
  • the plurality of sensors receive and measure parameters, such as temperature of the patient and surroundings, respiratory rate, galvanic skin voltage, electrical activity of muscle tissue, and the like.
  • the received data namely cardiac signals and data measured by the sensors, is then communicated to the base unit 1 via the communication interface. Data can also be received from third-party systems, medical devices, and other data sources to gather insights about the user’s health.
  • the base unit Fig.1 comprises processor, memories, databases, modules, and data.
  • the memories may include any non-transitory computer-readable medium including, for example, volatile memory (e.g., RAM), and/or non-volatile memory (e.g., EPROM, flash memory, etc.).
  • the memories may include an external memory unit, such as a flash drive, a compact disk drive, an external hard disk drive, or the like.
  • the device transmits the acquired and processed data over the communication interface over a network.
  • the network is a wireless network.
  • the network can also be an individual network or a collection of many such individual networks, interconnected with each other and functioning as a single large network, e.g., the Internet or an intranet.
  • the network can include different types of networks, such as intranet, local area network (LAN), wide area network (WAN), the internet, and such.
  • the network may also include individual networks, such as, but not limited to, Global System for Communication (GSM) network, Universal Telecommunications System (UMTS) network, Long Term Evolution (LTE) network, etc.
  • GSM Global System for Communication
  • UMTS Universal Telecommunications System
  • LTE Long Term Evolution
  • the network includes various network entities, such as base stations, gateways, servers, and routers, etc.
  • the data received by the base unit Fig.1 is stored, analyzed, interpreted or a combination, thereof.
  • the base unit is a microprocessor
  • machine learning algorithms within the microprocessor can analyze and interpret the data and provide the interpretation to the patient or doctor.
  • the base unit is a cloud computing environment
  • the doctor or medical professional can retrieve the data and interpret the data.
  • the received data from the plurality of leads and plurality of sensors is communicated to multiple other devices by the communication interface each of which is capable of storing, analyzing, and displaying interpretations of the received data.
  • the base unit Fig.1 and sensing unit Fig.3 is connected to the user being monitored.
  • the number of the leads can be attached and detached based on the requirement of the patient, doctor, or based on the situation.
  • Fig. 1 depicts an example of an enclosure of a base ECG unit.
  • Fig.3 is an example sensing unit that can be attached to the base unit or attached to the skin of the user, in accordance with an implementation of the present subject matter.
  • the plurality of leads along with the plurality of sensors measure various parameters, such as cardiac signals, environmental factors, and the like.
  • the received data is then communicated by the communication interface to another device or system. In an example, the communication of the data is in real time.
  • Data can, therefore, be accessed and streamed by the doctor or medical professional at a remote location.
  • Data from the wearable can inform doctors about all events that have happened to patients at their home and local environments. Streaming data can alert doctors about impending events, show efficacy of medicines, general health and activity status of the patient, like triggers, underlying behavioral and psychological problems and also determine the next course of action. Further, streaming of data can inform the doctors about possible triggers for events, which can allow doctors to prescribe preventive lifestyles.
  • the wearable incorporates flexible circuitry utilizing printed circuit boards (PCBs) or flexible electronics, ensuring secure electrical connections and component integration. This allows the wearable to bend and fold without any risk of circuitry damage.
  • a robust hinge mechanism is integrated 3, 6, 8, 10, 12, 13, 16 enabling smooth and repeated folding and unfolding of the various units of the wearable device Fig.1, Fig.2, Fig 3, and Fig.4.
  • the compact component placement ensures optimal space utilization while maintaining full functionality.
  • the wearable contains an inbuilt foldable display, powered by foldable OLED or e-paper technology, offers real-time visual feedback during recording while remaining flexible for folding. Efficient power management ensures optimal battery life by automatically transitioning to low-power or standby mode upon folding.
  • Wireless connectivity such as Bluetooth, facilitates real-time data transmission to smartphones, tablets, or other devices for monitoring and analysis without the need for physical connections.
  • all folding components are designed to withstand normal wear and tear, while protective packaging ensures safe transportation and storage.
  • the detachable lead ports and the main unit have magnetic connectors 3, 6, 8, 10, 12, 13, 16 that snap together effortlessly when brought close.
  • the magnets ensure a secure and reliable connection.
  • the locking system has a microcontroller that acts as the "brain" of the smart locking mechanism, that processes signals from the sensors and determines when the lead is properly aligned for locking.
  • the sensors are responsible for detecting the presence and alignment of the lead in the port. These sensors can include proximity sensors or other types of detection technologies.
  • There actuator functions as a locking block that physically secures the lead port onto the main ECG unit. It can be a small motor or magnet that engages when the lead is correctly aligned and locks the connection.
  • a power source represented by a battery, provides the energy needed for the locking system to function.
  • the wearable communicates with implantable devices to provide complete and real-time health information.
  • the communication system for the wearable ECG and implantable devices includes the selection of an appropriate wireless communication protocol (e.g., Bluetooth Low Energy), bidirectional communication for real-time data exchange, robust data encryption for security (HIPAA compliance), secure device pairing, data format standardization (HL7/DICOM), real- time data streaming, integration with medical software platforms, compliance with medical device certification standards, thorough interoperability testing, a user-friendly interface with alerts, power management optimization, and a comprehensive patient data management system for secure storage and long-term monitoring of ECG and implantable device data.
  • an appropriate wireless communication protocol e.g., Bluetooth Low Energy
  • HIPAA compliance robust data encryption for security
  • HIPAA compliance secure device pairing
  • data format standardization HL7/DICOM
  • real- time data streaming integration with medical software platforms, compliance with medical device certification standards, thorough interoperability testing, a user
  • the system extends beyond the functions described herein by seamlessly integrating a wide range of biomedical and other third-party devices and systems.
