REMOVAL OF ARTIFACTS IN NEUROPHYSIOLOGICAL SIGNALS

Abstract

Data processing unit, comprising: a processor; a network interface connected to the processor to enable data transmission over a network; a storage unit connected to the processor; Software stored in the memory unit, wherein an execution of the software by the processor configures the data processing unit to perform actions, comprising: Receiving digital EEG data based on digitizing analog electroencephalogram (EEG) signals from a patient's first electrode; Receiving digital welding sensor data based on digitizing the patient's analog welding sensor signals, which are simultaneous with the analog EEG signals; applying a transformation to the digital sweat sensor data based on a previously determined sweat stress profile of the patient; and an adaptive adjustment of the digital EEG data by subtracting the transformed welding sensor data from the digital EEG data.

Description

BACKGROUNDTechnical field

[0001] The present disclosure relates generally to signal processing and, more particularly, to detection, identification, and removal of artifacts in neurophysiological signals. Description of related technology

[0002] In recent years, medical interest in brain-computer interfaces and intelligent signal segmentation has increased. The brain's electrical activity is represented by electroencephalogram (EEG) signals. An EEG is a test that uses electrodes (which may be small, flat metal discs) attached to the scalp to detect electrical activity in the brain. Brain cells communicate with each other via electrical impulses and are constantly active, even during sleep. Such activity manifests as voltage fluctuations caused by the ionic current within the brain's neurons.

[0003] As used here, EEG refers to the recording of the spontaneous electrical activity of the brain over a period of time, as recorded by multiple electrodes placed on the scalp. Diagnostic applications generally focus on the spectral content of the EEG—that is, the nature of the neural oscillations (sometimes referred to as "brainwaves") observable in EEG signals. An EEG is one of the most important diagnostic tests for epilepsy and can also play a role in diagnosing other brain and sleep disorders. To improve the accuracy of EEG measurements, electrical noise—referred to here as artifacts—that can occur during measurements and distort or overwhelm the actual brainwave signals should be removed.

[0004] The following documents are known as state of the art: US 8 538 512 B1, US 2014 / 0 316 230 A1, US 2016 / 0 015 289 A1. SUMMARY

[0005] According to various embodiments, a data processing unit, a non-transitory, computer-readable storage medium, a computer program, and a method for removing artifacts in electroencephalogram (EEG) signals are provided. Analog EEG signals of a patient are received from a first electrode and converted into digital EEG data. Analog sweat sensor signals of the patient, which are contemporaneous with the analog EEG signals, are received and converted into digital sweat sensor data. Based on a pre-determined sweat stress profile of the patient, a transformation is applied to the digital sweat sensor data. The digital EEG data is adaptively adjusted by subtracting the transformed sweat sensor data from the digital EEG data. The adjusted digital EEG data is sent to be displayed on a user interface of the computer unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The drawings show illustrative embodiments. They do not show all embodiments. Other embodiments may be used in addition to or instead of these embodiments. Obvious or unnecessary details may be omitted to save space or provide more effective illustration. Some embodiments may be implemented with additional components or steps and / or without all of the components or steps shown. Where the same reference number is used in different drawings, it refers to the same or similar components or steps. Fig. Figure 1 illustrates an example architecture for implementing a system to perform adaptive filtering of sweat artifacts during electronic brain monitoring. Fig. 2 shows, in accordance with an illustrative embodiment, signals from multiple EEG channels over time. Fig. 3 is, in accordance with an illustrative embodiment, an example graph of sweat over time in response to various stress-inducing factors for a patient. Fig. Figure 4 shows a block diagram showing various components of an illustrative user unit at a general level. Fig. 5 shows, in accordance with an illustrative embodiment, a general block diagram of a system for adaptive filtering of welding artifacts. Fig. Figure 6 illustrates an example process of an initial training phase, where a sweat stress profile is individually tailored to a patient. Fig. Figure 7 illustrates an example process for an active phase where sweat-related artifacts are identified and removed from EEG data. Fig. 8 is, in accordance with an illustrative embodiment, a functional block diagram representation of a computer hardware platform capable of communicating with various networked components. DETAILED DESCRIPTIONOverview

[0007] In the following detailed description, numerous specific details are set forth by way of example in order to provide a thorough understanding of the relevant teachings. However, it should be understood that the present teachings may be practiced without such details. In other instances, well-known methods, procedures, components, and / or circuits have been described at a relatively general level without detail to avoid unnecessarily obscuring aspects of the presented teachings.

[0008] The present disclosure relates to the detection, identification, and removal of artifacts in neurophysiological signals. Known, available methods for automated EEG analysis to detect brain activity currently face a number of challenges that limit their reliability. Such drawbacks include biological artifacts and susceptibility to environmental noise / interference, resulting in reduced reliability. As used herein, the term "artifact" refers to any non-event signals that interfere with the relevant event signal.

[0009] One potential source of artifacts is patient perspiration. The applicants have determined that perspiration introduces a low-frequency noise source that interferes with the analysis of an EEG signal measured, for example, from the patient's forehead. Although skin rubbing may reduce the effects of skin potentials and thus mitigate the adverse effects discussed here, it is time-consuming, potentially harmful, and may not be an option in certain circumstances. For example, the EEG may be performed while asleep, or the patient may simply be unable to regularly rub their skin during an EEG analysis.

