Addressable serial electrode arrays for neural stimulation and / or recording applications, and wearable patch systems with on-board motion sensing and magnetically attached disposables for rehabilitation and physical therapy applications.

Abstract

The present disclosure provides devices, systems, and methods for physical therapy. Electrical signals are applied to a subject through electrodes of a device attached to the body part to be treated. The device can be operated remotely. The device can use artificial intelligence to evaluate and design a physical therapy routine using the electrodes.

Description

[Technical field]

[0001] cross reference This application claims the benefit of U.S. Provisional Application No. 63 / 328,349, filed April 7, 2022, which is incorporated by reference herein in its entirety. [Background technology]

[0002] Neurostimulation technology uses invasive or non-invasive approaches to apply electromagnetic energy to anatomical targets and induce neuromodulation of corresponding neural circuits for neurorehabilitation and other applications. However, many neurostimulation systems have a low number of electrodes, limiting the number and precision with which muscles and neural targets can be activated, and are often not portable or wearable.

[0003] Wiring many electrodes onto a transcutaneous device for neural or electrical stimulation and recording is difficult. Electrodes take up a lot of space. Routing numerous wires to the intended stimulation site can be difficult and costly. For wearable and transcutaneous applications, multi-layer flexible circuit boards pose a risk of reduced flexibility. Summary of the Invention

[0004] In some embodiments, the invention provides a device, the device comprising a first article comprising: 1) a first top surface, the first top surface comprising a set of first connectors, and 2) a first bottom surface, the first bottom surface comprising a plurality of electrodes; and a second article comprising: 1) a circuit board operatively connected to the plurality of electrodes, 2) a second top surface, the second top surface comprising a plurality of visualization aids, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes, and 3) a second bottom surface, the second bottom surface comprising a set of second connectors. and a second article having a second bottom surface, each of the first connectors independently configured to couple to one of the second connectors, the first connector and the second connector configured to form a connection that holds the first article and the second article together when the first connector is coupled to the second connector, and when the first article and the second article are operably connected, the first article and the second article together form a wearable, the wearable having a wearable size and a wearable shape adapted to fit a human body part.

[0005] In some embodiments, the invention provides a device, the device comprising: a) a first article, the first article being disposable, the first article comprising: 1) a bottom layer comprising A) a hydrogel, and B) a plurality of electrodes in contact with the hydrogel, the plurality of electrodes being connected in sequence, the plurality of electrodes being flexible, the plurality of electrodes being silver ink, the plurality of electrodes being configured to stimulate the muscle tissue and the neural target upon application of electrical stimulation to the muscle tissue and the neural target; 2) a middle layer, the middle layer comprising a dielectric material, a) a middle layer stacked on top of, in contact with, and operably connected to the bottom layer; and 3) a top layer, wherein the top layer comprises a set of first connectors, the first connectors being magnetic, the top layer stacked on top of, in contact with, and operably connected to the middle layer, the bottom layer having a bottom layer shape, the middle layer having a middle layer shape, the top layer having a top layer shape, the bottom layer and middle layer are operably connected, and when the middle layer and top layer are operably connected, the bottom layer shape, the middle layer shape, and the top layer shape substantially overlap; and b) a second article, the second article being durable, the second article comprising: 1) a second top surface, the second top surface comprising a plurality of visualization aids, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes, the plurality of visualization aids being LEDs, each visualization aid being configured to provide a visible signal when an electrode corresponding to the visualization aid emits an electrical signal; 2) a flexible circuit board comprising conductors, the flexible circuit board operably connected to the plurality of electrodes; and 3) a second bottom surface, , a second article comprising a second bottom surface, the second bottom surface comprising a set of second connectors, the second connectors being magnetic, and each of the first connectors being independently configured to couple to one of the second connectors; and c) a control system operably connected to the second article, the control system comprising: 1) a power source; 2) a processor operably connected to the power source and configured to operate the device; 3) a stimulator operably connected to the processor, operably connected to the power source and configured to deliver electrical stimulation to the flexible circuit board; 4) a control system operably connected to the processor;and 5) a memory system operatively connected to the processor and configured to record information regarding use of the device. 6) a transmitter operatively connected to the processor and configured to wirelessly transmit to a receiver external to the device a record of stimuli applied to a human body part in physical contact with the device and movements of the human body part in response to the stimuli. The device further comprises a control system including a power source, a processor, a stimulator, a wireless receiver, a memory system, and a transmitter operatively connected to the processor and configured to wirelessly transmit to a receiver external to the device a record of stimuli applied to a human body part in physical contact with the device and movements of the human body part in response to the stimuli. the stem, the transmitter, and the motion detector are in a common housing, the first connector and the second connector are configured to form a connection that holds the first article and the second article together when the first connector is mated to the second connector, when the first article and the second article are operably connected, the first article and the second article together form a wearable, the wearable having a wearable size and a wearable shape adapted to fit a human body part, the first article has a first article shape and the second article has a second article shape, when the first article and the second article are operably connected, the first article shape and the second article shape substantially overlap, the bottom layer is at least 25 cm; 2 has a surface area of

[0006] In some embodiments, the invention provides a method, the method including: a) contacting a body part of a human subject with a plurality of electrodes, the plurality of electrodes operably connected to a flexible circuit board, and the body part having a shape; b) manipulating the flexible circuit board to substantially conform to the shape of the body part; c) operably connecting a plurality of visualization aids to the flexible circuit board, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes; and d) passing an electric current through the plurality of electrodes, the plurality of electrodes providing an electrical stimulus to the body part.

[0007] In some embodiments, the invention provides a method, the method including: a) contacting a body part of a human subject with a plurality of electrodes, the plurality of electrodes operably connected to a flexible circuit board, and the body part having a shape; b) manipulating the flexible circuit board to substantially conform to the shape of the body part; c) receiving instructions from a wireless user device to deliver electrical stimulation to the flexible circuit board; and d) selecting at least a portion of the plurality of electrodes and passing a current through a portion of the plurality of electrodes, the portion of the plurality of electrodes providing the electrical stimulation to the body part.

[0008] In some embodiments, the invention provides an addressable electrode system comprising an article comprising a power source; (i) a plurality of sequentially connected electrodes; (ii) a plurality of indicators; and (iii) a plurality of switches, each switch independently comprising a first terminal, a second terminal, and a third terminal, wherein the first terminal of the switch is electrically coupled to the power source, the second terminal of the switch is electrically coupled to one of the plurality of indicators, and the third terminal of the switch is electrically coupled to one of the plurality of sequentially connected electrodes, and the switches are configured to simultaneously supply power to the electrodes and the indicators when power is supplied to the first terminal of the switch by the power source.

[0009] Each patent, publication, and non-patent literature cited in this application is incorporated herein by reference in its entirety, as if each was individually incorporated by reference. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic diagram for implementing addressable electrodes for use in the wearable systems described herein. [Figure 2-1] FIG. 2 is a diagram illustrating a set of electrode configurations described herein. [Figure 2-2] FIG. 2 is a diagram illustrating a set of electrode configurations described herein. [Figure 2-3] FIG. 2 is a diagram illustrating a set of electrode configurations described herein. [Figure 2-4] FIG. 2 is a diagram illustrating a set of electrode configurations described herein. [Figure 2-5] FIG. 2 is a diagram illustrating a set of electrode configurations described herein. [Diagram 3] FIG. 1 illustrates a thin, lightweight, and flexible wearable system. [Figure 4] FIG. 1 illustrates a wearable medial forearm flexor system for nerve stimulation to induce finger and wrist flexion movements. [Diagram 5] FIG. 4 is another diagram of the wearable system of FIG. 3. [Figure 6] FIG. 4 is an exploded view of the wearable system of FIG. 3. [Figure 7] FIG. 1 illustrates a wireless control system and method for controlling neural stimulation via a wearable patch system described herein. [Figure 8] FIG. 2 is a side view of a second article (electronic patch system) and a first article (disposable system) described herein. [Figure 9] FIG. 1 is a diagram of a disposable system as described herein. [Figure 10] FIG. 1 illustrates an electrode array of a disposable system described herein. [Figure 11] FIG. 1 illustrates a wrist-worn embodiment of the wearable patch system described herein. [Figure 12] FIG. 2 illustrates a stimulation module as described herein. [Figure 13] FIG. 1 illustrates a disposable electrode array configured to be attached to a patient's forearm and thumb. [Figure 14] 14A-14E show a disposable electrode array that is attached to a patient's forearm and thumb. [Figure 15] 15A-15C show a durable / reusable electronic patch attached to a disposable electrode array. [Figure 16] 16A-16C illustrate fixation of an assembly including a durable / reusable patch and corresponding disposable electrode array attached to a patient's forearm. [Figure 17] FIG. 1 shows a stimulator module attached to a patient's wrist. [Figure 18] FIG. 13 shows the connections via cables to the stimulator, flexor and extensor patches, and disposable thumb electrode array. [Figure 19] FIG. 1 illustrates a graphical user interface (GUI) of an application that allows a user to control stimulation. [Figure 20] FIG. 1 illustrates a GUI of an application that allows a user to control manual stimulation. [Figure 21] FIG. 13 illustrates user selection of multiple locations on a patch for delivering electrical stimulation, with each location indicated by an LED. [Figure 22] FIG. 13 illustrates a GUI of an application that allows a user to control thumb stimulation. [Diagram 23] FIG. 13 shows a GUI of an application for executing a stimulation sequence. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Neurostimulation can use electrical impulses to cause muscle contraction or modulate nerves, spinal cord, or spinal roots. The device can generate and deliver the impulses via electrodes to the skin near the muscle and nerve target to be stimulated. The electrodes can be pads that attach to or contact the skin. The delivered impulses can induce action potentials in the nerve or cause the muscle to contract. Neurostimulation is used, for example, in rehabilitation (such as physical or occupational therapy), strength training, and evaluation of nerve function. For example, neurostimulation can be used to address muscle atrophy, improve muscle weakness, or modulate spinal pathways to facilitate rehabilitation of motor or sensory function.

[0012] Neurostimulation technologies provide good results in neurorehabilitation and physical therapy applications, including post-stroke and injury cases. These results include improved movement, sensation, and independence for people living with disabilities. Neurostimulation systems with fewer electrodes are limited in the number and precision with which muscles and neural targets can be activated, and are often not portable or wearable. Additionally, systems / devices that do not include on-board sensing or a microprocessor capable of running AI or machine learning algorithms to recognize user intent may not be intuitive or functional in assistive and rehabilitation applications. User-directed neurostimulation to facilitate goal-directed rehabilitation tasks is much more effective than passive stimulation.

[0013] Another challenge with neurostimulation devices and systems is that manually placing electrodes is difficult and time-consuming, which hinders the location and stimulation of small muscles, such as those controlling the hand and fingers, as well as dorsal root targets in the spinal cord, which are important for promoting plasticity and recovery.

[0014] The present disclosure provides systems and methods that solve the aforementioned problems in neurostimulation. Disclosed herein is a system that provides a wearable, user-driven device with dynamic (electronically configurable) electrical stimulation to a target body area. The system can be, for example, a thin, lightweight, flexible wearable patch system. Non-limiting examples of body areas treated with the system include feet, lower legs, upper legs, lower back, buttocks, lower back, upper back, abdomen, chest, shoulders, upper arms, forearms, wrists, hands, base of the head, head (e.g., crown, forehead), and spinal cord. An exemplary device or system includes one or more articles, for example, patches. The articles are placed, for example, on the back, front, outer, inner areas of the body area. The one or more articles in the device or system can include a set of electrodes (e.g., arranged in an array, sequence, or matrix) for contacting the skin and providing electrical stimulation to the underlying muscles. In some embodiments, the articles are communicatively coupled to a unit that provides an electrical signal. In some embodiments, the article is communicatively coupled to a microprocessor configured to select electrical signals that stimulate specific muscles, nerves, or spinal roots corresponding to specific regions of the body. In some embodiments, the systems and methods disclosed herein are used for electrical muscle stimulation (EMS), electrocorticography, or electroencephalography.

[0015] A device or system can include multiple electrodes, in some embodiments, the electrodes are daisy-chained (e.g., wired together in sequence) and communicatively coupled to an indicator or control system via a switch (e.g., a solid state relay).

[0016] In some embodiments, the device or system includes a machine learning software module. The machine learning software module uses one or more machine learning algorithms to analyze the movement of the body part being electrically stimulated. The machine learning algorithm can be trained through a process of collecting data related to the electrical signal delivered to the muscle and / or nerve target (e.g., spatial information regarding which electrode on the patch delivered the signal), the movement of the body part, or both, to predict the movement of the body part. Non-limiting examples of the movement include extension, flexion, angular movement (e.g., rotation), clenching (e.g., of a fist), and relaxation of one or more body parts. In some embodiments, the movement is an exercise (e.g., a press, curl, extension, squat, throw, or hip hinge).