  • the smart electronic enclosure for the wearable is a compact and durable housing that protects all essential electronic components while providing advanced functionality. Its sleek design ensures user comfort and portability.
  • the exterior offers customization options, allowing personalization to suit individual preferences.
  • the enclosure is built with robust materials to withstand daily wear and tear, ensuring the longevity of the device.
  • Its smart features include wireless connectivity for seamless data transmission, LED indicators for status updates, and a user-friendly interface for real-time feedback.
  • the enclosure prioritizes user safety and data security through advanced encryption measures. With its cutting-edge design, the smart electronic enclosure enhances the overall usability and effectiveness of the wearable device.
  • the wearable can be a tattoo that comprises a flexible and skin-friendly substrate made of materials like flexible polymers or bio-compatible substances. It integrates diverse electronic components, such as sensors for monitoring biometric data like heart rate and temperature, microcontrollers, and communication modules for seamless data transmission, including Bluetooth.
  • the tattoo can be powered by batteries or energy-harvesting mechanisms, like body heat or motion, ensuring autonomous operation. Conductive traces or interconnects establish connections between the electronic components, while encapsulation or coating shields them from moisture and the environment.
  • the tattoo attaches securely to the skin through an adhesive layer, and in some cases, flexible printed circuit boards (PCBs) provide a more structured platform.
  • PCBs flexible printed circuit boards
  • wireless communication capabilities like Bluetooth or NFC enable data transfer.
  • the system and method of the invention can be implemented using any other structure of wearable devices, including other components either placed in proximity or remotely with respect to the wearable device.
  • the wearable device can have some of the components embedded within it like processing module and output module. For capturing physiological data, many of the sensors to be kept on to the skin of a patient are also embedded into it.
  • the system includes an input module, a processing module and an output module.
  • the input module receives a set of inputs comprising wearable device sensor data regarding an individual's physiological parameters, distributed IoT system’s data related to additional physiological data about the individual and his surroundings, enterprise system’s data related to medical information of the individual, sentiments expressed on social media by the individual, and direct user data entered by the individual.
  • the parameters indicative of surroundings of the individual includes user activity, vehicle information, weather information, and audio, or video input, or combination thereof.
  • the enterprise system inputs indicative of medical information of the individual include hospital data, diagnostics data, or insurance data, or combination thereof.
  • the user inputs provided by the individual include responses to health questionnaires, feedback, or self-reported information, or combination thereof.
  • the processing module conducts multimodal, multisource, and multilingual processing on the set of inputs, to generate a comprehensive profile of the individual, to identify recurring patterns, discerns triggers, stressors, reactions, and recovery phases, and to forecast potential adverse events.
  • the processing module also generates alert in case of anticipation of adverse event
  • the output module is coupled to the processing module and receives the processed information, and to present them as a comprehensive report.
  • the output module also optionally render the alerts when an adverse event is anticipated.
  • the processing module is also optionally adapted to process physiological parameters and additional physiological parameters along with one or more of Electroencephalography (EEG) data, Functional near-infrared spectroscopy (fNIRS) data, Skin temperature data, Skin Conductance Level (SCL) data, Skin Conductance Response (SCR) data, Body temperature data, Heart Rate (HR) data, Heart rate variability (HRV) data, Blood Pressure and trend data, Respiration and trend data, SpO2 data, Electrodermal Activity (EDA) data, Blood Volume Pulse (BVP) data, Respiratory Sinus Arrhythmia (RSA) data, Electromyography (EMG) data, Motion data, 3-axis accelerometer data, 3-axis gyroscope data, 3-axis magnetometer data, Piezo film data, Piezo cable data, Vibration data, Impact data, Altitude data, or other biomedical sensor data that can be combined with an ECG data.
  • EEG Electroencephalography
  • fNIRS Functional near-infrared spectroscopy
  • the processing module is also optionally adapted to instruct one or more implantable devices, such as pacemakers and Vagus Nerve Stimulators (VNS) on determination of adverse events.
  • the processing module is remotely coupled to a client computing device for communicating the data, view trends, and notifications for potential abnormalities.
  • the processing module is also optionally adapted to identify patterns based on combining two or more physiological parameters, thresholds, climatic conditions, changes to locations, motion, audio and video input.
  • the patterns may include normal patterns, abnormal patterns, disorder specific patterns, and unknown patterns.
  • the plurality of inputs maybe pre-processed for performing noise cancelation, filtering, and smoothening.
  • the processing module may also perform AI-based learning based on patterns of group of individuals with similar patterns, identify evolution of a disorder over a period of time from low grade to severe, correlating triggers, stressors, and effects of medication and therapy, and track underlying physical and mental health conditions that develop over a period of time. Predicting an adverse event comprises calculating a stress score of the individual based on sensor data, IoT application and network data, enterprise data, social media data and user input.
  • the implementation of the invention is further exemplified for automatic identification of epileptic seizures using BVP (Blood Volume Pulse and EDA (Electrodermal Activity) derived features. The subjects wore the sensors on their non-dominant hand for the duration of the recording period.
  • EDA and BVP were extracted from the distal and proximal phalanges of the hand using the respective sensors.
  • a seizure prediction and identification framework is defined which is used by models used by processing module to identify a seizure.
  • the characteristic features of a seizure are: ⁇ Segment of higher power: There is a lead time to an epileptic seizure, which forms the basis of identifying preictal windows. The lead time is reflected as a period of statistically significant change in the EDA and BVP Power signals, distinguishable from other periods of autonomic arousal.
  • ⁇ Stationary, upward trend The seizure signals both show a period of stationary, positive uptrend for a sustained period, culminating in a seizure event.
  • ⁇ Surge “Seizure activity is characterized by abnormal, excessive, brief episodes of electrical discharges that disturb the normal activity of the nervous system.” Seizures show power surges that can be clearly seen in the frequency spectrum of signals. Surges can be described as brief “overvoltage spikes or disturbances on an electrical waveform,” and show significantly higher power than surrounding segments.