[0010] The concepts discussed here provide an automatic and efficient way to detect and remove sweat-related artifacts that interfere with EEG signals in real time, thereby providing a more accurate assessment of a patient's cognitive state. To this end, a data processing unit, which may be a user unit or a central server interacting with a user unit, receives a patient's analog EEG signals from a first electrode and converts the EEG signals into digital EEG data. Analog sweat sensor signals from the patient, contemporaneous with the analog EEG signals, are received and converted into digital sweat sensor data. Based on a pre-determined sweat stress profile of the patient, a transformation is applied to the digital sweat sensor data.The digital EEG data is then adaptively adjusted by subtracting the transformed sweat sensor data from the digital EEG data. The adjusted EEG data can then be sent in real time to a user interface of the data processing unit. Reference will now be made in detail to examples illustrated in the attached drawings and discussed below. Example architecture

[0011] Fig. 1 illustrates an example architecture 100 for implementing a system for performing adaptive filtering of sweat artifacts during electronic brain monitoring. The architecture 100 may include a network 106 that allows patients 101(1) to 101(N) to exchange data with various resources via their respective user devices 102(1) to 102(N). Networked resources may include a patient database 110, an EEG analysis server 116 with an EEG artifact engine 103 executing thereon, and any other components connected to the network 106.

[0012] Network 106 may be, but is not limited to, a local area network (LAN), a virtual private network (VPN), a cellular network, the Internet, or a combination thereof. For example, network 106 may include a mobile network that may be communicatively coupled to a private network, sometimes referred to as an intranet, that provides various ancillary services, such as data communication with a patient database 110, libraries, and the Internet. For the purposes of this discussion, network 106 will be described by way of example and not limitation as a mobile network, such as may be operated by a network operator or service provider to provide a wide range of mobile data communication services and ancillary services or features to its subscriber customers and associated mobile device users.The network 106 enables the transmission of data collected from the patients 101(1) to 101(N) to the EEG artifact engine 103 on the EEG analysis server 116 and its recording in the patient database 110.

[0013] For the purposes of later discussion, several user units are included in the drawing to illustrate some examples of the units that can collect neurophysiological and / or sweat data from a patient via electrodes 105(1) through 105(N) and receive various services via network 106. Today, user units typically take the form of portable handsets, smartphones, personal digital assistants (PDAs), and smartwatches, although they can also be implemented in other form factors such as consumer electronics and medical electronics units.

[0014] In one embodiment, electrodes 105(1) and 105(N) are removably connected to the scalp of the patient 101(1) and 101(N), respectively. Various technologies can be used for the electrodes, which are divided into two groups: (i) electrophysiological signal electrodes and (ii) sweat monitoring electrodes, collectively referred to herein as electrodes. In some embodiments, an electrode is configured to both acquire electrophysiological signals and monitor sweat parameters. In other words, a single electrode channel is operable to provide both EEG data and sweat data.

[0015] With respect to the former, an electrode is configured to capture brain activity or related data, such as EEG data, of the patient. In various embodiments, such electrodes may include surface- or intracranially mounted electrodes for measuring electrophysiological and brain-related signals, such as an EEG, ECoG, EOG, EMG, and the like. In one embodiment, the sensors may also be magnetic electrodes configured to capture magnetic fields to obtain brainwave signals similar to those that can be obtained, for example, by an electrode mounted on the patient's scalp.

[0016] For example, EEG electrodes are placed at various locations on a patient's scalp to measure the patient's neurophysiological signals. There may be multiple such electrodes, each providing a different channel of data to the associated user units (e.g., 102(1) and 102(N)). In addition, there are sweat electrodes operable to measure the amount of sweat in the vicinity of a corresponding EEG electrode.

[0017] Depending on the desired accuracy of the EEG data filtering discussed here, the ratio of sweat electrodes to the number of EEG electrodes can vary in different embodiments. For example, the ratio of sweat electrodes to EEG electrodes can be 1:1 or 1:many (where the EEG electrodes are within a predetermined radius of a sweat electrode), or each EEG electrode can have a sweat sensor embedded in it. In other words, the same electrode provides both sweat data and patient neurophysiological signals. Since sensitivity is inversely proportional to the distance between the EEG electrode and the sweat electrode, a compromise can be found to obtain good sensitivity and a difference between individual sweat values ​​that is greater than the inaccuracy in individual sweat measurements.In one embodiment, a distance regression calculation is performed to adjust the filtering (e.g., the transformation) based on the distance between the EEG electrode and the sweat sensor electrode.

[0018] In various embodiments, the electrodes (e.g., the sweat and / or EEG electrodes) may be connected to their respective user units via physical lines (e.g., cable lines) and / or wirelessly via short-range wireless data transmission. For example, the wired data transmission between the electrodes of the patient 101(1) may occur via separate lines via a wiring harness (e.g., in the form of a bundle of wires) that is physically connected to the user unit 102(1).

[0019] In other embodiments, data transmission between electrodes 105(N) and user unit 102(N) occurs via a short-range wireless data transmission technology. Various short-range wireless technologies for data transmission with user unit 102(N) include, but are not limited to, DECT (Digital Enhanced Cordless Telecommunications), NFC (Near Field Communication), ZigBee, and Bluetooth. DECT technology (conventionally used for wireless residential telephone systems) consumes significantly less power than cellular systems. NFC is a set of standards for smartphones and similar devices to establish radio data transmission between the devices by bringing them into contact or close proximity, typically no more than a few centimeters. Since the range is short, significant energy savings are achieved.While Bluetooth operates over longer distances than NFC, it still saves a significant amount of power compared to cellular systems. ZigBee is a standard for low-cost, low-power wireless mesh networks. The ZigBee mesh network provides high reliability and a longer range. This allows the user unit 102(N) to conserve power even over longer distances. In general, lower power consumption allows for smaller batteries with longer lifespans, resulting in more ergonomic form factors for the user unit 102(N) and / or the electrodes 105(N). For example, the electrodes may be part of a helmet that can be easily worn on a patient's head. The wireless technology discussed here provides greater patient comfort and mobility while avoiding accidental disconnection of an electrode attached to a receiver.