[0017] In some embodiments, the wearable system is configured to interoperate with a brain-computer interface (BCI). A device (e.g., headgear) can derive electrical signals from the brain and provide the signals to the wearable system. The wearable system can include a processor configured to execute instructions based at least in part on the signals to stimulate specific electrodes. An algorithm (e.g., a machine learning algorithm) can interpret the signals from the brain to determine body movements associated with the signals. Based on this interpretation, a controller can drive the electrodes to stimulate these body movements. Subjects with illnesses, injuries, or conditions that prevent the brain from performing motor functions can use the wearable systems disclosed herein to restore these motor functions. Wearable Systems

[0018] The devices and systems disclosed herein include wearables, such as articles, patches, textiles, fabrics, and other wearable systems. In some embodiments, the wearable systems disclosed herein (e.g., wearable patch systems) can be configured to be incorporated into medical articles or other textile products, such as apparel (e.g., casts, wrappings, braces, bandages, gowns, shirts, pants, underwear, watches, bracelets, headbands, headgear, or athletic wear).

[0019] The wearable system can be configured to be worn by a subject. The subject can be a human subject or a non-human subject. For example, the subject can be a mammalian subject, such as a dog, cat, horse, cow, pig, sheep, goat, rabbit, elephant, camelid (e.g., camel, alpaca, or llama), mouse, hamster, guinea pig, ferret, buffalo, donkey, deer, or monkey. The wearable system can be worn by a bird or a reptile, such as a chicken, duck, goose, emu, eagle, alligator, or crocodile.

[0020] In some embodiments, the system comprises a first article configured to deliver an electrical signal to a body part and a second article configured to provide an indication as to where on the body the electrical signal is being delivered.

[0021] In some embodiments, the electrical signal is configured for neural stimulation (e.g., muscle tissue or neural target stimulation). For example, the electrical signal provided by the systems disclosed herein can induce muscle contraction. The electrical signal is provided by electrodes placed on the surface of the skin near the muscle to be stimulated. In some embodiments, the system provides therapy by repeating muscle contractions and improving blood flow to promote repair of damaged muscles. Muscle strength is also improved by the repeated cycles of contraction and relaxation. The systems disclosed herein can be used for many types of electrical muscle stimulation, including transcutaneous electrical nerve stimulation (TENS), electrical muscle stimulation (EMS), electrical stimulation for tissue repair (ESTR), interferential current (IFC), neuromuscular electrical stimulation (NMES), functional electrical stimulation (FES), spinal cord stimulation (SCS), and iontophoresis.

[0022] In some embodiments, the first article has a first article shape. The second article has a second article shape. When the first and second articles are operably connected, the first and second article shapes may substantially overlap. In some embodiments, the second article shape may be larger than the first article shape. The first and / or second article shapes may conform to at least a portion of a body part of a subject (e.g., a human subject). For example, the first and / or second article shapes may be bent, flexed, or folded to provide sufficient electrical and / or physical contact or coupling with the body part. The first article may be disposable. The second article may be durable. The first article may include a bottom layer, a middle layer, and a top layer. The middle layer may be stacked on top of, in contact with, and operably connected to the bottom layer. The top layer may be laminated above, in contact with, and operatively connected to the middle layer.

[0023] In some embodiments, the first article comprises a plurality of electrodes and a layer of hydrogel.

[0024] FIG. 3 illustrates a thin, lightweight, flexible wearable system described herein. The wearable system may include a first article and a second article, such as, for example, a patch or a textile, that may be physically and / or communicatively and / or electrically coupled. The second article may be positioned above the first article such that a portion of the second article contacts the subject's skin and the surface of the first article is visible. In some embodiments, the first article includes a disposable system. The disposable system includes, for example, a number of disposable magnetic connectors for coupling the first article to the second article. The magnetic conductor is conductive and thus may be configured to pass an electric current through the wearable system. Thus, the magnetic connector may provide both magnetic and electrical coupling. In some embodiments, the second article includes an electronic patch system.

[0025] The electronic patch system can include multiple patches. The patches can be made from darts of textile (e.g., fiber-based material such as fabric or cloth). The patches can include at least one indicator (or visualization aid). Multiple indicators can be disposed or arranged on the second article. The indicators can be disposed above (e.g., disposed on) or below the patches such that the indicators and patches correspond. The patches (and indicators) can be arranged in an array (e.g., a square or rectangular array, or any shape suitable for the intended body site). The indicators can include audio, visual, tactile, or audiovisual indicators, or combinations thereof. For example, in some embodiments, the indicators generate light and / or sound to alert the subject or user of where and when an electrical signal is being delivered to the subject's body via one or more electrodes of the wearable patch system. The indicators can include, for example, light emitting diodes (LEDs). The indicators of the second article can correspond to one or more electrodes of the second article. The indicators can generate a light or visible signal at the same time that the corresponding electrodes emit a signal. For example, one or more electrodes of a first article can be positioned directly beneath a particular indicator of a second article. When neural stimulation is simultaneously provided by multiple electrodes in different spatial regions of the patch system, the indicators in the corresponding regions can simultaneously provide audiovisual signals (e.g., light up). In this way, patterns of stimulation of muscle tissue and neural targets from different regions of the body underlying the patch can be displayed on the surface of the second article.

[0026] The electronic patch system may include a stimulator module or a control system. The stimulator module may control the delivery of electrical signals to the body part and may monitor and record data resulting from the neural stimulation caused by the delivered electrical signals. The stimulator module may include multiple modular components within a housing.

[0027] The stimulator module can include a processing unit or processor (e.g., a microprocessor). In some embodiments, the processor is external and operably coupled to the stimulator module. The processor can receive instructions (e.g., provided by a medical professional or automatically provided by a computer program, e.g., algorithmically) indicating a set of electrodes on a first article of the wearable system for providing neural stimulation to an area of ​​the body covered by the wearable system. The instructions can also include a corresponding set of indicators for providing audio, visual, tactile, audiovisual signals, or combinations thereof to a user (e.g., a subject or wearer of the patch, an administrator, or a medical professional). The processor can determine which patch (and indicator) corresponds to which electrode set, for example using personality resistors.

[0028] The processor can implement instructions configured to control the neural stimulation. For example, the control instructions can define which electrodes provide stimulation and when. The control instructions can provide a pattern of electrical stimulation. For example, the control instructions can instruct to provide electrical pulses to a particular muscle or neural target periodically (e.g., every 1 second, every 5 seconds, every 10 seconds, every 20 seconds, or every 30 seconds).

[0029] The processor can implement instructions that define characteristics of the electrical impulse. For example, the electrical impulse can have a frequency of 1-10 Hz, 11-20 Hz, 21-30 Hz, 31-40 Hz, 41-50 Hz, 51-60 Hz, or 61-70 Hz. The impulse duration can be 50-100 microseconds (μs), 100-150 μs, or 150-200 μs. The duration of the stimulus can be 0 s-5 s, 5 s-10 s, 10 s-20 s, or 20 s-30 s.

[0030] The instructions may be received from an external computing device (e.g., a mobile computing device) connected via a wired or wireless network. The network may be, for example, an Ethernet network, a Wi-Fi network, a Bluetooth network, etc.

[0031] The stimulator module can include a motion sensing unit, which can include an inertial measurement unit (IMU). The IMU can include an inertial sensing integrated circuit (IC) having, for example, six or nine degrees of freedom (DOF) for sensing different types of rotational motion (e.g., surge, heave, and sway) and translational motion (e.g., roll, pitch, and yaw) about a set of axes. The inertial measurement unit can include, for example, one or more accelerometers, gyroscopes, and / or magnetometers.

[0032] The stimulator module can include a machine learning software module that provides one or more machine learning models that can be configured to process measurements from the IMU, electrical signals provided via the electrodes, or instructions for providing electrical signals to predict movement of the part of the body to which the patch is attached. In some cases, the machine learning software module is external to the stimulator module. For example, the machine learning software module can be located on a server that is communicatively coupled to the stimulator module (e.g., via a wired or wireless network).

[0033] In some embodiments, the stimulator module (or control system) or processor is operably connected to a power source. In some embodiments, the power source and the stimulator module are in a common housing. The power source can include a battery. The power source can include an AC adapter. In some embodiments, the stimulator includes sufficient electrical components to draw voltage and power from a mains power source when plugged into a wall outlet.

[0034] The stimulator module may further include a memory system operatively connected to the processor and configured to record information regarding use of the wearable patch system. The memory system may further include one or more machine learning models executable by the processor on data recorded in the memory.

[0035] The stimulator module may further include a transmitter operatively connected to the processor and configured to wirelessly transmit information regarding use of the device to a recipient external to the device. The transmitter may transmit information via radio waves, such as over a Wi-Fi network or a Bluetooth network.

[0036] The disposable system can include a dielectric disposed over the plurality of electrodes. The dielectric can include any material that acts as an electrical insulator and can be polarized by an applied electric field, such as porcelain, glass, mica, metal oxides, plastics, etc. Exemplary plastics can include thermoplastic polymers such as polyvinyl chloride (PVC or vinyl) or polyethylene terephthalate (PET). The electrodes can be flexible electrodes. The electrodes can include pastes or inks made of one or more conductive materials or substances (e.g., metals such as silver or gold). In some embodiments, the electrodes are coated with carbon nanotubes.

[0037] 3 illustrates a device worn on a forearm. The exemplary device includes two sets of communicatively and physically coupled first and second items, the first set configured to be worn on the back of the forearm and the second set configured to be worn on the inside of the forearm. The wearable devices on the back and inside of the forearm can provide electrical stimulation to muscles and neural targets configured to induce wrist, hand, and finger movements.

[0038] FIG. 4 shows a wearable flexor system on the medial forearm for stimulating muscles and nerve targets that induce finger and wrist flexion type movements as described herein. Similar to FIG. 3, the embodiment of FIG. 4 includes an electronic patch system and a disposable system. The flexor patch system includes openings that allow for wrist pronation and supination. In some cases, the same wearable system can be worn on the opposite side of the arm to stimulate additional muscles and nerve targets that perform complex movements.

[0039]

[0013] Figure 5 shows another view of the wearable system of Figure 3. Figure 5 shows a first wearable system placed on the inside of the forearm and a second wearable system placed on the outside of the forearm. Figure 5 further shows the first wearable patch system, the second wearable patch system, and the stimulator module side-by-side.

[0040] FIG. 6 shows an exploded view of the wearable device described herein. FIG. 6 shows the second article (electronic patch system), the stimulator module, the first article (disposable system), and the cable. The disposable may have conductive paste or ink (e.g., silver ink) electrodes on one side and a magnetic connector on the other side. This arrangement can create a durable / reusable mechanical and electrical connection to the electronic patch system. The stimulator (wrist-worn) may include a cloth or woven strap that includes an adhesive unit such as Velcro or a magnetic snap. In some embodiments, a forearm strap (not shown) may be included that wraps around the device when worn on the user's arm. These straps can keep all the elements of the wearable patch system in good contact. These straps can be stretchy woven and can have, for example, Velcro or magnetic snap connections.

[0041] In some embodiments, at least two stimulators can be connected to multiple patches. For example, the wearable system can include a first wearable patch system, a second wearable patch, and a first stimulator module, as shown in FIG. 5. The first and second wearable patch systems are attached to the patient's forearm and controlled by the first stimulator module connected thereto. The wearable system can further include a third wearable patch system connected with the second stimulator module. For example, the third wearable patch can be small and can provide neural stimulation to the patient's spinal cord when attached to the patient's neck and controlled by the second stimulator. When the patient wears a patch on the forearm and a patch near the spinal cord, both generate neural stimulation, and the effectiveness of the stimulation can exceed that of the forearm muscle stimulation alone.

[0042] In some embodiments, a wireless control system can be used to control the neural stimulation via one or more stimulators. When multiple patches are connected to the stimulator, a user can use the wireless control system to control the neural stimulation. A user can send a wireless signal to the stimulator, which can control the delivery of electrical signals from designated patches to body parts and monitor and record data resulting from the neural stimulation caused by the delivered electrical signals. For example, the stimulator can control two forearm patches to generate neural stimulation in a given pattern and a cervical patch to generate another given pattern. The stimulator can control simultaneous stimulation from multiple patches. Alternatively, the stimulator can control stimulation from different patches to occur in any configuration (e.g., order or spatial arrangement) provided by the user. A user can save the configuration of neural stimulation generated from different patches and its pattern for future use without having to perform manual configuration repeatedly.

[0043] In some embodiments, the wearable patch can include a transmitter operably connected to a processor in the patch and configured to transmit wirelessly to other wearable patches. The transmitter can transmit information via radio waves, such as a Wi-Fi network or a Bluetooth network. Thus, the wearable patch can communicate with and control other patches. For example, the forearm patch and the neck patch can include a transmitter operably connected to a processor in the patch. When the forearm patch generates neural stimulation, the forearm patch can automatically control the neck patch to generate neural stimulation. The dual stimulation can significantly improve the effectiveness of motor or sensory rehabilitation and strength training. In addition, stimulating the spinal cord in this manner increases the plasticity and / or adaptability of the body, improving rehabilitation.

[0044] 7 shows a wireless control system and method for controlling neural stimulation via a wearable system. The communication system or mobile device can include an application for configuring, controlling, and / or monitoring the neural stimulation from the wearable patch system (e.g., during physical therapy, occupational therapy, exercise instruction, or medical rehabilitation) and collecting data related to the electrical signals delivered to the skin and the movement of the body part in response to the delivered electrical signals. The data can be provided to an external server, such as, for example, a cloud server, for further analysis (e.g., machine learning analysis, etc.).