  • ⁇ State transition Seizures show clear transition in emotional states, with the seizure state showing hyper internal excitement. The steps mentioned in Fig.6 are elaborated in the following sections. It is to be noted that all the steps were processing using input module and the processing module of the system. Data acquired from subjects underwent an extensive pre-processing and exploration phase, to make it suitable for the machine learning models.
  • ⁇ Statistical and common signal-based features Various linear and non-linear features related to EDA and Heart Rate Variability (HRV) were used, which includes Tonic Phasic, SMNA, Amplitude, Rise Time, Recovery Time etc for EDA and time and frequency domain features for HRV.
  • HRV Heart Rate Variability
  • ⁇ Rules based features They form the backbone of rules-based and ML models, showing the best demarcation among classes. Rules-based features like Power, Convergence and Divergence, Area Between Curve and Rate of Change are used which are described in details further.
  • the Power component of the EDA and the BVP signals were significantly changed during a seizure, and these changes in the Power signal provides signatures of a seizure.
  • the first step in the process is to convert the EDA and the BVP signals to frequency domain.
  • STFT Short-Time Fourier-Transform
  • the hop size is the difference between the window length M and the overlap length L.
  • X_scaled X_std * (max - min) + min g. Add the corresponding elements in the two matrices, to derive a single Power matrix for each subject as shown in Fig.7. This signal will be used as the base for deriving related parameters.
  • the Divergence points indicate from when the short-term Power signal became stronger compared its overall long-term trend, while the Convergence points show when the short term behavior of the Power signal is in sync with its normal, long term behavior.
  • a Divergence point is the start of a segment, while a corresponding Convergence point that shows when a Fast line is crossing over the Slow line, is the end of that particular segment.
  • the segment that is clearly distinct than all other segments are sought to find the one that hold signatures of a seizure.
  • Rate of Change To understand the nature of change in the Power signal, the maximum rate of change is calculated as the percentage increase or decrease in Power over a given period of time. To derive the maximum ROC, the current values of the Power signal are compared with the values in another time period, which in current implementation is every 1 minutes. It was consistently found that the DC segment that contains the seizure to have the highest rate of change, especially during or prior to the electrographic onset of a seizure.
  • the AWC is calculated as the integral of the difference of the two rates of change.
  • the formula is given below: Automatic labelling of seizure segments To distinguish the probable segment that contains a seizure from other parts of a signal, a product of AWC, Power, ROC of Power, Statistical Moments and Z-Score was computed for each of the DC segments.
  • top 5 segments with the highest value are labelled as Suspected Seizure Activity (SSA), while the rest will be labelled Neutral once the rules mentioned above executed sequentially. Neutral segments will internally be divided into Neutral and Anxious. Exemplary graphical representation of Top DC segments is illustrated using Fig.8. Peaks (1), (2) and (3) shown in the above image are likely to contain a seizure and their corresponding time segments will be labelled as SSA. Recursive Feature Elimination (RFE) and Principal Component Analysis (PCA) is used to identify the most important statistical and signal-based features to use in our model. All rules- based features were mandatorily added to the final list of selected features.
  • RFE Recursive Feature Elimination
  • PCA Principal Component Analysis
  • the seizure dataset was automatically labelled using a script that computed the rules-based features, and then identified segments where the product of all the rules-based features was the highest.
  • the top five such segments were labelled as Suspected Seizure Activity (SSA), implying that these segments have the highest probability of containing a seizure.
  • SSA Suspected Seizure Activity
  • the time period enclosed within the start time of a seizure segment and the end time of the segment is considered as the seizure window, and the data points within this time segment is labelled as SSA. Rest of the data points are labelled as Neutral.
  • the Neutral class was further divided into Neutral and Anxious. The time enclosed within the start time of a seizure window and the actual seizure event will be defined as the pre-ictal period.
  • Model for prediction and detection of seizure A hybrid approach was used to automatically predict and detect a seizure. In the first step, it was predicted whether a seizure will be present in the data and if so what is the pre-ictal time. Secondly, the actual event in a specific segment was detected and confirmed. The dataset and its constituent features, which were uniformly used across algorithms, were fed to a group of supervised and unsupervised algorithms to get the desired output.
  • Model Name Description 1 Ensemble Model (ML In ensemble modelling multiple algorithms are e e s. t l ⁇ Model Name Description split according to a specified parameter.
  • Support Vector Machine Support Vector Machine (SVM) is a type of ML , the seizure might be present, further the exact narrow window is identified in which a seizure likely presented itself.
  • SVM Support Vector Machine Support Vector Machine
  • LSTM Long Short Term Memory
  • LSTM cell consists of 64 units. Cell was layered sequentially 2 times with dropout layer after each cell. The dropout rate was set to 0.2 between layers to reduce overfitting. A fully- connected dense layer was used at the end. AdamOptimizer was used to train the model using default learning rate of 0.01. A batch size of 32 was used for training, validation, and testing. The sequence length was chosen as 100 as it was faster and also yielded good result.
  • Fig.9 illustrates a graphical exemplary representation of a DC segment for identification of seizure.
  • the section enclosed within A and B have been identified as the active seizure zone by the voting classifiers.
  • the section within D and C denotes the complete seizure window, where D represents the start of the pre-ictal period, while C represents return to baseline power post seizure.
  • S is the time identified by the LSTM AutoEncoder as the minutes with the maximum reconstruction loss, showing significant anomaly in the signal.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Physics & Mathematics (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Molecular Biology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Biophysics (AREA)
  • Veterinary Medicine (AREA)
  • Neurology (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Physiology (AREA)
  • Neurosurgery (AREA)
  • Data Mining & Analysis (AREA)
  • Accounting & Taxation (AREA)
  • Computing Systems (AREA)
  • Bioethics (AREA)
  • Cardiology (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Business, Economics & Management (AREA)
  • Evolutionary Computation (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