[0020] The applicants have found that sweat introduces a low-frequency, high-amplitude noise source that impairs the analysis of an EEG signal measured, for example, on the patient's forehead (e.g., 101(1)). In this context, Fig. 2 Signals from multiple EEG channels over time. Experts in EEG signal data know that such data possess a "signature," in which signals exhibit a predetermined frequency and amplitude range. An artifact in the form of sweat creates low-frequency, high-amplitude shifts in the EEG data, as shown by the data from channels 1, 2, and 11. Accordingly, in a current EEG measurement, different locations on the body produce different results.

[0021] Looking again at Fig. 1, each user unit (e.g., 102(1)) is configured to receive the electronic data collected from the respective patient (either directly via a wire or wirelessly) and to process the received data. For example, the user unit 102(1) may receive EEG signals and sweat sensor signals of a patient 101(1) from one or more of the electrodes 105(1). The EEG signals and the sweat sensor signals are simultaneous (e.g., synchronous with each other). The collected data is initially in analog form. Accordingly, the user unit 102(1) converts the analog EEG signals and the analog sweat sensor signals into corresponding digital data. In various embodiments, this data may then be (i) further processed by the user device 102(1) or (ii) sent to the EEG analysis server 116 via the network 106 for further processing by the EEG artifact engine 103.

[0022] For example, the EEG artifact engine 103 may interact with a user device (e.g., 102(1) or 102(N)) to process the digital EEG data and the sweat data in real time to provide adjusted digital EEG data from which artifacts associated with the patient's perspiration have been removed. For this purpose, the EEG artifact engine 103 may receive a sweat stress profile 111 from a patient database 110.

[0023] The patient database 110 is connected via the network 106 for data transmission. In one embodiment, the patient database 110 provides online access to a variety of functions associated with a patient's account, such as health data, the sweat stress profile 111, prescription data, password control, etc.

[0024] In one embodiment, the sweat stress profile 111 stored in the patient database 110 can be created during a training phase, whereby a sweat stress profile is created individually for a patient. In this way, the patient's response (e.g., 101(1)) to controlled stress-inducing factors is identified. For example, a patient can wear a virtual reality (VR) or augmented reality (AR) interface to accelerate physical exertion and response to stress-inducing aural, visual, and physical factors. During the training phase, the sweat stress profile is identified over time in this controlled environment. Fig. Figure 3 is, in accordance with an illustrative embodiment, an example graph 300 of sweat over time in response to various stress-inducing factors for a patient. In some embodiments, an EEG is performed during this phase to calibrate the filters individually for the patient.

[0025] For example, a patient may perform a motor task while wearing a VR headband integrated with the EEG. The patient also wears a sweat sensor. At time 10, the amount of sweat increases above a predetermined threshold, resulting from controlled changes in the physiological environment (e.g., temperature increase, movement, etc.). At this time, the filter is turned ON and adaptively changed to reduce the error between the generated output and the actual output using a stepwise approach.

[0026] During the training phase, the patient's electrodermal sensitivity is determined (e.g., how sweat affects neurophysiological signal levels). Some patients may be strongly affected (e.g., their skull width may be comparatively small, or the amplitude of their neural signals may be comparatively small, which may affect the measured signal levels differently than individuals with a larger skull width and / or larger amplitudes of the observed neurophysiological signals). Furthermore, patients may differ from one another in terms of their personality and level of anxiety during the test. Thus, in one embodiment, a baseline value—sometimes referred to herein as a sweat stress profile—is determined for each patient and stored in the patient database 110.In this way, a data processing unit can determine how a particular patient's EEG response is affected by their sweat production.

[0027] In one embodiment, the appropriate filter to be used for the digital EEG data is determined during the training phase. For example, different types of filters such as Least Mean Square (LMS), Recursive Least Square (RLS), Kalman, etc., can be evaluated to determine the one or more that best match the patient's physiological response to sweat. In this context, it should be noted that RLS algorithms are faster to synthesize but are also typically more complex than LMS algorithms. If the sweat artifact is limited to a narrowband signal, an LMS algorithm may perform superiorly in terms of accuracy. Kalman filters can be more complex than RLS algorithms but also perform better because they do not require stationarity and can thus be implemented even when the artifacts to be removed fluctuate greatly over time.Finally, if the artifact signal is determined to be nonlinear, a nonlinear filter (particle filter, batch filter, denoising or stacked auto-encoder, etc.) can be used. Deep neural network-based models can be used once a sufficient amount of training data has been generated.

[0028] Although the functionality of the EEG artifact engine 103 has been discussed in connection with an EEG analysis server 116, with further reference to Fig. 1, in one embodiment, some or all of the functionality of the EEG artifact engine 103 may be contained within the user device (e.g., 102(1) or 102(N)) itself. In other words, a user device may provide the functionality of adaptive filtering of sweat artifacts during electronic brain monitoring independently of the EEG analysis server 116.

[0029] Although the patient database 110, the data processing unit (e.g., 102(1)), and the EEG analysis server 116 are shown on different platforms merely by way of example, it should be apparent that in various embodiments, the patient database 110, the EEG analysis server 116, and the user unit (e.g., 102(1)) may be combined in various combinations. In other embodiments, these data processing platforms may be implemented by virtual data processing units in the form of virtual machines or software containers deployed in a cloud, thereby providing an elastic architecture for processing and storage.

[0030] Each of the databases and data processing units discussed here complies with the U.S. Health Insurance Portability and Accountability Act (HIPAA), which regulates the protection of sensitive patient data. Example user unit

[0031] As in connection with Fig. As discussed in Figure 1, the adaptive filtering of sweat artifacts during the simultaneous activity of an electronic brain monitoring system in the architecture can have 100 different types of user units. For this purpose, Fig. 4 is a block diagram showing various components of an illustrative user unit 400 at a general level. For purposes of discussion, the diagram shows the user unit 400 in the form of a wireless data transmission unit that communicates via the network 106. Fig. 1 can exchange data.