[0045] The communication system (e.g., a mobile device) can be configured to provide an analysis of the performance of a human subject using the wearable system in therapy and determine a proposed therapy regimen based on the analysis. For example, the communication system can process or provide processing of the human subject's movement data to determine whether the subject is performing an exercise properly or to determine the person's ability to perform a complex movement or sequence of movements. The analysis can be performed using machine learning or statistical methods. Additionally, the communication system can analyze the movements of the human subject undergoing therapy induced by providing neural stimulation and compare them to movements performed by healthy individuals. The analyzed movement characteristics can indicate a particular condition requiring a particular therapy regimen. For example, the analysis can determine that a person is unable to effectively grasp an object or perform an exercise with a proper range of motion. The system can recommend a therapy to correct these conditions. The proposed therapy regimen can include activation of electrodes in a pattern determined to be likely to result in physical therapy improvement based on the analysis.

[0046] The application can be a desktop application or a mobile application (e.g., an ANDROID® application or an iOS® application). The application for configuring, controlling, and / or monitoring neurostimulation facilitates the configuration, control, or monitoring via a user interface. A non-limiting example of a user interface is a graphical user interface (GUI). In some embodiments, the GUI includes a layout. The layout can be two-dimensional. In some embodiments, the layout is three-dimensional (e.g., when multiple wearables are placed at different locations on the body and the electrodes can be addressed using a three-dimensional coordinate system rather than two-dimensional). The layout can include a grid. The grid can use Cartesian or polar coordinates. In some embodiments, the three-dimensional grid uses Cartesian, cylindrical, or spherical coordinates. A point on the grid can include two individual electrodes in the electrode arrangement. The electrode arrangement can be incorporated into or embedded in an article (e.g., a first article of a wearable system).

[0047] The electrical or neural stimulation impulses can be initiated by gestures provided to the UI. The gestures can include physical contact (e.g., with the subject's body or with an instrument or tool). The gestures can be generated by a human hand, or one or more fingers of a human hand. In some embodiments, the gestures can be generated by another body part. For example, if the subject is paralyzed or paraplegic, the gestures can be generated by one or both eye movements (including winking or blinking). The gestures can include touches. Touching a point (e.g., a grid) on the user interface of the UI can generate a control signal that can be used to contact the subject and provide neural or electrical stimulation to the subject via electrodes corresponding to the points on the user interface. The touches can include one or more taps. Tapping a point on the grid can generate an electrical impulse having the same duration as the tap. Repeated tapping of a point on the GUI can generate repeating impulses provided by the corresponding electrodes (e.g., as a time-series sequence). Tapping multiple points on the GUI can generate impulses that are spatially distributed on the grid. A touch also includes a press (e.g., of a finger or stylus) on the UI. An impulse may last as long as the duration of the press. A touch may include pressing against the surface of a device providing the UI while sliding or dragging a finger or stylus. Sliding or dragging a finger or stylus over a point on the UI stimulates or activates (e.g., momentarily) corresponding electrodes, generating an impulse. A gesture may include a combination of taps, presses, and drags (such as when multiple fingers or hands are used).

[0048] In one example, a user presses down at a location on the screen and then slides their finger across the screen (e.g., in a shape or pattern such as a line, circle, or figure eight). As the finger slides across the screen, corresponding electrodes are momentarily activated to mirror the pattern of finger movement. In this way, dynamic electrical stimulation is applied to the subject's body that mimics the finger movement, and the electrical impulses can be thought of as "moving" across the portion of the body covered by the electrode array.

[0049] A gesture can stimulate multiple electrodes simultaneously (e.g., in a spatial arrangement). For example, multiple fingers, digits, or an instrument (such as a stylus) can tap, press, slide, and / or drag multiple points on the UI.

[0050] Gestures can be performed or provided using input / output (I / O) devices such as a keyboard, a mouse (e.g., by clicking, or by clicking and dragging), a trackball, a microphone (e.g., for voice-controlled or speech-controlled neurostimulation), a motion capture device (e.g., a camera that can monitor gestures without physical contact with the device containing the UI), a joystick, or other I / O device. The UI can allow a user to enter a series of commands (e.g., a script) or upload a computer program containing instructions for neurostimulation by particular electrodes of the array.

[0051] The UI can provide control of multiple electrode arrays (e.g., incorporated into multiple wearable patches). The UI can achieve this control using a split screen, with each split presenting a grid containing one separate electrode array. In some embodiments, the UI can display multiple grids simultaneously (e.g., 2, 3, 4, 5, or 10 grids). In some embodiments, multiple UI windows can be displayed and interacted with on multiple screens or monitors.

[0052] A control signal can be provided to one or more stimulation modules. The control signal can be provided over a network (e.g., Bluetooth, Wi-Fi, or a wired network (e.g., Ethernet)). The control signal can encode information related to the type of gesture performed and / or spatial location information related to the gesture (e.g., to provide a spatial pattern of recommended neural stimulation impulses for the subject). The control signal, when executed by the processor, can provide this information as instructions to connect corresponding electrodes of the electrode array to power (e.g., via a power source such as a battery or an alternating current (A / C) adapter) using switches connected to and corresponding to the electrodes such that current is provided to the electrodes. For example, the instructions can place a first plurality of switches in an "on" position and a second (non-overlapping) plurality of switches in an "off" position, thereby causing the electrodes connected to the "on" switches to generate neural stimulation. In this manner, the current provided to the electrodes provides the subject with a spatial pattern of recommended neural stimulation impulses corresponding to the gesture provided to the UI.

[0053] Non-limiting examples of mobile devices or communication systems include mobile phones, smart phones, personal digital assistants (PDAs), laptop computers, desktop computers, and other types of computing devices.

[0054] The mobile device can be used to provide dynamic (e.g., electronically movable) stimulation to a subject wearing a wearable system. Dynamic stimulation can enable rapid motion point mapping on a body part (e.g., the forearm) using a touchpad or screen area of ​​the mobile device. This technique can include, for example, an operator (e.g., the subject wearing the patch system or a healthcare provider) sliding a finger or stylus over a touchscreen control object or area to change where stimulation occurs (i.e., which electrode) on the subject's body part on a selected patch. A particular section of the stylus or grid can be repeatedly tapped to provide repeated electrical impulses to a particular area covered by the wearable system. An indicator (e.g., LED) indicates the current stimulation (activated electrode) location on the electronic patch system of the wearable patch system.

[0055] The user interface allows for the impulses delivered to the subject to be altered. For example, the user interface allows for the alteration of the frequency of the impulses, the impulse duration, and / or the stimulus duration. These characteristics can be dynamically altered. For example, the impulse frequency and / or the impulse duration can be manually altered throughout the stimulus. In some cases, patterns for altering the impulse frequency and / or the impulse duration can be programmed using the user interface.

[0056] FIG. 8 shows a side view of the second article (electronic patch system) and the first article (disposable system) described herein. The second article has an alternating arrangement of indicators 802 (e.g., LEDs) and solid state relays 801. The LEDs are disposed on a flex circuit or printed circuit board (PCB) 806. The underside of layer 810 can be an electrical connection system including metal (e.g., gold) plated contacts 805 connected to magnets 803. The first article can include a vinyl surface 812 used as a dielectric, on which is disposed a foam having a ferromagnetic disk 808 and conductive adhesive 809, and a silver ink electrode 811 adhered to the skin with a hydrogel 820. In some embodiments, alternative adhesives and conductive materials for the electrodes can be used.

[0057] Removing hydrogel from metal electrodes is difficult, time consuming, and often leaves behind unwanted residue. The disclosed system can solve this problem using magnetic connections by creating a multi-layer disposable with a hydrogel layer (bottom / skin side), a silver ink that forms a flexible electrode, a dielectric (e.g., vinyl or polyethylene terephthalate (PET)) layer, and a magnetic (and electrical) connection layer on top for attachment to a durable electronic patch system. The disposable system can then be removed and discarded after a set number of uses without leaving any residue on the electronic patch system.

[0058] The electronic patch system can be made, for example, of a flexible circuit board structure with gold plated copper as the conductor for carrying the stimulation energy. The plating, for example nickel-copper-nickel, can be adhered to the flexible circuit conductor using, for example, conductive epoxy or conductive pressure sensitive adhesive (also known as conductive transfer tape or conductive double-sided tape, e.g., 3M 9711S).

[0059] FIG. 9 shows a diagram of a disposable system described herein. The disposable system can include a magnetic attachment and an electrode array. The magnetic attachment can automatically align the electronic patch system with the electrical contacts on the disposable electrode array. The electrode array can include a plurality of electrodes configured to deliver an electrical signal to human skin.

[0060] FIG. 10 shows an electrode array of the disposable system described herein. The electrode array can be formed to accommodate various body sites. The electrode array can also be conformed (e.g., by bending or folding) to a body part or part of a body part while maintaining mechanical contact between the body and each electrode (allowing the body part, muscle, or neural target under the electrode array to slide laterally as it flexes) and, in addition, maintaining electrical contact between the body and the electrode, which allows the transmission of stimulation current to the electrode through the conductive plating on the magnet in the magnetic attachment. FIG. 10 additionally shows the shape of the thumb electrode array, which can cover the thumb rotators as well as the lateral and oblique adductor muscles.

[0061] In some embodiments, the electrodes are arranged in a square or rectangular configuration. This configuration can be used for neuromuscular electrical stimulation (NMES) or transcutaneous electrical nerve stimulation (TENS). The stimulation module can support a wide variety of waveforms in both NMES or TENS modes.

[0062] 11 shows a wrist-worn embodiment of the wearable system. The wrist-worn embodiment can include, for example, a stimulator unit 1101, an electrode array with hydrogel 1102, and a wrist strap 1103.

[0063] FIG. 12 illustrates an example of a stimulation module. The stimulation module can include a microprocessor, an inertial motion sensor (IMU), a battery, accessible electrodes, and stimulation circuitry. While the embodiment of FIG. 12 is configured to be worn on the wrist, in other embodiments, the stimulation module is worn on other parts of the body. In some embodiments, the stimulation module is embedded in a patch. The wearable patch system can also be linked, for example wirelessly, to a brain-computer interface (BCI) system.

[0064] The stimulation modules are available for applications including upper limb stimulation, lower limb stimulation (both quadriceps and dorsiflexion), and cervical stimulation (base of the neck).

[0065] The microprocessor can also store stimulation sequences (e.g., a sequence of electrical pulses delivered from the electrodes to a muscle or neural target). A particular stimulation sequence may induce a particular movement sequence. For example, a stimulation pattern may cause the user to open their hand for a few seconds while the user positions themselves around an object. A stimulation pattern may then be applied that causes the user to close their hand around the object. This configuration allows a useful sequence to be triggered with a single command.

[0066] The microprocessor can handle stimulus multiplexing: electrical stimulation pulses can be nested or interlaced in the time domain to allow multiple muscles or nerve targets to be stimulated simultaneously, although the individual pulses will not be coincident.

[0067] (Addressable Electrodes) The devices and systems described herein include, for example, multiple electrodes. Features of these electrodes include, for example, wiring configurations such as daisy chain or sequential, and communicative coupling to indicators or control systems via switches (e.g., solid state relays). In some embodiments, the electrodes are configured to be placed on the body (e.g., in contact with the skin), or in some embodiments, are implantable.

[0068] FIG. 1 shows a schematic diagram for implementing the electrodes used in the systems described herein.

[0069] The electrodes or electrode pairs 101 can be daisy-chained (e.g., wired together in sequence) to reduce the amount of wiring required. Suitable electrode configurations include anodes and cathodes, multi-phase electrodes, and return electrodes. An electrode can be an anode or a cathode by doubling the circuit. The electrodes are connected to an indicator (e.g., an LED) or controller 103 using a switch 104 (e.g., a solid-state relay, a metal-oxide-semiconductor field effect transistor (MOSFET), or other switch topology). In some embodiments, the addressable indicator and / or controller 103 and switch 104 are integrated into a single integrated circuit (IC), further reducing the size and complexity of the architecture.

[0070] FIG. 2 shows a non-limiting example set of electrode configurations described herein. Electrodes can be incorporated into flexible circuit boards or textiles that are worn on body parts. One of the key challenges in textiles is routing a large number of conductive threads and / or densely spaced conductive threads. Using electrodes in textile form allows many electrodes to be daisy-chained with only a few conductive pathways. These configurations can be used for electrocorticography (ECoG), electrical muscle stimulation (EMS), or electroencephalography.

[0071] The electrodes can be electrically coupled to the stimulation module, for example, using wires, coiled wires, retractable wires, or within a fabric or textile. In some embodiments, a return or ground electrode is required to complete the electrode circuit via additional wires and separate electrodes. In some embodiments, a pair of electrodes includes a return electrode and a ground electrode integrated as a frame around the active (e.g., stimulating or recording) electrode. The frame can be, for example, a circular (201), rectangular, or square frame. The electrodes can be arranged in an array. In some cases, the electrode array can be arranged in a serpentine flex circuit, for example, a winding path 203 that includes a series of electrodes. In some embodiments, the electrodes are arranged within a rigid board in a pack / patch configuration (204b). In this configuration, one or more electrodes are placed under a puck, which is connected to another puck by a short flexible connector (e.g., a coiled wire, etc.) so that it can bend, flex, and stretch for wearable applications. In some embodiments, the electrodes contact the skin through a layer of hydrogel.