L'invention concerne un système de prédiction et de prise en charge de troubles chroniques. Le système comprend un module d'entrée, un module de traitement et un module de sortie. Le module d'entrée reçoit un ensemble d'entrées comprenant des données de capteur de dispositif portable concernant les paramètres physiologiques d'un individu, des données distribuées du système d'IdO relatives à des données physiologiques supplémentaires concernant l'individu et son environnement, des données du système d'entreprise associées à des informations médicales de l'individu, des sentiments exprimés sur les médias sociaux par l'individu et des données d'utilisateur directes entrées par l'individu. Le module de traitement effectue un traitement multimodal, multisource et multilingue sur l'ensemble d'entrées, génère un profil d'individu, identifie des motifs récurrents, discerne des déclencheurs, des facteurs de stress, des réactions et des phases de récupération et prévoit des événements indésirables potentiels et génère éventuellement une alerte en cas d'anticipation d'événement indésirable. Le module de sortie restitue les informations traitées et présente un rapport complet et restitue éventuellement les alertes lorsqu'un événement indésirable est anticipé.
PCT/IB2023/060566 2022-10-19 2023-10-19 Réseau d'ido agnostique de dispositif pour la prédiction et la prise en charge de l'épilepsie et la prévention du risque de mort subite inattendue en épilepsie WO2024084429A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IN202241022874 2022-10-19
IN202241022874 2022-10-19