[0032] User unit 400 may include one or more antennas 402; a transceiver 404 for cellular, WLAN, and / or wired data transmission; a user interface 406; one or more processors 408; hardware 410; and memory 416. In some embodiments, antennas 402 may include an uplink antenna that transmits radio signals to a base station and a downlink antenna that receives radio signals from the base station. In some other embodiments, a single antenna may both transmit and receive radio signals. The same or different antennas may be used for WLAN data transmission. These signals may be processed by transceiver 404, sometimes collectively referred to as a network interface configured to receive and transmit digital data.In one embodiment, the transceiver 404 additionally includes a short-range wireless data transmission capability such as DECT, NFC, ZigBee, Bluetooth, etc. In one embodiment, the user unit 400 does not include an antenna 402, and data transmission with external components occurs via a wired data transmission.

[0033] In one embodiment, user device 400 includes a user interface 406 that enables a user to provide input and receive output from user device 400. For example, user interface 406 may include a data output device (e.g., a visual display, speaker, haptic device, etc.) that may be used to provide notifications from EEG artifact engine 103 of EEG analysis server 116.

[0034] User interface 406 may also include one or more data input devices. The data input devices may include, but are not limited to, combinations of one or more of keypads, keyboards, mouse devices, touch-sensitive screens, microphones, speech recognition packages, and any other suitable devices or electronic / software selection interfaces. Data input devices may also include ports, such as USB ports, configured to connect to the wired electrodes discussed herein.

[0035] The user device 400 may include one or more processors 408, which may be a single-core processor, a multi-core processor, a CISC (Complex Instruction Set Computer) processor, or another type of processor.

[0036] Hardware 410 may include a power source and digital signal processors (DSPs), which may include single-core or multi-core processors. Hardware 410 may also include network processors that manage high-speed data transmission interfaces, such as data transmission interfaces that interact with peripheral components. The network processors and the peripheral components may be linked by a switching fabric. Hardware 410 may further include hardware decoders and encoders, analog-to-digital converters (ADCs), a network interface controller, and / or a USB controller. Furthermore, the hardware may include a temperature sensor, a humidity sensor, an accelerometer, and a global positioning system (GPS).For example, the temperature and humidity sensors can be used to determine the ambient temperature and humidity in the patient's environment. Furthermore, the accelerometer and GPS can be used to determine the patient's position and whether the patient is moving.

[0037] The memory 416 may be implemented using computer-readable media, such as computer storage media. Storage media includes volatile and non-volatile, removable and non-removable media, implemented using any method or technology for storing data, such as computer-readable instructions, data structures, program modules, or other data.Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disks (DVDs), high-definition video storage disks, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store data for access by a data processing device.

[0038] Memory 416 may store various software components or modules executable or accessible by the processor(s) 408 and controller of user device 400. The various components of memory 416 may include software 418 and an operating system 460. Software 418 may include various applications 420, such as an EEG analysis application 440 with multiple modules. Each module may include routines, program instructions, objects, components, and / or data structures that perform tasks or implement abstract data types.

[0039] For example, the EEG analysis application 440 of the user unit 400 may include a sweat module 442 operable to interpret the digital data from the wearable sweat sensor. In one embodiment, a wearable sweat sensor primarily measures the galvanic skin response (GSR), which may be performed by placing two electrodes a predetermined distance apart (e.g., one inch) on the body and / or scalp surface and measuring the recorded electrical resistance between the two electrodes when a very weak current is constantly passed through them. The galvanic skin potential (GSP) may be measured entirely without externally applied current between two electrodes if the electrodes are connected to a voltage amplifier. Both GSR and GSP vary with the emotional state of the subject (e.g.,Arousal), as well as body temperature and humidity. The wearable sweat sensor can be integrated into the EEG headband or worn on any other part of the body.

[0040] A neurophysiological module 444 may be present, operable to interpret the digital EEG data. A temperature module 446 may be present, operable to interpret the ambient temperature conditions provided by a temperature sensor. A humidity module 448 may be present, operable to determine the humidity in the patient's environment. In addition, an NA module 450, a K module 452, and a CI module 454 may be present, operable to determine the level of biomarkers present in sweat, such as sodium (Na + ), potassium (K + ) Chloride (Cl -), glucose or lactose, and the like from the digital sweat sensor data. A combination of biomarker content and GSR signal can be used to separate the emotional and cognitive components from the physiological components present in sweat secretion, which can then be used to correct the EEG signal.

[0041] In a particular embodiment, the EEG analysis application 440 includes an adaptive adaptation module 456 operable to apply a transformation to the sweat sensor data based on a pre-determined sweat stress profile of the patient. The EEG analysis application 440 then adaptively subtracts the transformed sweat sensor data from the digital EEG data.

[0042] Although an EEG analysis application 440 has been depicted as part of the user device 400, in some embodiments, the functions performed by the EEG analysis application 440 may be performed by the EEG artifact engine 103. For example, the functions associated with the adaptive adjustment module 456 may be performed by the EEG artifact engine 103 of the EEG analysis server 116. The operation of the EEG analysis server 116 is discussed in more detail below.

[0043] The operating system 460 may include components that enable the user device 400 to receive and send data via various interfaces (e.g., user controls, data transfer interface, and / or memory input / output devices), and to process data using the processor(s) 408 to generate output. The operating system 460 may include a presentation component that presents the output (e.g., displays the data on an electronic display of the user device 400, stores the data in the memory 416, transmits the data to another electronic device, etc.). In addition, the operating system 460 may include other components that perform various additional functions generally associated with an operating system. Example block diagram

[0044] Fig. 5 shows, in accordance with an illustrative embodiment, a general block diagram of a system 500 for adaptive filtering of sweat artifacts. The system 500 includes a summing node 510, which may represent the EEG analysis application 440 of a user device 400 or an EEG artifact engine 103. A patient's EEG signals are received from various electrodes. Block 502 represents a single channel that has been digitized. The patient's sweat sensor signals are received simultaneously. Block 504 represents the digitized sweat sensor data. The adaptive filter 506—sometimes referred to herein as a transform—receives the sweat sensor data 504 and adaptively adjusts the sweat sensor data based on the control signal received by the adaptive algorithm 508. The adaptive algorithm 508 is configured during a training phase in which the sweat stress profile is determined.The patient's sweat stress profile is used to generate the adaptive algorithm to adjust filter 506. Accordingly, the adaptive filter 506 generates a digital signal level for each sample of the digital EEG data. For each sample of the digital EEG data, summing node 510 subtracts the transformed sweat sensor data therefrom.