[0072] Daisy chains can be used with implanted electrodes to solve wiring problems for stimulation or biosignals or neural recording. For example, a set of stereo electroencephalography (SEEG) or deep brain stimulation (DBS) electrodes 205 can be wired in a daisy chain fashion. In addition, daisy chain arrays can be used with electrocorticography (ECoG) arrays 207a or strips 207b.

[0073] (Machine Learning) An adaptive controller containing machine learning algorithms (e.g., neural networks) can learn to associate specific stimulation patterns with specific body movements. The controller can then drive the stimulation electrodes in spatiotemporal patterns such that the subject mimics the hand or body movements (joint angles and force levels) made by a therapist, caregiver, or user (healthy side).

[0074] The wearable system can classify body movements (e.g., exercises or portions thereof, or functional movements or portions thereof) using a classification model. The classifier can be trained using data collected from the IMU of the wearable system to associate IMU measurements (e.g., accelerometer measurements, gyroscope measurements, and / or magnetometer measurements of translation (or displacement), rotation, velocity, acceleration, jerk, position, or other quantities) obtained while the wearable device is worn with specific movements of the body part on which the wearable device is worn. The classifier can be trained based on IMU measurements from a target subject or population of subjects (who may or may not be similar to the target subject, where similarity can be based on demographic information, medical information, or similarity of injury, disease, or health condition).

[0075] In some embodiments, the classifier is a neural network. The neural network can include a long short-term memory (LSTM) network. The LSTM can be used to analyze the movements of a subject undergoing rehabilitation or strength training. As the subject's condition improves, the LSTM becomes less involved in training or fine-tuning the machine learning model, as it may forget IMU measurements previously taken during training. In this way, the LSTM generates accurate predictions of the individual's movements.

[0076] In some embodiments, the machine learning model predicts movement by processing patterns of electrical impulses (e.g., provided during nerve stimulation or electrical muscle stimulation) and correlating the patterns with movement data (e.g., from an IMU). The machine learning model is trained to predict a particular movement produced by a particular pattern of electrical impulses. Additionally, the machine learning model can also be trained to predict a particular pattern of electrical impulses that produced an observed movement or set of movements. The adaptive controller can observe a set of movements performed by a therapist or other professional and generate a corresponding series of impulses such that the subject mimics the therapist's movements.

[0077] The machine learning model can further be trained using both the observed movement data (e.g., from the IMU) and the corresponding electrical impulses. For example, if the movements the subject produces from a set of electrical impulses deviate from those made by a healthy individual, the machine learning model can predict the health condition, disease, disorder, or injury afflicting the subject and prescribe a course of treatment. The machine learning model can be implemented while the subject is undergoing treatment to test the effectiveness of the treatment. For example, the IMU and electrical impulse data can be tested repeatedly over and over during successive treatment appointments. The machine learning model can predict whether the treatment will be successful, how long the treatment will take, or can be used to determine whether a new treatment is needed.

[0078] In addition to analyzing IMU data and / or electrical impulse data, the machine learning models described herein can analyze medical data (e.g., from an electronic health record (EHR)), demographic data, visual data (e.g., video of exercise performance), or other data.

[0079] Training Phase The machine learning software modules described herein can be configured to undergo at least one training phase, in which the machine learning software modules are trained to perform one or more tasks including data extraction, data analysis, and generation of output.

[0080] In some embodiments of the software application described herein, the software application includes a training module that trains the machine learning software module. The training module is configured to provide training data to the machine learning software module, the training data including, for example, IMU measurements and corresponding body movements associated with the IMU measurements (e.g., accelerometer measurements, gyroscope measurements, and / or magnetometer measurements of translation (or displacement), rotation, velocity, acceleration, jerk, position, or other quantities). In some embodiments, the training data includes simulated IMU data with corresponding simulated body movements. In some embodiments, the machine learning software module utilizes automated statistical analysis of the data to determine which features to extract and / or analyze from the IMU measurements. In some embodiments, the machine learning software module determines which features to extract and / or analyze from the IMU measurements based on the training the machine learning software module receives.

[0081] In some embodiments, the machine learning software module is trained using the dataset and targets in a supervised learning manner. In some embodiments, the dataset is divided into a training set, a test set, and in some embodiments, a validation set. A target is specified that includes the correct classification for each input value in the dataset. For example, a set of IMU data from one or more individuals is repeatedly presented to the machine learning software module, and for each sample presented during training, the output generated by the machine learning software module is compared to the desired target. The difference between the target and the set of input samples is calculated, and the machine learning software module is modified so that the output is closer to the desired target value. In some embodiments, a backpropagation algorithm is utilized so that the output is closer to the desired target value. After a number of training iterations, the output of the machine learning software module closely matches the desired target for each sample in the input training set. When new input data that was not used during training is then presented to the machine learning software module, the module can generate an output classification value that indicates which category the new sample is most likely to fall into. The machine learning software module can generalize from the training to interpret new input samples not seen before. This functionality of the machine learning software module allows for the classification of almost any input data that has a mathematically formalizable relationship to the category to which the data is assigned.

[0082] In some embodiments, the machine learning software module utilizes an individualized learning model based on the fact that the machine learning software module was trained on data from a single individual, such that the machine learning software module utilizing the individualized learning model is configured to be used for the single individual whose data the module was trained on.

[0083] In some embodiments, the machine training software module utilizes a global training model based on which the machine training software module has been trained with data from multiple individuals, such that a machine training software module utilizing the global training model is configured for use with multiple individuals.

[0084] In some embodiments, the machine training software module utilizes a simulated training model that is based on the machine training software module being trained with simulated IMU measurement data.

[0085] In some embodiments, as the availability of IMU data changes, the use of the training model also changes. For example, if there is an insufficient amount of suitable motion data available to train the machine training software module to the desired accuracy, a simulated training model can be used. This lack of data may be the case early in the implementation when few suitable IMU measurements with associated outliers are initially available. As additional data becomes available, the training model can be changed to a global model or an individual model. In some embodiments, a mixture of training models can be used to train the machine learning software module. For example, a simulated global training model can be used, utilizing a mixture of multiple subject data and simulated data to meet the training data requirements.

[0086] In some embodiments, unsupervised learning is used to train a machine learning software module to use input data, such as IMU data, to output, for example, a diagnosis or a body movement. In some embodiments, unsupervised learning includes feature extraction performed by the machine learning software module on the input data. The extracted features can be used to represent input for visualization, classification, subsequent supervised training, and more generally, subsequent storage or analysis. In some embodiments, each training case consists of multiple IMU data.

[0087] Machine learning software modules suitable for unsupervised training include k-means clustering, multinomial mixtures, affinity propagation, discrete factor analysis, hidden Markov models, Boltzmann machines, restricted Boltzmann machines, autoencoders, convolutional autoencoders, recurrent neural network autoencoders, and long short-term memory autoencoders.

[0088] The machine learning software module may include a training phase and a prediction phase. The training phase provides data for training the machine learning algorithm. Non-limiting examples of the types of data input to the machine learning software module for training include medical image data, clinical data (e.g., from health records), encoded data, encoded features, and metrics derived from IMU data. In some embodiments, the data input to the machine learning software module is used to build a hypothesis function for determining the presence of body movement. In some embodiments, the machine learning software module is configured to determine whether a result of the hypothesis function has been achieved and, based on the analysis, make a decision regarding the data on which the hypothesis function was built. That is, the results tend to reinforce the hypothesis function with respect to the data on which the hypothesis function was built, or tend to contradict the hypothesis function with respect to the data on which the hypothesis function was built. In some embodiments, depending on how close the results tend to the results determined by the hypothesis function, the machine learning algorithm may adopt, adjust, or abandon the hypothesis function with respect to the data on which the hypothesis function was built. Thus, the machine learning algorithms described herein dynamically learn through a training phase what features of the input (e.g., data) are most predictive in determining whether features of a subject's recorded IMU data predict a particular body movement.

[0089] For example, a machine learning software module may be provided with data to train on, e.g., so that the module can determine the most salient features of received IMU data to operate on. The machine learning software modules described herein may be trained on how to analyze IMU data rather than analyzing the IMU data using predefined instructions. Thus, the machine learning software modules described herein dynamically learn through training what features of the input measurements are most predictive in determining whether an IMU feature is indicative of body movement.

[0090] In some embodiments, the machine learning software module is trained by repeatedly presenting the IMU data to the machine learning software module along with body movements, such as, for example, flexing, relaxing, stretching, contracting muscles, etc. The IMU data may include accelerometer, gyroscope, or magnetometer data generated in response to electrical stimulation of a human body part.

[0091] In some embodiments, training begins when a machine learning software module is provided with IMU data and asked to determine the presence or absence of body motion. The predicted body motion is then compared to the true body motion corresponding to the IMU data. Optimization techniques, such as gradient descent and backpropagation, are used to update the weights of each layer of the machine learning software module to bring closer the match between the probability of body motion predicted by the machine learning software module and the presence of body motion. This process is repeated with new IMU data and body motion until the accuracy of the network reaches a desired level. In this case, training begins when a machine learning software module is provided with corresponding IMU data and asked to determine the presence or absence of body motion. Optimization techniques are used to update the weights of each layer of the machine learning software module to bring closer the match between the probability of body motion predicted by the machine learning software module and the true body motion. This process is repeated with new IMU data and body motion until the accuracy of the network reaches a desired level. The output data is then compared to the true body motion corresponding to the IMU data. Optimization techniques are used to update the weights of each layer of the machine learning software module to bring closer the match between the probability of body motion predicted by the machine learning software module and the actual body motion. This process is repeated with new IMU data and body movements until the network reaches a desired level of accuracy. After training with the appropriate body movements described above, the machine learning module can analyze the IMU measurements and determine the presence of body movements, the type and location of the body movements, and the conditions associated therewith.

[0092] In some embodiments, the machine learning software module receives the IMU data and directly determines a probability of the subject's physical movement, where the probability of the physical movement includes a probability that the IMU measurement is associated with the subject's physical movement.

[0093] In some embodiments, the user's body movements are input by a user of the system. In some embodiments, the user's body movements are input by an entity other than the user. In some embodiments, the entity may be a health care provider, a health care professional, a family member, or an acquaintance. In some embodiments, the entity may be a system, device, or additional system that analyzes the IMU measurements described herein and provides data regarding physiological anomalies.

[0094] In some embodiments, a strategy is provided for collecting training data to ensure that the IMU measurements represent a wide range of conditions, providing a wide training data set for the machine learning software module. For example, a predetermined number of measurements during a set period of time may be useful as part of the training data set. In addition, these measurements may be defined to provide a certain amount of time between measurements. In some embodiments, the training data set may include IMU measurements taken with varying subject physical conditions.

[0095] b. Prediction phase After training, the machine learning algorithm is used to determine the presence or absence of physical movements for which the system was trained, for example, using a predictive phase. With appropriate training data, the system can identify the type of physical movement and the current state associated with such physical movement. For example, IMU measurements of the subject's brain are taken and appropriate data derived from the IMU measurements are submitted to the system for analysis using the described trained machine learning algorithm. In some embodiments, the machine learning software algorithm detects physical movements associated with health conditions.

[0096] As an example, a subject is known to have a health condition that affects athletic performance, and IMU measurements are recorded before and after treatment. Data from the IMU measurements, and / or features and / or metrics derived from said data, are submitted for analysis to a system that employs trained machine learning algorithms as described to determine the efficacy of treatment using a predictive phase.

[0097] In the prediction phase, the hypothesis function constructed and optimized in the training phase is used to predict the probability of body movement. In some embodiments, in the prediction phase, a machine learning software module can be used to analyze data derived from the IMU measurements independent of any system or device described herein. In some embodiments, the new data record provides a signal window that is longer than the signal window required to determine the presence or absence of body movement of the subject. In some embodiments, the longer signal can be cut to an appropriate size, for example 10 seconds, and then used in the prediction phase to predict the probability of body movement of the new patient data.

[0098] In some embodiments, the probability threshold can be used in combination with the final probability to determine whether a given recording is consistent with the trained body movement. In some embodiments, the probability threshold is used to adjust the sensitivity of the trained network. For example, the probability threshold can be 1%, 2%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 98%, or 99%. In some embodiments, the probability threshold is adjusted if the accuracy, sensitivity, or specificity falls below a predefined adjustment threshold. In some embodiments, the adjustment threshold is used to determine the parameters of the training period. For example, if the accuracy of the probability threshold falls below the adjustment threshold, the system can extend the training period and / or request additional measurements and / or body movements. In some embodiments, the additional measurements and / or body movements can be included in the training data. In some embodiments, additional measurements and / or body movements can be used to refine the training data set.