Publications (1)

Publication Number Publication Date
WO2024084429A1 true WO2024084429A1 (fr) 2024-04-25

Family

ID=90737207

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2023/060566 WO2024084429A1 (fr) 2022-10-19 2023-10-19 Réseau d'ido agnostique de dispositif pour la prédiction et la prise en charge de l'épilepsie et la prévention du risque de mort subite inattendue en épilepsie

Country Status (1)

Country Link
WO (1) WO2024084429A1 (fr)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180113987A1 (en) * 2016-10-20 2018-04-26 Jiping Zhu Method and system for quantitative classification of health conditions via wearable device and application thereof
US20210151179A1 (en) * 2017-08-03 2021-05-20 Rajlakshmi Dibyajyoti Borthakur Wearable device and iot network for prediction and management of chronic disorders

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180113987A1 (en) * 2016-10-20 2018-04-26 Jiping Zhu Method and system for quantitative classification of health conditions via wearable device and application thereof
US20210151179A1 (en) * 2017-08-03 2021-05-20 Rajlakshmi Dibyajyoti Borthakur Wearable device and iot network for prediction and management of chronic disorders

Similar Documents

Publication Publication Date Title
Anikwe et al. Mobile and wearable sensors for data-driven health monitoring system: State-of-the-art and future prospect
US10813565B2 (en) Wearable monitor
US10213152B2 (en) System and method for real-time measurement of sleep quality
Hadjem et al. An ECG monitoring system for prediction of cardiac anomalies using WBAN
CN105411554A (zh) 一种无线式无创伤人体生理参数采集、检测及智能诊断系统
Rajan et al. Viable investigations and real-time recitation of enhanced ECG-based cardiac telemonitoring system for homecare applications: a systematic evaluation
US10959662B2 (en) Seizure prediction using cardiovascular features
Crifaci et al. ECG and GSR measure and analysis using wearable systems: Application in anorexia nervosa adolescents
Periyaswamy et al. Ambulatory cardiac bio-signals: from mirage to clinical reality through a decade of progress
JP2022508124A (ja) 脳卒中後症状を有する患者における心房不整脈をモニタリングし特性評価する長期非侵襲的システム
Chowdhury et al. Wearable Real-Time Epileptic Seizure Detection and Warning System
WO2024084429A1 (fr) Réseau d'ido agnostique de dispositif pour la prédiction et la prise en charge de l'épilepsie et la prévention du risque de mort subite inattendue en épilepsie
US20240156389A1 (en) Health monitoring system and method
US20220338801A1 (en) Wearable System for Automated, Objective and Continuous Quantification of Pain
WO2022040811A1 (fr) Procédé et appareil de rétablissement à la suite d'une commotion
US20210052229A1 (en) Systems and methods to detect cardiac events
Belaid et al. A new design of real-time monitoring and spectral analysis of EEG and ECG signals for epileptic seizure detection
Almazaydeh et al. A panoramic study of obstructive sleep apnea detection technologies
Singh et al. Non-EEG-Based Seizure Detection: A Comprehensive Review
Gorman A novel non-EEG wearable device for the detection of epileptic seizures
Hey et al. Mobile Sensors for Multiparametric Monitoring in Epileptic Patients
Shukla et al. Algorithms and Devices for Seizure Prediction and Diagnosis

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23879314

Country of ref document: EP

Kind code of ref document: A1