[0045] In one embodiment, the coefficients and / or hyperparameters of the adaptive filter 506 are determined during training, where the desired output is pre-calculated or estimated from other environmental variables and / or data from other sensors (e.g., temperature sensor, humidity sensor, etc.). The error function—sometimes referred to herein as the error signal—is estimated from the difference between the desired output and the actual output. The coefficients and / or hyperparameters are optimized so that the error function is minimized. The trained adaptive algorithm 508 is then used to remove the artifact from the EEG test signal. In some embodiments, the adaptive filter 506 is further adjusted based on environmental conditions during patient assessment. Example process

[0046] Following the above overview of the example architecture 100 and the example block diagram 500, it may be helpful to provide a general discussion of an example process. For this purpose, the Fig. 6 and Fig. 7 illustrates processes 600 and 700 for detecting, identifying, and removing artifacts in neurophysiological signals during electronic brain monitoring of a patient. Specifically, Fig. 6 an example process of an initial training phase, where a sweat stress profile is individually tailored to a patient, and Fig. Figure 7 illustrates an example process for an active phase where sweat-related artifacts are identified and removed from EEG data.

[0047] The processes 600 and 700 are illustrated as collections of blocks in logical flowcharts, which represent sequences of operations that can be implemented in hardware, software, or a combination thereof. In the context of software, the blocks represent computer-executable instructions that, when executed by one or more processors, perform the stated operations. In general, computer-executable instructions may include routines, programs, objects, components, data structures, and the like that perform functions or implement abstract data types. The order in which the operations are described is not intended to be limiting, and any number of the described blocks may be combined in any order and / or performed in parallel to implement the process. For purposes of discussion, the processes 600 and 700 are described with respect to the architecture 100 of Fig. 1 described.

[0048] In block 602, a patient is placed in a controlled environment to create a sweat stress profile typical for the patient. To this end, the controlled environment may involve the patient wearing a VR or AR interface to accelerate physical exertion and response to stress-inducing audio, video, and physical factors. In some embodiments, the ambient temperature and / or humidity of the room are controlled to identify how the patient responds to the introduced stress-inducing factors.

[0049] In block 604, analog EEG signals from the patient are received from one or more electrodes by a data processing unit. As discussed above, in various embodiments, the data processing unit may be a user device, such as user device 102(1) or EEG analysis server 116. These analog signals are converted into digital EEG training data.

[0050] In block 606, analog sweat sensor signals from the patient that are contemporaneous with the analog EEG signals are received by the data processing unit. For example, the sweat sensor signals are contemporaneous with the EEG signals in that the sweat sensor signals are synchronous with the EEG signals. These analog sweat sensor signals are converted into digital sweat sensor training data. In some embodiments, different time periods of this training phase expose the patient in question to different stress-inducing factors. For example, during a time period 0 < T < 5 min, soothing music is played, cheerful scenes are displayed in the VR or AR interface, and pleasant environmental conditions are maintained; during the time period 5 min ≤ T < 10 min, the environmental conditions may be deteriorated (e.g.,higher temperature and / or humidity); during the time period 10 min ≤ T < 15 min, the VR or AR interface may display stressful scenes to accelerate physical exertion and response to stress-inducing aural, visual, and physical factors; etc.

[0051] In block 610, the digital EEG training data, the digital sweat sensor training data, and the stress-inducing factors are correlated.

[0052] In block 612, a sweat stress profile is created for the patient in question based on the correlation performed in block 610. In various embodiments, the sweat stress profile for the patient in question can be stored in the user unit (e.g., 102(1)) and / or sent over the network 106 to be stored in the patient database 110.

[0053] With reference to Fig. 7, an example process 700 for an active phase is illustrated, wherein sweat-related artifacts are identified and removed from EEG data by a data processing unit. The data processing unit may be a user unit such as user unit 102(1) or EEG analysis server 116. In various embodiments, process 700 may be independent of process 600 or performed after process 600 (i.e., to use the sweat stress profile from process 600). Accordingly, in block 702, the sweat stress profile of the subject patient is received. In various embodiments, the sweat stress profile may be received via network 106 from patient database 110 or directly from the user unit (e.g., 102(1)).

[0054] In block 704, analog EEG signals from the patient are received from one or more electrodes by a data processing unit, which in various embodiments may be a user unit (e.g., 102(1)) or the EEG analysis server 116. These analog signals are converted into digital EEG data.

[0055] In block 706, the data processing unit receives analog sweat sensor signals from the patient, which are synchronous with the analog EEG signals. These analog sweat sensor signals are converted into digital sweat sensor data.

[0056] In block 708, in one embodiment, data regarding environmental conditions is received by the data processing unit. The environmental condition data may include, for example, but is not limited to, (i) ambient temperature, (ii) ambient humidity, (iii) ambient airflow, etc., from one or more environmental sensors proximate the subject patient.

[0057] In block 710, a transformation is applied to the digital welding sensor data. In one embodiment, the transformation is based on the previously determined welding stress profile received by the data processing unit (e.g., in block 702). In some embodiments (e.g., where the ambient conditions were measured in block 708), the ambient conditions also serve as the basis for the transformation. Accordingly, block 710 provides transformed digital welding sensor data that can be used as a reference for filtering.