[0099] (Examples of machine learning techniques) In some embodiments, the systems, methods, computer-readable media, and techniques disclosed herein use various machine learning techniques. In some embodiments, ML involves identifying and recognizing patterns in existing data to facilitate prediction of subsequent data. ML can include ML models, such as ML algorithms. Machine learning, whether analytical or statistical, can provide deductive or inductive inference based on actual or simulated data. The ML model can be a trained model. The ML technique can include one or more supervised, semi-supervised, self-supervised, or unsupervised ML techniques. For example, the ML model can be a trained model trained by supervised learning (e.g., various parameters are determined as weights or scaling coefficients). ML can include one or more of regression analysis, regularization, classification, dimensionality reduction, ensemble learning, meta-learning, association rule learning, cluster analysis, anomaly detection, deep learning, and very deep learning.Non-limiting examples of ML include k-means, k-means clustering, k-nearest neighbors, learning vector quantization, linear regression, nonlinear regression, least squares regression, partial least squares regression, logistic regression, stepwise regression, multivariate adaptive regression splines, ridge regression, principal component regression, least absolute value shrinkage and selection (LASSO), least angle regression, canonical correlation analysis, factor analysis, independent component analysis, linear discriminant analysis, multidimensional scaling, non-negative matrix decomposition, principal component analysis, principal coordinate analysis, projection pursuit, Sammon mapping, t-distributed stochastic neighbor embedding, AdaBoosting, boosting, gradient boosting, bootstrap aggregation, ensemble averaging, decision trees, conditional decision trees, boosted decision trees, gradient boosted decision trees, random forests, stacked generalization, Bayesian networks ... These include: ISM Belief Networks, Naive Bayes, Gaussian Naive Bayes, Multinomial Naive Bayes, Hidden Markov Models, Hierarchical Hidden Markov Models, Support Vector Machines, Encoders, Decoders, Autoencoders, Stacked Autoencoders, Perceptrons, Multilayer Perceptrons, Artificial Neural Networks, Feedforward Neural Networks, Convolutional Neural Networks, Recurrent Neural Networks, Long Short-Term Memory, Deep Belief Networks, Deep Boltzmann Machines, Deep Convolutional Neural Networks, Deep Recurrent Neural Networks, Generative Adversarial Networks, Vision Transformers, Long Short-Term Memory Networks (LSTM), and Masked Autoencoders.

[0100] training Training the ML model may, in some embodiments, include selecting one or more untrained data models for training using the training dataset. The selected untrained data models may include any type of untrained ML models for supervised, semi-supervised, self-supervised, or unsupervised machine learning. The selected untrained data models may be specified based on inputs (e.g., user inputs) that specify relevant parameters to use as predictors or other variables to use as potential explanatory variables. For example, the selected untrained data models may be specified to generate outputs (e.g., predictions) based on the inputs. Conditions for training the ML model from the selected untrained data models may be similarly selected, such as, for example, limits on the complexity of the ML model or limits on the refinement of the ML model beyond a certain point. The ML models may be trained (e.g., via a computer system such as a server) using the training dataset. In some embodiments, a first subset of the training dataset may be selected for training the ML model. The selected untrained data models may then be trained with the first subset of the training dataset using appropriate ML techniques based on the type of ML model selected and any conditions specified for training the ML model. In some embodiments, due to the processing power requirements of training the ML model, the selected untrained data model is trained using additional computing resources (e.g., cloud computing resources). In some embodiments, such training can continue until at least one aspect of the ML model is validated and meets selection criteria for use as a predictive model.

[0101] (verification) In some embodiments, one or more aspects of the ML model may be validated using a second subset of the training dataset (e.g., different from the first subset of the training dataset) to determine the accuracy and robustness of the ML model. Such validation may include applying the ML model to the second subset of the training dataset to make predictions derived from the second subset of the training data. The ML model may then be evaluated to determine whether performance is sufficient based on the derived predictions. The sufficiency criteria applied to the ML model may vary depending on the size of the training dataset available for training, the performance of previous iterations of the trained model, or user-specified performance requirements. If the ML model does not achieve sufficient performance, additional training may be performed. The additional training may include refinement of the ML model or retraining on a different first subset of the training dataset, after which the new ML model may be validated and evaluated again. When the ML model achieves sufficient performance, in some embodiments, the ML model may be stored for current or future use. The ML model may be stored as a set of parameter values ​​or weights for analysis of additional inputs (e.g., additional relevant parameters for use as additional predictor variables, additional explanatory variables, additional user interaction data, etc.), which may also include an indication of the appropriateness of the analysis logic or model, in some embodiments. In some embodiments, multiple ML models may be stored to generate predictions under different sets of input data conditions. In some embodiments, the ML models may be stored in a database (e.g., associated with a server).

[0102] (Deep Learning) The systems, methods, computer-readable media, and techniques disclosed herein may implement one or more deep learning techniques. Deep learning is an example of ML that may be based on a series of algorithms that model high-level abstractions in the data by using multiple processing layers composed of complex structures or other multiple nonlinear transformations. In some embodiments, a dropout method may be used to reduce overfitting. At each training stage, individual nodes are dropped out (e.g., ignored) from the net with probability 1-p or kept with probability p, leaving a reduced network. Input and output edges to the dropped-out nodes may also be removed. In some embodiments, the reduced network is trained on the data of that stage. The removed nodes may then be reinserted into the network with their original weights.

[0103] (Decision Tree) The systems, methods, computer-readable media, and techniques disclosed herein may implement one or more decision trees or random forest techniques. Decision trees may be supervised ML algorithms that can be applied to both regression and classification problems. Decision trees may mimic the decision-making process of the human brain. For example, a decision tree may grow from a root (basic condition), and when the tree meets a condition (internal node / feature), the tree branches into multiple branches. The end of the branch that does not branch further is the outcome (leaf). A decision tree may be generated using a training dataset according to the following operations: (1) Starting from a root node (the entire dataset), the algorithm may split the dataset into two branches using a decision rule or branching criterion. (2) Each of the two branches may generate a new child node. (3) For each new child node, the branching process may be repeated until the dataset cannot be further split. (4) Each branching criterion may be selected to maximize information gain (e.g., quantifying how much the branching criterion reduces the degree of label mixing in the child node). The label may be the data or the classification predicted by the decision tree.

[0104] (Random Forest) Random forest regression is an extension of decision tree models that tends to yield more robust predictions by extending the use of training data partitions. Whereas decision trees make a single pass through the data, random forest regression can bootstrap 50% of the data (e.g., with replacement) to build many trees. Rather than using all explanatory variables as split candidates, a random subset of the candidate variables can be used for splits to generate trees with different data and different variables. The predictions from the trees are then averaged to produce a final prediction, collectively called a forest. A random forest model can contain many trees (e.g., 100 trees), with the number of terms sampled per split (e.g., 3, 6, 10, etc.), the minimum number of splits per tree (e.g., 1, 2, 4, 10, etc.), and the minimum split size (e.g., 16, 32, 64, 128, 256, etc.). Random forests can be trained in a similar manner to decision trees. Training a random forest can include the following operations: (1) Randomly select k features from the total number of features. (2) Create a decision tree from these k features using the same operations as for generating decision trees. (3) Repeat the previous two operations until the target number of trees is created.

[0105] (Long short term memory (LSTM)) The systems, methods, computer-readable media, and techniques disclosed herein may implement one or more long short-term memory (LSTM) techniques. LSTMs may be artificial neural networks used in the fields of artificial intelligence and deep learning. Unlike standard feed-forward neural networks, LSTMs may use feedback connections. LSTM architectures may provide short-term memory for recurrent neural networks (RNNs). Such RNNs may process not only single data points (such as images) but also entire sequences of data (such as audio or video). The connection weights and biases of an RNN may change once per episode of training, similar to how physiological changes in synaptic strength store long-term memories. Activation patterns in the network may change once per time step, similar to how moment-to-moment changes in electrical firing patterns in the brain store short-term memories. LSTM architectures may provide RNNs with short-term memory that can persist for many (e.g., thousands) of time steps.

[0106] In some embodiments, the LSTM unit may include a cell, an input gate, an output gate, and a forget gate. The cell may store values ​​over any time interval, and the input gate, output gate, and forget gate may control the flow of information into and out of the cell. The forget gate may be used to determine what information from the previous state to discard by assigning a value between 0 and 1 to the previous state compared to the current input (e.g., a (rounded) value of 1 means to keep the information, and a value of 0 means to discard the information). The input gate may use the same system as the forget gate to determine what new information to store in the current state. The output gate may control what information of the current state to output (e.g., by assigning a value between 0 and 1 to the information, taking into account the previous and current states). By selectively outputting relevant information from the current state, the LSTM network may maintain long-term dependencies that are useful for making predictions in both the current and future time steps. LSTM networks are well-suited for classification, processing, and prediction based on time series data, since there may be delays of unknown duration between important events in a time series. LSTMs can solve the vanishing gradient problem that can occur when training traditional RNNs. This relative insensitivity to gap length can be an advantage of LSTMs over RNNs, Hidden Markov Models, and other sequence learning methods in many applications.

[0107] In some embodiments, the LSTM may be used with one or more different types of neural networks (e.g., a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), etc.). In some embodiments, the CNN, LSTM, and DNN are complementary in modeling capabilities and may be combined in a unified architecture. For example, in such a unified architecture, the CNN may be suitable for reducing frequency variation, the LSTM may be suitable for temporal modeling, and the DNN may be suitable for mapping features to a more separable space. For example, input features to an ML model using LSTM techniques in a unified architecture may include segment features for each of a plurality of segments. To process the input features of each of the plurality of segments, the segment features of the segment may be processed using one or more CNN layers to generate a first feature of the segment. The first feature may be processed using one or more LSTM layers to generate a second feature of the segment. The second feature may be processed using one or more fully connected neural network layers to generate a third feature of the segment, and the third feature may be used for a classification operation. In some embodiments, to process the first features using one or more LSTM layers to generate second features, the first features may be processed using a linear layer to generate reduced features having dimensions reduced from dimensions of the first features. The reduced features may be processed using one or more LSTM layers to generate second features. Short-term features having a first number of context frames may be generated based on the input features, and the features generated using the one or more CNN layers may include long-term features having a second number of context frames that is greater than the first number of context frames of the short-term features. In some embodiments, the one or more CNN layers, the one or more LSTM layers, and the one or more fully connected neural network layers may be jointly trained to determine training values ​​for parameters of the one or more CNN layers, the one or more LSTM layers, and the one or more fully connected neural network layers.In some embodiments, the input features include log-mel features having multiple dimensions. The input features include one or more context frames that indicate the temporal context of the signal (e.g., input data). Such implementation of the unified architecture can exploit the complementary advantages associated with each of the CNN, LSTM, and DNN. For example, the convolutional layer can reduce the spectral variability of the input and aid in modeling the LSTM layer. Having a DNN layer after the LSTM layer can reduce the variability of the hidden state of the LSTM layer. The unified architecture can be jointly trained to improve overall performance. Training with the unified architecture also eliminates the need for separate CNN, LSTM, and DNN architectures. Adding multi-scale information to the unified architecture can capture information at different time scales.

[0108] (Support Vector Machine) The systems, methods, computer-readable media, and techniques disclosed herein may implement one or more support vector machine learning techniques. In machine learning, a support vector machine (SVM) is a supervised learning model with an associated learning algorithm that analyzes data for classification and regression analysis. SVM may be a robust predictive method based on statistical learning. SVM may be suitable for domains characterized by the presence of large amounts of data, noisy patterns, or the absence of a general theory.

[0109] SVMs can map input vectors into a high-dimensional feature space by a preselected nonlinear mapping function. In this high-dimensional feature space, an optimal separating hyperplane can be constructed. The optimal hyperplane can then be used to determine class separation, regression fitting, accuracy of density estimation, etc. More formally, SVMs construct a hyperplane or set of hyperplanes in a high- or infinite-dimensional space and can be used for classification, regression, or other tasks such as outlier detection.

[0110] Support vectors can be defined as the data points that are closest to the decision surface (or hyperplane). Thus, support vectors may be the most difficult data points to classify and may be directly related to the optimal location of the decision surface. Given a set of training examples, each marked as belonging to one of two categories, an SVM training algorithm can build a model that assigns new examples to either category, making the algorithm a non-probabilistic binary linear classifier. SVMs can map training examples to points in a space that maximizes the width of the gap between the two categories. New examples can then be mapped into the same space and predicted to belong to a category based on which side of the gap the example lies in. In addition to performing linear classification, SVMs can efficiently perform non-linear classification, using what is called the kernel trick, implicitly mapping inputs into a high-dimensional feature space.

[0111] In a support vector machine, the dimensionality of the feature space can be large. For example, a 200-dimensional input space can be mapped to a 1.6 billion-dimensional feature space by a fourth-order polynomial mapping function. SVMs are useful for discovering knowledge from huge amounts of input data.

[0112] (Gradient Boosting) The systems, methods, computer-readable media, and techniques disclosed herein can implement one or more gradient boosting techniques. Gradient boosting is a machine learning technique used in regression and classification tasks, among others. Gradient boosting provides a predictive model in the form of an ensemble of weak predictive models, usually decision trees. When decision trees are weak learners, the resulting algorithm is called a gradient boosted tree. Gradient boosted tree models are built incrementally, similar to other boosting methods, but generalize other methods by allowing optimization of any differentiable loss function.