[0058] In block 712, the transformed digital sweat sensor data is subtracted from the digital EEG data, thereby removing the sweat-related artifact from the digital EEG data. Accordingly, the data processing unit adaptively adjusts the digital EEG data. In this way, an appropriate amount is subtracted from the digital EEG data for each sample to remove sweat artifacts.

[0059] In block 714, the adaptively adjusted digital EEG data is displayed on a user interface of the user unit (e.g., 102(1)). The display of the adjusted digital EEG data may occur in real time. Example computer platform

[0060] As discussed above, functions related to the detection, identification, and removal of artifacts in neurophysiological signals may be performed using one or more data processing units connected via a wireless or wired data transmission for data transmission, as in Fig. 1, and in accordance with the process 600 of Fig. 6 and process 700 from Fig. 7 is, Fig. Figure 8 is a functional block diagram of a computer hardware platform that can communicate with various networked components such as patient user units 102(1) to 102(N), patient databases such as database 110, etc. In particular, Fig. 8 a network or host computer platform 800, such as can be used to implement a server such as the EEG analysis server 116 from Fig. 1 to be realized.

[0061] The computer platform 800 may include a central processing unit (CPU) 804, a hard disk drive (HDD) 806, a random access memory (RAM) and / or a read-only memory (ROM) 808, a keyboard 810, a mouse 812, a display 814, and a data communication interface 816 connected to a system bus 802.

[0062] In one embodiment, the HDD 806 has capabilities including storing a program that can perform various processes in a manner described herein, such as the EEG artifact engine 840. The EEG artifact engine 840 may include various modules configured to perform various functions. As discussed in connection with the discussion of the user unit 400 of Fig.As mentioned in Figure 4, in various embodiments, some or all of the functions described below may be performed by the user unit 400 itself. In other words, the EEG analysis application 440 of the user unit 400 may perform the functionality of the EEG artifact engine 840 discussed below.

[0063] For example, the EEG artifact engine 840 may include a sweat module 842 operable to interpret the digital sweat sensor data received from the user unit (e.g., 102(1)). There may be a neurophysiology module 844 operable to interpret the digital EEG data received by the user unit (e.g., 102(1)). There may be a temperature module 846 operable to interpret the ambient temperature conditions provided by a temperature sensor. There may be a humidity module 848 operable to determine the humidity in the patient's environment. In addition, there may be an NA module 850, a K module 852, and a Cl module 854 operable to calculate the Na + -, K + - or Cl --content of the digital sweat sensor data. Additional modules (not shown) may be available to measure additional parameters such as glucose, lactose, etc.

[0064] In one embodiment, a presentation module 858 is provided, operable to present the adaptively adjusted digital EEG data to a suitable recipient, e.g., a responsible caregiver, a physician, or the patient themselves. For example, the adaptively adjusted digital EEG data may be sent over the network 106 to a patient's user unit for display on a user interface of the unit. Presentation of the adjusted digital EEG data may occur in real time.

[0065] In one embodiment, a program such as Apache™ may be stored as a web server for operating the system. In one embodiment, the HDD 806 may store an executing application that includes one or more library software modules, such as those for the Java™ Runtime Environment program for implementing a JVM (Java™ Virtual Machine). conclusion

[0066] The descriptions of the various embodiments of the present teachings have been presented for illustrative purposes and are not intended to be exhaustive or limited to the disclosed embodiments. Those skilled in the art will appreciate that numerous changes and modifications may be made without departing from the spirit and scope of the described embodiments. The terminology used herein has been chosen to best explain the principles of the embodiments, practical application, or technical improvement over commercially available technology, or to enable others skilled in the art to understand the embodiments disclosed herein.

[0067] While the above describes the best mode contemplated and / or other examples, it should be understood that various changes may be made therein, and that the subject matter disclosed herein may be embodied in various forms and examples, as well as that the teachings may be applied to numerous applications, only a few of which have been described herein. It is intended that the following claims claim all applications, changes, and modifications that fall within the true spirit and scope of the present teachings.

[0068] The components, steps, features, objects, benefits, and advantages discussed herein are for illustrative purposes only. Neither they nor any discussions related to them are intended to limit the scope. While various advantages have been discussed herein, it should be understood that not all embodiments necessarily incorporate all of the advantages. Unless otherwise stated, all measurements, values, ratings, positions, magnitudes, sizes, and other specifications recited in this description, including the following claims, are approximate, not exact. Rather, they are intended to have a reasonable range consistent with the functions to which they relate and with what is conventional in the related art.

[0069] Numerous other embodiments are also conceivable. These include applications with fewer, additional, and / or different components, steps, features, objects, benefits, and advantages. They also include implementations in which the components and / or steps are arranged and / or ordered in a different manner.

[0070] Aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, as well as combinations of blocks in the flowcharts and / or block diagrams, may be implemented by computer-readable program instructions.

[0071] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine such that the instructions, executed by the processor of the computer or other programmable data processing device, produce a means for implementing the functions / steps defined in the flowchart block(s) and / or block diagrams.These computer-readable program instructions may also be stored on a computer-readable storage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a manner such that the computer-readable storage medium having instructions stored thereon comprises an article of manufacture, including instructions that implement aspects of the function / step specified in the flowchart block(s) and / or block diagrams.

[0072] The computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of process steps to be performed on the computer, other programmable apparatus, or other device to produce a computer-implemented process such that the instructions executing on the computer, other programmable apparatus, or other device implement the functions / steps defined in the block(s) of flowcharts and / or block diagrams.