[0113] (k nearest neighbor method) The systems, methods, computer-readable media, and techniques disclosed herein may implement one or more K-nearest neighbor (KNN) techniques. KNN is a non-parametric classification method. In KNN classification, the output is class membership. An object is classified by majority vote of its neighbors and assigned to the most common class among its k nearest neighbors (k is a positive integer, usually small). If k=1, the object is assigned to the class of its single nearest neighbor. In KNN regression, the output is the object's property value. This value is the average of the values ​​of its k nearest neighbors. KNN is a type of classification in which functions are locally approximated and computation is deferred until the function is evaluated. Because the algorithm relies on distance for classification, accuracy can be improved by normalizing the training data when features represent different physical units or are at widely different scales.

[0114] The systems, methods, computer-readable media, and techniques disclosed herein may implement one or more Monte Carlo methods. Monte Carlo is a broad class of computational algorithms that rely on repeated random sampling to obtain a numerical result. The underlying concept is to use randomness to solve problems that, in principle, may be deterministic.

[0115] (One-hot encoding) The systems, methods, computer-readable media, and techniques disclosed herein can implement one or more one-hot encoding techniques. One-hot encoding can be used to handle categorical data. For example, an ML model can use numerical input variables. The categorical variables can be transformed in the pre-processing part. The categorical data can be either nominal or ordinal. Ordinal data can have a ranking order of values, so it can be transformed into numerical data by ordinal encoding.

[0116] (Example) In the following example, control of neural stimulation by a wearable system is described. The wearable system includes multiple components, specifically, a disposable electrode array for extensors, an extensor patch corresponding to the disposable electrode array for extensors, a disposable electrode array for flexors, a flexor patch corresponding to the disposable electrode array for flexors, a stimulator, a disposable electrode array for thumbs, a flexor / extensor cable, and a thumb dongle, as shown in FIG. 6. The stimulator is paired with a user's personal device (e.g., a mobile device). The user can wirelessly control the neural stimulation of the wearable patch system using the personal device.

[0117] Example 1: A disposable electrode array is attached to the patient's skin. FIG. 13 illustrates disposable electrode arrays configured to be attached to a patient's forearm and thumb. As illustrated, the disposable electrode arrays include a thumb disposable electrode array 1310, an extensor disposable electrode array 1320, and a flexor disposable electrode array 1330. The thumb disposable electrode array 1310 is configured to be worn around the patient's thumb. The extensor disposable electrode array 1320 is configured to be attached to the extensor compartment (also referred to as the posterior compartment) of the forearm to stimulate muscles (e.g., superficial and deep muscles) and neural targets within the compartment. The flexor disposable electrode array 1330 is configured to be attached to the flexor compartment (also referred to as the anterior compartment) of the forearm to stimulate muscles and neural targets within the compartment. As illustrated, both the extensor disposable electrode array 1320 and the flexor disposable electrode array 1330 include multiple electrodes, each attached to a corresponding magnetic connector. Disposable electrode arrays 1310, 1320, and 1330 are attached to a transparent sheet that can be detached from the electrodes when attached to the patient's skin.

[0118] 14A-14E show disposable electrode arrays attached to a patient's forearm and thumb. FIG. 14A shows a flexor disposable electrode array 1410 attached to the flexor compartment of a patient's forearm. FIG. 14B shows an extensor disposable electrode array 1420 attached to the extensor compartment of a patient's forearm. FIG. 14C shows a side view of a patient's arm with an extensor disposable electrode array 1420 and a flexor disposable electrode array 1410 attached and wrapped around the arm. As shown in FIG. 14C, when the flexor and extensor disposable electrode arrays are attached, the majority of the muscles and nerve targets of the forearm are covered.

[0119] 14D and 14E show a disposable electrode array that is attached to and wrapped around a patient's thumb. The magnetic connector 1440 is placed over the ligament just below the bone that connects to the thumb. The Velcro area 1430 is between the palms and the remaining flap 1460 is wrapped around the thumb. The narrow end of the flap 1450 has Velcro that is attached to the other Velcro area 1430 between the palms. Each electrode of the disposable electrode array 1470 makes direct contact with the patient's skin.

[0120] Example 2: A durable / reusable patch is applied to the patient's skin. After the disposable flexor and extensor electrode arrays are attached to the patient's forearm, a corresponding durable / reusable electronic patch is placed over the disposable electrode array. FIGS. 15A-15C show the durable / reusable electronic patch attached to the disposable electrode array. FIG. 15A shows a durable / reusable flexor patch 1510 corresponding to the flexor disposable electrode array 1410 attached to the patient's forearm, overlapping the disposable electrode array. FIG. 15B shows a durable / reusable patch 1520 corresponding to the extensor disposable electrode array 1420 attached to the patient's forearm, overlapping the disposable electrode array. FIG. 15C shows a side view of the patient's arm with the flexor patch 1510 and extensor patch 1520 attached. The flexor patch 1510 has the same number of magnetic connectors as the flexor disposable electrode array 1410, such that when the patch 1510 is placed over the electrode 1410, each magnetic connector in the patch can be aligned with a corresponding magnetic connector on the electrode. Similarly, the durable / reusable extensor patch 1520 has the same number of magnetic connectors attached to the extensor disposable electrode array 1420. The thumb disposable electrode array does not require a patch to function.

[0121] As further shown in FIG. 15C, when a durable / reusable patch is placed on a disposable electrode array, the electrodes may lift slightly off the skin. To correct this lift, the assembly can be secured after the patch is placed on the disposable electrode array. FIGS. 16A-16C show how an assembly including a durable / reusable patch and a corresponding disposable electrode array is attached and secured to the patient's forearm. Multiple securing mechanisms (e.g., straps) can be used, with as much of the assembly as possible touching the forearm. If both flexor and extensor patches are used simultaneously, three forearm straps can be used, as shown in FIGS. 16A and 16C. If only extensor patches are used, two forearm straps can be used. This component is not required for the thumb electrode.

[0122] Example 3: Establishing a connection between a stimulator and a durable / reusable patch. The stimulator module 1710 is attached to a wrist strap and placed on the patient's wrist, with the LED sign 1720 of the stimulator module 1710 located near the patient's finger.

[0123] The stimulator module 1710 includes three cable ports configured to connect to the flexor and extensor patches and the thumb disposable electrode array. FIG. 18 shows the connection of the stimulator, the flexor and extensor patches, and the thumb disposable electrode array via cables. The flexor cable 1810 is connected between the stimulator module 1850 and the flexor patch 1820. The extensor cable 1830 is connected between the stimulator module 1850 and the extensor patch 1840. The thumb dongle 1860 is connected between the stimulator module 1850 and the thumb disposable electrode array. In this way, the stimulator establishes an electrical connection between the flexor and extensor patches and the thumb disposable electrode array 1870.

[0124] Example 4: Control of neurostimulation by a wearable patch system. Once the stimulator module is connected to the flexor and extensor patches and the thumb disposable electrode array, the user can operate a personal device (e.g., a mobile device) and control the patches and electrodes via a mobile, web, or desktop application. FIG. 19 shows a graphical user interface (GUI) 1900 of an application that allows a user to control stimulation. The GUI displays the name of the device 1910, which refers to the stimulator with which the patient's personal device is paired, and the device's battery level. The user can select the name of the patient for whom stimulation will be performed. The application can store records for multiple patients. In a home care environment, the patient wearing the patch can operate the mobile device to control the stimulation. In a medical setting, a medical professional can operate a web or desktop application to select the name of the patient wearing the patch and control the stimulation.

[0125] The GUI also allows the user to select between manual stimulation 1940 and stimulation sequences 1950. In the manual stimulation section 1940, the user can select stimulation areas 1920 including flexors, extensors, and thumbs. Each of these stimulation categories corresponds to an electrode and patch as shown in Figures 13-18. The user can control which electrode stimulates by selecting the stimulation site. The user can save desired inputs for specific actions during manual stimulation, such as location and amplitude of stimulation. These saved actions can be executed sequentially or simultaneously by selecting a stimulation sequence 1950.

[0126] FIG. 20 shows a GUI 2000 of an application that allows a user to control manual stimulation. The user can select either flexor or extensor stimulation, and the selection is displayed in the stimulation selection section 2010. A stimulation indicator 2020 (e.g., "on / off") is also displayed in the GUI and changes automatically depending on the state of the stimulation amplitude slider 2030. For example, when the slider 2030 is dragged all the way to the bottom, indicating the minimum value of amplitude to be applied, the stimulation indicator 2020 automatically turns off. When the slider is dragged up from the minimum, the button automatically turns on. Alternatively, the stimulation indicator 2020 can be turned on / off independently of the amplitude slider 2030.

[0127] As described above in accordance with FIG. 18, the flexor and extensor patches have LEDs that indicate the location of the electrodes that will generate electrical stimulation. The user can control the location of the electrodes by navigating the control area 2040. The user can move their finger over the control area 2040 to find the desired spot to stimulate. The LED on the patch that corresponds to the desired spot can light up to indicate the location of the electrodes to stimulate. The user can also test different locations of the LEDs on the patch and see where the patient responds to stimulation by navigating the control area 2040.

[0128] The GUI also provides a keep LED on function 2050. Once the user has identified a desired spot for electrical stimulation, indicated by an LED on the patch, the user may tap the keep LED on function 2050 to save the location of the LED. The selected LED, which corresponds to the location of the desired spot for stimulation, may change to a different color. The user may continue to navigate on the control area 2040 and identify a different desired spot for electrical stimulation. Another LED on the patch, representing the user's newly selected spot, may illuminate.

[0129] FIG. 21 illustrates a user selection of multiple locations on a patch for performing electrical stimulation, with each location indicated by an LED. The location of the blue LED 2110 indicates the first desired spot that the user selected and saved by tapping the "keep LED on" button 2050. The location of the green LED 2120 represents the new location to which the user is moving on the control area 2040. As the user's finger slides on the control area 2040, the location of the green LED 2120 changes accordingly. Once the user identifies a second desired spot that they wish to stimulate, they tap the keep LED on 2050 again to save this spot. In response, the green LED 2120 turns blue, indicating that the corresponding spot has been selected and saved.

[0130] The GUI allows the user to save a pattern at one or more selected positions. As shown in FIG. 20, the user can tap Save Pattern 2100. The desired pattern can be saved to the user's profile. The user can load this pattern even after adjusting the input. In some embodiments, only one pattern can be saved at a time for each selected motion. If another pattern is saved over an already saved pattern, the newly saved pattern may overwrite the selected motion. If the user wants to delete the previously selected positions, indicated by the blue LEDs on the patch, they can tap Turn LEDs Off 2060 or tap Turn All Off 2070 to delete all selected positions. Since each pattern is already saved to the user's profile, the user can tap Load Pattern 2090 to restore the previously saved patterns even after turning off the blue LEDs.

[0131] FIG. 22 shows a GUI 2200 of an application that allows a user to control thumb stimulation. Instead of selecting either extensor or flexor stimulation, the user can select thumb stimulation, which is shown in the stimulation selection section 2210. A stimulation indicator 2220 (e.g., on / off) is also displayed in the GUI and changes automatically depending on the state of the stimulation amplitude slider 2230. For example, when the slider 2230 is dragged all the way to the bottom, indicating the minimum value of amplitude to be applied, the stimulation indicator 2220 automatically turns off. When the slider is dragged up from the minimum, the button automatically turns on. Alternatively, the stimulation indicator 2220 can be turned on / off independently of the amplitude slider 2230.

[0132] The GUI 2200 provides an electrodes section 2240 that includes a visual representation of the five subsystem electrodes in the disposable thumb electrode (see five subsystem electrodes 1470 in FIG. 14E). Each electrode can be selected to control a different section of the thumb movement. The user can select a single electrode or multiple electrodes at once, depending on the movement selected in the movement selection section 2260. As shown, for example, electrodes 2 and 3 are selected simultaneously to generate electrical stimulation. The user can also tap Turn All Off 2250 to remove the selected positions.

[0133] The GUI 2200 also provides load pattern 2270 and save pattern 2280 with similar functionality as in the flexor and extensor modes (e.g., 2090 and 2100 in FIG. 20). A user can save a desired pattern with a given amplitude and stimulation area by tapping save pattern 2280. A saved pattern can be loaded by tapping load pattern 2270.

[0134] As described above, a user can save desired inputs for specific actions during manual stimulation, such as stimulation location, amplitude, etc. A stimulation sequence function can be selected by the user so that the system can execute the saved actions sequentially or simultaneously.

[0135] FIG. 23 shows a GUI 2300 of an application for executing a stimulation sequence 2300. Instead of selecting a manual stimulation, the user can select a stimulation sequence 2310 that can execute one or more saved movements. A stimulation indicator 2320 (e.g., on / off) is displayed in the GUI and changes automatically depending on the state of a stimulation amplitude slider 2330. The functions of the stimulation indicator 2320 and the stimulation amplitude slider 2330 are similar to those described in FIG. 20 and FIG. 22. A movement selection section 2340 includes three saved movements for the user to select. As shown, for example, a movement in which the user opens his / her hand during manual stimulation is saved. Similarly, the user can operate a scroll-down menu to select one of the saved movements. Also, in a movement action section 2350, the user can select how to execute the movements, for example, to play them in sequence, to stimulate together, etc.