[0073] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams may represent a module, segment, or portion of instructions comprising one or more executable instructions for performing the particular logical function(s). In some alternative implementations, the functions indicated in the block may occur in a different order than shown in the figures. For example, two blocks shown in succession may actually execute substantially concurrently, or the blocks may sometimes execute in reverse order depending on the corresponding functionality.It is further understood that each block of the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by special purpose hardware-based systems that perform the specified functions or steps, or by combinations of special purpose hardware and computer instructions.

[0074] Although the foregoing has been described in connection with exemplary embodiments, it should be understood that the term "exemplary" is intended merely as an example and not as the best or optimal. Unless explicitly stated in the foregoing, no statement or illustration is intended or should be interpreted as imparting to the public any purpose of any component, step, feature, object, advantage, benefit, or equivalent, regardless of whether such purpose is so recited in the claims.

[0075] Rather, it should be apparent that the terms and expressions used herein have the usual meanings assigned to such terms and expressions in their respective fields of research and science, except where different specific meanings have been defined herein. Relative terms such as "first" and "second" and the like may be used merely to distinguish one entity or action from another, without necessarily requiring or implying any actual such relationship or sequence between such entities or actions. The terms "comprises," "comprising," and any other variations thereof are intended to express a non-exclusive inclusion, such that a process, method, article, or apparatus comprisingthat has a list of elements, not only includes those elements, but may also include other elements not expressly listed or inherent in such process, procedure, article, or device. An element preceded by "a," and derivatives thereof, does not preclude, without further limitation, the presence of additional, identical elements in the process, procedure, article, or device from which it is composed.

[0076] The Summary of Disclosure is provided to enable the reader to quickly examine the nature of the technical disclosure. It is understood, however, that it is not intended to interpret or limit the scope or meaning of the claims. Moreover, it is apparent from the foregoing Detailed Description that various features are grouped together in various embodiments in order to streamline the disclosure. This method of disclosure does not imply that the claimed embodiments are intended to have more features than are expressly recited in each claim. Rather, as the following claims make clear, inventive subject matter lies in fewer than all of the features of a single disclosed embodiment.Thus, the following claims form a part of the detailed description, with each claim standing on its own as separately claimed subject matter.