[0136] (Embodiment)

[0023] Embodiment 1. A device, comprising: a first article having: 1) a first top surface, the first top surface having a set of first connectors, and 2) a first bottom surface, the first bottom surface having a plurality of electrodes; and a second article having: 1) a circuit board operatively connected to the plurality of electrodes; 2) a second top surface, the second top surface having a plurality of visualization aids, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes; and 3) a second bottom surface, the second bottom surface having a set of second connectors, the first connectors being connected to the first and second electrodes. and a second article having a second bottom surface, each of the first and second connectors independently configured to couple to one of the second connectors, the first connector and the second connector configured to form a connection that holds the first and second articles together when the first connector is coupled to the second connector, and when the first and second articles are operably connected, the first and second articles together form a wearable, the wearable having a wearable size and a wearable shape adapted to fit a human body part.

[0137] Embodiment 2. The device of embodiment 1, wherein the visualization aid is a light.

[0138] Embodiment 3. The device of embodiment 1 or 2, wherein the visualization aid is an LED.

[0139] Embodiment 4 A device according to any one of embodiments 1 to 3, wherein each visualization aid is configured to provide a visible signal when an electrode corresponding to the visualization aid emits an electrical signal.

[0140] Embodiment 5. The device of any one of embodiments 1 to 4, wherein the plurality of electrodes are flexible.

[0141] Embodiment 6. The device of any one of embodiments 1 to 5, wherein the plurality of electrodes is a silver ink.

[0142] Embodiment 7 A device described in any one of embodiments 1 to 6, wherein the multiple electrodes are configured to stimulate the muscle tissue and the neural target upon application of electrical stimulation to the muscle tissue and the neural target.

[0143] Embodiment 8. A device according to any one of embodiments 1 to 7, wherein the multiple electrodes are connected in series.

[0144] Embodiment 9. A device according to any one of embodiments 1 to 8, wherein the first connector and the second connector are magnetic.

[0145] Embodiment 10. A device described in any one of embodiments 1 to 9, wherein the first article has a first article shape and the second article has a second article shape, and when the first article and the second article are operably connected, the first article shape and the second article shape substantially overlap.

[0146]

[0026] Embodiment 11: The first bottom surface is at least 25 cm 2 11. The device of any one of embodiments 1 to 10, having a surface area of

[0147] Embodiment 12. The device of any one of embodiments 1 to 11, wherein the circuit board is flexible.

[0148] Embodiment 13. A device according to any one of embodiments 1 to 12, wherein the circuit board comprises a conductor.

[0149]

[0023] Embodiment 14. The device of embodiment 13, wherein the conductor is gold-plated copper.

[0150] Embodiment 15. The device of embodiment 13 or 14, wherein the conductor is bonded to the plating.

[0151] Embodiment 16. A device according to any one of embodiments 13 to 15, wherein the conductor is joined to a nickel-copper-nickel plating.

[0152] Embodiment 17. The device of any one of embodiments 1 to 16, wherein the second article is durable.

[0153] Embodiment 18. A device according to any one of embodiments 1 to 17, wherein the first article is disposable.

[0154] Embodiment 19. A device according to any one of embodiments 1 to 18, wherein the first article is configured to contact human skin.

[0155] Embodiment 20. A device described in any one of embodiments 1 to 19, wherein the first article comprises a bottom layer, an intermediate layer, and a top layer, the intermediate layer being stacked on top of the bottom layer, in contact with the bottom layer, and operably connected to the bottom layer, and the top layer being stacked on top of the intermediate layer, in contact with the intermediate layer, and operably connected to the intermediate layer.

[0156]

[0046] Embodiment 21. The device of embodiment 20, wherein the bottom layer comprises a plurality of electrodes.

[0157]

[0046] Embodiment 22. The device of embodiment 20 or 21, wherein the bottom layer comprises a hydrogel and a plurality of electrodes.

[0158] Embodiment 23. The device of any one of embodiments 20-22, wherein the intermediate layer comprises a dielectric material.

[0159] Embodiment 24 A device described in any one of embodiments 20 to 23, wherein the top layer comprises a set of first connectors.

[0160] Embodiment 25. A device described in any one of embodiments 20 to 24, wherein the bottom layer has a bottom layer shape, the middle layer has a middle layer shape, and the top layer has a top layer shape, the bottom layer and middle layer are operably connected, and when the middle layer and top layer are operably connected, the bottom layer shape, middle layer shape, and top layer shape substantially overlap.

[0161] Embodiment 26c) A device described in any one of embodiments 1 to 25, further comprising a control system operably connected to a second item, the control system comprising: 1) a power source; 2) a processor operably connected to the power source and configured to operate the device; and 3) a stimulation device operably connected to the processor, operably connected to the power source, and configured to send electrical stimulation to the circuit board.

[0162]

[0046] Embodiment 27. A device as described in embodiment 26, wherein the power source, processor, and stimulator are within a common housing.

[0163]

[0046] Embodiment 28. The device of embodiment 26 or 27, wherein the control system further comprises a wireless receiver operably connected to the processor and configured to receive instructions from a user to send electrical stimulation to the circuit board.

[0164] Embodiment 29. A device described in any one of embodiments 26 to 28, wherein the control system further comprises a memory system operably connected to the processor and configured to record information regarding use of the device.

[0165] Embodiment 30. A device described in any one of embodiments 26 to 29, wherein the control system further comprises a transmitter operably connected to the processor and configured to wirelessly transmit information regarding the use of the device to a recipient external to the device.

[0166] Embodiment 31. The device of embodiment 30, wherein the information regarding the use of the device is a record of stimuli applied to a human body part in physical contact with the device, and movements of the human body part in response to the stimuli.

[0167] Embodiment 32. A device described in any one of embodiments 26 to 31, wherein the control system further comprises a motion detector operably connected to the processor and configured to detect motion of a human body part in physical contact with the device.

[0168]

[0023] Embodiment 33. A device comprising: a) a first article, the first article being disposable, the first article comprising: 1) a bottom layer comprising A) a hydrogel, and B) a plurality of electrodes in contact with the hydrogel, the plurality of electrodes being connected in sequence, the plurality of electrodes being flexible, the plurality of electrodes being silver ink, the plurality of electrodes being configured to stimulate the muscle tissue and the neural target upon application of electrical stimulation to the muscle tissue and the neural target; 2) a middle layer, the middle layer comprising a dielectric material, the middle layer laminated on the bottom layer. 1) a first article, comprising: a) a middle layer, in contact with and operably connected to the bottom layer; and 2) a top layer, wherein the top layer comprises a set of first connectors, the first connectors being magnetic, the top layer being stacked on top of, in contact with and operably connected to the middle layer, wherein the bottom layer has a bottom layer shape, the middle layer has a middle layer shape, and the top layer has a top layer shape, the bottom layer and middle layer are operably connected, and when the middle layer and top layer are operably connected, the bottom layer shape, the middle layer shape, and the top layer shape substantially overlap; and b) a second article, comprising: , the second article is durable, and the second article comprises: 1) a second top surface, the second top surface comprising a plurality of visualization aids, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes, the plurality of visualization aids being LEDs, and each visualization aid configured to provide a visible signal when an electrode corresponding to the visualization aid emits an electrical signal; 2) a flexible circuit board comprising conductors, the flexible circuit board operatively connected to the plurality of electrodes; and 3) a second bottom surface, the second bottom surface comprising a plurality of visualization aids each of which independently corresponds to one of the plurality of electrodes, the plurality of visualization aids being LEDs, and each visualization aid configured to provide a visible signal when an electrode corresponding to the visualization aid emits an electrical signal. a second article having a second bottom surface, the bottom surface having a set of second connectors, the second connectors being magnetic, and each of the first connectors being independently configured to couple to one of the second connectors; and c) a control system operably connected to the second article, the control system including: 1) a power source; 2) a processor operably connected to the power source and configured to operate the device; 3) a stimulator operably connected to the processor, operably connected to the power source, and configured to deliver electrical stimulation to the flexible circuit board; 4) a control system operably connected to the processor;a wireless receiver configured to receive instructions from a user to send electrical stimuli to the flexible circuit board; 5) a memory system operably connected to the processor and configured to record information regarding use of the device; 6) a transmitter operably connected to the processor and configured to wirelessly transmit to a receiver external to the device a record of stimuli applied to a human body part in physical contact with the device and movement of the human body part in response to the stimuli; and 7) a motion detector operably connected to the processor and configured to detect movement of a human body part in physical contact with the device, comprising a power source, a processor, a stimulator, a wireless receiver, a memory system, a transmitter, and and a control system, wherein the motion detector is within a common housing, the first connector and the second connector are configured to form a connection that holds the first article and the second article together when the first connector is mated to the second connector, and when the first article and the second article are operably connected, the first article and the second article together form a wearable, the wearable having a wearable size and a wearable shape adapted to fit a human body part, the first article having a first article shape and the second article having a second article shape, when the first article and the second article are operably connected, the first article shape and the second article shape substantially overlap, and the bottom layer has a thickness of at least 25 cm, 2 The device has a surface area of

[0169] Embodiment 34 A system comprising a device according to any one of embodiments 1 to 33, and a communication device configured to wirelessly operate the device.

[0170]

[0046] Embodiment 35. The system of embodiment 34, wherein the communications device comprises non-transitory computer-executable code encoded on a computer-readable medium, the non-transitory computer-executable code configured to cause the device to operate based on instructions provided by a user.

[0171]

[0046] Embodiment 36. The system of embodiment 35, wherein the non-transitory computer executable code is configured to monitor the performance of a human subject using the device in physical therapy.

[0172] Embodiment 37. The system of embodiment 35 or 36, wherein the non-transitory computer executable code is configured to provide an analysis of a human subject's performance using the device in therapy and determine a proposed treatment regimen based on the analysis, the proposed treatment regimen including activation of a plurality of electrodes in a pattern determined to be likely to provide improvement in physical therapy based on the analysis.

[0173] Embodiment 38. A system described in any one of embodiments 35 to 37, wherein the communication device comprises a touch screen that displays a grid, the grid corresponding to a layout of the multiple electrodes.

[0174] Embodiment 39. A system as described in any one of embodiments 35 to 38, wherein the communication device has a touchscreen displaying a grid, the grid corresponding to a layout of a plurality of electrodes, and when a point on the grid of the touchscreen is touched, an electrode among the plurality of electrodes that corresponds to the point on the grid is activated.

[0175] Embodiment 40. A method comprising contacting a device according to any one of embodiments 1 to 33 with a human subject.

[0176] Embodiment 41 A method comprising: a) contacting a body part of a human subject with a plurality of electrodes, the plurality of electrodes operably connected to a flexible circuit board and the body part having a shape; b) manipulating the flexible circuit board to substantially conform to the shape of the body part; c) operably connecting a plurality of visualization aids to the flexible circuit board, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes; and d) passing an electric current through the plurality of electrodes, the plurality of electrodes providing an electrical stimulus to the body part.

[0177]

[0081] Embodiment 42. The method of embodiment 41, wherein the body part comprises muscle tissue and a neural target.

[0178]

[0046] Embodiment 43. The method of embodiment 41 or 42, wherein the plurality of electrodes are disposed within a hydrogel, and the hydrogel is in contact with the body part.

[0179]

[0081] Embodiment 44. The method of any one of embodiments 41-43, wherein the plurality of electrodes is a silver ink.

[0180]

[0081] Embodiment 45. The method according to any one of embodiments 41 to 44, wherein each visualization aid is configured to provide a visible signal when an electrode corresponding to the visualization aid emits an electrical signal.

[0181]

[0081] Embodiment 46. The method of embodiment 45, wherein the plurality of visualization aids are LEDs.

[0182]

[0071] Embodiment 47. The method of any one of embodiments 41-46, further comprising operably connecting a current source to the flexible circuit board.

[0183]

[0046] Embodiment 48. The method of any one of embodiments 41 to 47, wherein the flexible circuit board is operably connected to a current source.

[0184]

[0081] Embodiment 49. The method of any one of embodiments 41 to 48, further comprising tracking movement of the body part in response to electrical stimulation.

[0185] Embodiment 50. The method of any one of embodiments 41 to 49, further comprising tracking movement of the body part in response to the electrical stimulation and designing a physical therapy regimen based at least in part on the movement of the body part in response to the electrical stimulation, the physical therapy regimen comprising a pattern of electrical signals applied via the plurality of electrodes.

[0186] Embodiment 51 A method comprising: a) contacting a body part of a human subject with a plurality of electrodes, the plurality of electrodes operably connected to a flexible circuit board and the body part having a shape; b) manipulating the flexible circuit board to substantially conform to the shape of the body part; c) receiving instructions from a wireless user device to send electrical stimulation to the flexible circuit board; and d) selecting at least a portion of the plurality of electrodes and passing an electrical current through a portion of the plurality of electrodes, the portion of the plurality of electrodes providing the electrical stimulation to the body part.

[0187]

[0081] Embodiment 52. The method of embodiment 51, wherein the body part comprises muscle tissue and a neural target.

[0188]

[0046] Embodiment 53. The method of embodiment 51 or 52, wherein the plurality of electrodes are disposed within a hydrogel, and the hydrogel is in contact with the body part.

[0189]

[0081] Embodiment 54. The method of any one of embodiments 51-53, wherein the plurality of electrodes is a silver ink.