Claims

[1] Data processing unit comprising: a processor; a network interface connected to the processor to enable data transmission over a network; a memory unit connected to the processor; software stored in the storage unit, wherein execution of the software by the processor configures the data processing unit to perform actions comprising: receiving digital EEG data based on digitizing analog electroencephalogram (EEG) signals of a patient from a first electrode; receiving digital sweat sensor data based on digitizing analog sweat sensor signals of the patient that are synchronous with the analog EEG signals; applying a transformation to the digital sweat sensor data based on a previously determined sweat stress profile of the patient; and an adaptive adjustment of the digital EEG data by subtracting the transformed sweat sensor data from the digital EEG data. [2] Data processing unit according to claim 1, wherein the adaptive adjustment of the digital EEG data occurs in real time with the reception of the analog EEG and sweat sensor signals. [3] A data processing unit according to claim 1, wherein the sweat sensor data is sodium (Na + ) measurement data, potassium (K + ) measurement data and chlorine (Cl - ) measurement data. [4] The data processing unit of claim 1, wherein the analog sweat sensor signals are based on a galvanic skin response by measuring an electrical resistance between the first electrode and a second electrode when a current flows between the first electrode and the second electrode. [5] Data processing unit according to claim 1, wherein: the first electrode is part of a group of a plurality of electrodes configured to receive analog EEG signals from different parts of a scalp of the patient; and analog welding sensor signals are received from a welding sensor electrode for one or more electrodes of the plurality of electrodes that are within a predetermined distance from the welding sensor. [6] The data processing unit of claim 1, wherein execution of the software by the processor further configures the data processing unit to perform actions comprising: performing an adjustment of the distance regression of the transformation based on a distance between the first electrode and a weld sensor electrode before applying the transformation to the digital weld sensor data. [7] The data processing unit of claim 1, wherein execution of the software by the processor further configures the data processing unit to perform actions comprising: receiving environmental data of an environment of the patient comprising at least one of: (i) an ambient temperature, (ii) an ambient humidity, and (iii) an ambient airflow from one or more environmental sensors, wherein applying a transformation to the digital sweat sensor data is further performed based on the received environmental data. [8] Data processing unit according to claim 1, wherein: the patient’s sweat stress profile is determined in advance during a training phase; and execution of the software by the processor further configures the data processing unit to perform actions comprising: introducing stress-inducing factors for the patient in a controlled environment; receiving analog EEG training signals of a patient from the first electrode in response to the stress-inducing factors and converting the analog EEG training signals into digital EEG training data; receiving analog sweat sensor signals from the patient that are synchronous with the analog EEG training signals in response to the stress-inducing factors and converting the analog sweat sensor training signals into digital sweat sensor data; and Creating a sweat stress profile of the patient based on the digital EEG training data and the digital sweat sensor training data. [9] The computing device of claim 8, wherein the stress-inducing factors are introduced via at least one of: (i) a virtual reality (VR) interface of the computing device, (ii) an augmented reality (AR) interface of the computing device, (iii) a control of an ambient temperature in a controlled environment, (iv) a control of an ambient humidity of the controlled environment, and (v) a control of an ambient air flow of the controlled environment. [10] The data processing unit according to claim 1, wherein the analog EEG signals and the analog sweat sensor signals are received wirelessly and in real time by the data processing unit via a short-range wireless data transmission technology. [11] A non-transitory computer-readable storage medium, specifically containing computer-readable program code comprising computer-readable instructions which, when executed, cause a computer unit to perform a method for removing artifacts in electroencephalogram (EEG) signals, the method comprising: receiving digital EEG data based on digitizing analog EEG signals of a patient from a first electrode; receiving digital sweat sensor data based on digitizing analog sweat sensor signals of the patient that are synchronous with the analog EEG signals; applying a transformation to the digital sweat sensor data based on a previously determined sweat stress profile of the patient; and an adaptive adjustment of the digital EEG data by subtracting the transformed sweat sensor data from the digital EEG data. [12] The non-transitory computer-readable storage medium of claim 11, wherein the adaptive adjustment of the digital EEG data occurs in real time with the receipt of the analog EEG and sweat sensor signals. [13] A non-volatile computer-readable storage medium according to claim 11, wherein the sweat sensor data comprises sodium (Na + ) measurement data, potassium (K + ) measurement data and chlorine (Cl - ) measurement data. [14] The non-volatile computer-readable storage medium of claim 11, wherein the analog sweat sensor signals are based on a galvanic skin response by measuring an electrical resistance between the first electrode and a second electrode when a current flows between the first electrode and the second electrode. [15] A non-transitory computer-readable storage medium according to claim 11, wherein: the first electrode is part of a group of a plurality of electrodes configured to receive analog EEG signals from different parts of a scalp of the patient; and analog welding sensor signals are received from a welding sensor electrode for one or more electrodes of the plurality of electrodes that are within a predetermined distance from the welding sensor. [16] The non-transitory computer-readable storage medium of claim 11, further comprising performing an adjustment of the distance regression of the transform based on a distance between the first electrode and a weld sensor electrode before applying the transform to the digital weld sensor data. [17] The non-transitory computer-readable storage medium of claim 11, further comprising receiving environmental data of an environment of the patient comprising at least one of: (i) an ambient temperature, (ii) an ambient humidity, and (iii) an ambient airflow from one or more environmental sensors, wherein applying a transformation to the digital sweat sensor data is further based on the received environmental data. [18] The non-transitory computer-readable storage medium of claim 11, wherein determining the sweat stress profile during a training phase comprises: introducing stress-inducing factors for the patient in a controlled environment; receiving analog EEG training signals of a patient from the first electrode in response to the stress-inducing factors and converting the analog EEG training signals into digital EEG training data; receiving analog sweat sensor signals from the patient that are synchronous with the analog EEG training signals in response to the stress-inducing factors and converting the analog sweat sensor training signals into digital sweat sensor data; and Creating a sweat stress profile of the patient based on the digital EEG training data and the digital sweat sensor training data. [19] A non-transitory computer-readable storage medium according to claim 18, wherein the stress-inducing factors are introduced via at least one of: (i) a virtual reality (VR) interface of the computing unit, (ii) an augmented reality (AR) interface of the computing unit, (iii) a control of an ambient temperature in a controlled environment, (iv) a control of an ambient humidity of the controlled environment, and (v) a control of an ambient air flow of the controlled environment. [20] The non-transitory computer-readable storage medium of claim 11, wherein the analog EEG signals and the analog sweat sensor signals are received wirelessly and in real time by the computer unit via a short-range wireless data transmission technology. [21] A method for removing artifacts in electroencephalogram (EEG) signals, the method comprising: receiving digital EEG data based on digitizing analog EEG signals of a patient from a first electrode; receiving digital sweat sensor data based on digitizing analog sweat sensor signals of the patient that are synchronous with the analog EEG signals; applying a transformation to the digital sweat sensor data based on a previously determined sweat stress profile of the patient; and an adaptive adjustment of the digital EEG data by subtracting the transformed sweat sensor data from the digital EEG data. [22] The method of claim 21, wherein the adaptive adjustment of the digital EEG data occurs in real time with the reception of the analog EEG and sweat sensor signals. [23] The method of claim 21, wherein the sweat sensor data is sodium (Na + ) measurement data, potassium (K + ) measurement data and chlorine (Cl - ) measurement data. [24] The method of claim 21, wherein the analog sweat sensor signals are based on a galvanic skin response by measuring an electrical resistance between the first electrode and a second electrode when a current flows between the first electrode and the second electrode. [25] The method of claim 21, wherein: the first electrode is part of a group of a plurality of electrodes configured to receive analog EEG signals from different parts of a scalp of the patient; and analog welding sensor signals are received from a welding sensor electrode for one or more electrodes of the plurality of electrodes that are within a predetermined distance from the welding sensor. [26] The method of claim 21, further comprising performing an adjustment of the distance regression of the transformation based on a distance between the first electrode and a weld sensor electrode before applying the transformation to the digital weld sensor data. [27] The method of claim 21, further comprising receiving environmental data of an environment of the patient, comprising at least one of: (i) an ambient temperature, (ii) an ambient humidity, and (iii) an ambient airflow from one or more environmental sensors, wherein applying a transformation to the digital sweat sensor data is further based on the received environmental data. [28] The method of claim 21, wherein determining the sweat stress profile during a training phase comprises: introducing stress-inducing factors for the patient in a controlled environment; receiving analog EEG training signals of a patient from the first electrode in response to the stress-inducing factors and converting the analog EEG training signals into digital EEG training data; receiving analog sweat sensor signals from the patient that are synchronous with the analog EEG training signals in response to the stress-inducing factors and converting the analog sweat sensor training signals into digital sweat sensor data; and Creating a sweat stress profile of the patient based on the digital EEG training data and the digital sweat sensor training data. [29] The method of claim 28, wherein the stress-inducing factors are introduced via at least one of: (i) a virtual reality (VR) interface of the computing unit, (ii) an augmented reality (AR) interface of the computing unit, (iii) a control of an ambient temperature in a controlled environment, (iv) a control of an ambient humidity of the controlled environment, and (v) a control of an ambient air flow of the controlled environment. [30] The method of claim 21, wherein the analog EEG signals and the analog sweat sensor signals are received wirelessly and in real time by the computer unit via a short-range wireless data transmission technology. [31] A computer program comprising computer program code stored on a computer-readable medium for, when loaded into and executed on a computer system, causing the computer system to perform all the steps of the method according to any one of claims 21 and 30.

Images