[0190] Embodiment 55. A method as described in any one of embodiments 51 to 54, further comprising operably connecting a plurality of visualization aids to the flexible circuit board, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes, and each visualization aid being configured to provide a visible signal when the electrode corresponding to the visualization aid emits an electrical signal.

[0191] Embodiment 56. A method as described in any one of embodiments 51 to 55, wherein the flexible circuit board is operably connected to a plurality of visualization aids, each of the plurality of visualization aids independently corresponding to one of the plurality of electrodes, and each visualization aid is configured to provide a visible signal when the electrode corresponding to the visualization aid emits an electrical signal.

[0192]

[0046] Embodiment 57. The method of embodiment 56, wherein the plurality of visualization aids are LEDs.

[0193]

[0046] Embodiment 58. The method of any one of embodiments 51 to 57, further comprising operably connecting a current source to the flexible circuit board.

[0194]

[0046] Embodiment 59. The method of any one of embodiments 51 to 58, wherein the flexible circuit board is operably connected to a current source.

[0195] Embodiment 60. The method of any one of embodiments 51-59, further comprising tracking movement of the body part in response to electrical stimulation.

[0196] Embodiment 61 The method of any one of embodiments 51 to 60, further comprising tracking movement of the body part in response to the electrical stimulation and designing a physical therapy regimen based at least in part on the movement of the body part in response to the electrical stimulation, the physical therapy regimen comprising a pattern of electrical signals applied via a plurality of electrodes.

[0197] Embodiment 62 An addressable electrode system comprising: a power source; and an article comprising: (i) a plurality of sequentially connected electrodes; (ii) a plurality of indicators; and (iii) a plurality of switches, each switch independently comprising a first terminal, a second terminal, and a third terminal, wherein the first terminal of the switch is electrically coupled to the power source, the second terminal of the switch is electrically coupled to one indicator of the plurality of indicators, and the third terminal of the switch is electrically coupled to one electrode of the plurality of sequentially connected electrodes, and the switches are configured to simultaneously supply power to the electrodes and the indicators when power is supplied to the first terminal of the switch by the power source.

[0198]

[0046] Embodiment 63. The system of embodiment 62, wherein the article is a textile.

[0199] Embodiment 64. The system of embodiment 62 or 63, wherein the article is divided into a plurality of patches.

[0200]

[0046] Embodiment 65. The system of embodiment 64, wherein each indicator of the plurality of indicators independently corresponds to one patch of the plurality of patches.

[0201] Embodiment 66. The system of embodiment 65, wherein the indicator is disposed on the patch.

[0202] Embodiment 67. A system described in any one of embodiments 62 to 66, wherein a plurality of serially connected electrodes are connected in a daisy chain configuration.

[0203] Embodiment 68. The system of any one of embodiments 62 to 67, wherein the switch includes a solid-state relay.

[0204]

[0071] Embodiment 69. The system of any one of embodiments 62-67, wherein the switch comprises a metal oxide semiconductor field effect transistor (MOSFET).

[0205] Embodiment 70. The system of any one of embodiments 62-69, wherein the one indicator of the plurality of indicators and the one switch of the plurality of switches are integrated into a single integrated circuit (IC).

[0206] Embodiment 71. A system described in any one of embodiments 62 to 70, wherein the second terminal of each switch is electrically coupled to at most one indicator.

[0207] Embodiment 72. A system described in any one of embodiments 62 to 71, wherein the second terminal of each switch is electrically coupled to at most one electrode.

[0208] Embodiment 73. A system described in any one of embodiments 62 to 72, wherein each electrode of the plurality of serially connected electrodes is electrically coupled to a return electrode.

[0209]

[0081] Embodiment 74. The system of embodiment 73, wherein each return electrode is integrated with one of the electrodes of the plurality of serially connected electrodes.

[0210] Embodiment 75. A system described in any one of embodiments 62 to 74, wherein the multiple continuously connected electrodes comprise at least a first electrode and a second electrode, the first electrode being connected to the second electrode by a coiled wire, and the coiled wire being flexible.

[0211] Embodiment 76. A system described in any one of embodiments 62 to 75, wherein each indicator generates visible light.

[0212]

[0076] Embodiment 77. The system of embodiment 76, wherein each indicator is a light emitting diode (LED).

[0213] Embodiment 78. A method for providing neural stimulation comprising: a) receiving a user gesture at a user interface (UI), the gesture being physical contact with the user interface; b) generating a control signal based at least in part on the gesture; c) generating a recommended spatial pattern of neural stimulation impulses for the subject from the control signal; and d) supplying a current to at least one of a plurality of electrodes in contact with the subject to provide the spatial pattern of neural stimulation impulses to the subject.

[0214]

[0076] 79. The method of claim 78, wherein the user interface comprises a graphical user interface (GUI).

[0215]

[0083] Embodiment 80: The method of embodiment 79, wherein the GUI includes a layout, the layout corresponding to a spatial configuration of the multiple electrodes.

[0216]

[0076] 81. The method of claim 80, wherein the layout includes at least two dimensions.

[0217] 82. The method of claim 81, wherein the layout is a grid.

[0218]

[0083] The method of any one of embodiments 78-81, further comprising displaying a user interface on the communication device.

[0219]

[0071] Embodiment 84. The method of embodiment 83, wherein the communication device is a mobile device.

[0220]

[0082] Embodiment 85. The method of any one of embodiments 78 to 84, wherein the user interface is provided by a software application.

[0221]

[0046] Embodiment 86: The method of embodiment 85, wherein the software application is a mobile application.

[0222] Embodiment 87. A method according to any one of embodiments 78 to 86, wherein physical contact with the user interface occurs with a portion of the user interface, the portion of the user interface corresponding to a location on the subject of at least one of the plurality of electrodes.

[0223]

[0082] Embodiment 88. The method of embodiment 87, wherein the physical contact occurs from a hand or a stylus.

[0224]

[0082] Embodiment 89. The method of embodiment 88, wherein the physical contact comprises one or more taps.

[0225]

[0046] Embodiment 90. The method of embodiment 89, wherein one or more taps are performed simultaneously.

[0226]

[0046] Embodiment 91. The method of embodiment 89, wherein the one or more taps are contiguous in time.

[0227]

[0093] Embodiment 92. The method of embodiment 87, wherein the physical contact comprises a prolonged depression.

[0228]

[0096] Embodiment 93. The method of embodiment 92, wherein the physical contact further comprises dragging.

[0229]

[0081] Embodiment 94. The method of any one of embodiments 78 to 93, wherein the control signal is provided over a network.

[0230]

[0096] 35. The method of claim 34, wherein the network is a wired or wireless network.

[0231]

[0096] Embodiment 96. The method of embodiment 95, wherein the network is Ethernet, Bluetooth, or Wi-Fi.

[0232] Embodiment 97. The method of any one of embodiments 78 to 96, wherein the plurality of electrodes comprises an array of electrodes.

[0233]

[0082] Embodiment 98. The method of embodiment 97, wherein the array of electrodes is embedded in the article.

[0234]

[0081] Embodiment 99. The method of embodiment 98, wherein the article is a textile.

[0235]

[0046] Embodiment 100. The method of embodiment 99, wherein the textile is incorporated into a wearable device.

[0236] Embodiment 101 A method according to any one of embodiments 78 to 100, wherein generating the spatial pattern of the recommended neural stimulation impulses is executed by a processor.

[0237]

[0071] Embodiment 102. The method of embodiment 101, wherein the control signal provides a set of instructions that, when executed by the processor, generates a recommended spatial pattern of neural stimulation impulses.

[0238] Embodiment 103 A system for providing neural stimulation, the system comprising: a communication device; a stimulation module; an electrode array; and a power source; the communication device comprising: a software application configured to provide a user interface, the user interface configured to accept a gesture that is physical contact with the user interface and generate a control signal based at least in part on the gesture; and a transmitter configured to transmit the control signal to the stimulation module; the electrode array comprising a plurality of electrodes, each electrode electrically coupled to a power source via a switch, each switch configured to connect or disconnect a corresponding electrode to the power source or from the power source; the stimulation module comprising: a receiver configured to receive the control signal; a processor operably coupled to the receiver, the processor configured to generate a recommended pattern of neural stimulation impulses from the control signal; and a switching unit configured to configure each switch to connect or disconnect a corresponding electrode in response to the recommended pattern of neural stimulation impulses.

[0239] Embodiment 104 A system comprising a first device described in embodiment 1 and a second device described in embodiment 1.

[0240] Embodiment 105. The system of embodiment 105, wherein a second device is operably coupled to the subject's neck, the second device configured to provide a second neural stimulation to the spinal cord.

[0241]

[0081] Embodiment 106. The system of embodiment 106, wherein the second device is configured to provide a second neural stimulation in response to a first neural stimulation provided by the first device.

[0242] Embodiment 107. The system of embodiment 107, wherein the second device is configured to provide a second neural stimulation simultaneously with the first neural stimulation provided by the first device.

[0243]

[0082] Embodiment 108. The system of embodiment 107, wherein the second device is configured to provide a second neural stimulation after the first neural stimulation provided by the first device.

[0244] Embodiment 109. A system described in any one of embodiments 107 to 109, wherein the second device is configured to provide a second neural stimulation for a duration longer than the duration of the first neural stimulation provided by the first device.

[0245] Embodiment 110. A system described in any one of embodiments 107 to 110, wherein the second device is configured to provide a second neural stimulation for a duration equal to the duration of the first neural stimulation provided by the first device.

[0246] Embodiment 111 A device comprising: a first article having 1) a plurality of electrodes and 2) a layer of hydrogel; and a second article having 1) a circuit board operably connected to the plurality of electrodes, and 2) a second upper surface comprising a plurality of visualization aids, each visualization aid of the plurality of visualization aids independently corresponding to one of the plurality of electrodes, wherein when the first article and the second article are operably connected, the first article and the second article together form a wearable, and the wearable has a wearable size and a wearable shape adapted to fit a human body part.

Claims

1. A system, A first article, wherein the first article is flexible and, 1) A first upper surface having a set of first connectors, and 2) A first bottom surface comprising a plurality of electrodes Equipped with, The number of the first set of connectors is the same as the number of electrodes. A first article wherein each first connector of the set of first connectors is positioned in the same location as the corresponding electrode of the plurality of electrodes, A second article, wherein the second article is flexible and, A flexible circuit board that can be operably connected to the plurality of electrodes of the first article, The second upper surface, and A second bottom surface, wherein the second bottom surface comprises a set of second connectors, the number of first connectors in the set is the same as the number of second connectors in the set, and each first connector in the set of first connectors is independently configured to connect to one of the second connectors in the set of second connectors. The second article comprising Equipped with, The set of first connectors and the set of second connectors are configured such that when the set of first connectors is coupled to the set of second connectors, they hold the first article and the second article together and form a connection portion that operably connects the flexible circuit board to the plurality of electrodes. A system wherein, when the first article and the second article are operably connected, the first article and the second article together form a wearable, the wearable having a wearable size and wearable shape adapted to fit a part of the human body.

2. The system according to claim 1, wherein the second upper surface is provided with a plurality of visual indicators, each of the plurality of visual indicators independently corresponds to one of the plurality of electrodes, and the plurality of visual indicators are lights.

3. The system according to claim 2, wherein each of the plurality of visual indicators is configured to supply a visible signal when an electrode corresponding to the visual indicator emits an electrical signal.

4. The system according to claim 1, wherein each connector of the set of first connectors forms a magnetic connection with the corresponding connector of the set of second connectors, and each magnetic connection is conductive.

5. The system according to claim 1, wherein the second article is durable and reusable.

6. The system according to claim 5, wherein the first article is configured to come into contact with human skin and is disposable.

7. A stimulator that can be operably connected to the second article, 1) A power supply comprising at least one of a battery or an AC adapter, 2) A processor that is operably connected to the power supply and configured to operate the system, 3) A stimulation circuit configured to be operably connected to the processor, operably connected to the power supply, and to supply an electrical stimulation signal to the flexible circuit board. The system according to claim 1, further comprising a stimulator having

8. The stimulator further comprises a wireless receiver operably connected to the processor and configured to receive instructions from a user, wherein the stimulator supplies the electrical stimulation signal to the flexible circuit board based on the instructions from the user. The system according to claim 7, wherein the system is operated remotely or locally by the user.

9. The system according to claim 7, wherein the stimulator further comprises a motion sensor operably connected to the processor and configured to detect the movement of a body part.

10. The system according to claim 1, further comprising, when executed, non-temporary computer executable code configured to provide an analysis of the performance of a human subject using a wearable in treatment, and to determine a proposed treatment regimen based on the analysis thereof, wherein the proposed treatment regimen includes activation of the plurality of electrodes in a pattern that is determined to be likely to provide an improvement in physiotherapy based on the analysis thereof.

11. The system according to claim 1, further comprising non-temporary computer executable code configured to cause a grid to be displayed on a touchscreen when executed, wherein the grid corresponds to the layout of the plurality of electrodes, and touching a point on the grid with the touchscreen activates one of the plurality of electrodes that corresponds to the point on the grid.

12. The system according to claim 9, further comprising non-temporary computer executable code configured, when executed, to track the movement of the body part in response to an electrical stimulus and to design a physiotherapy regimen at least in part on the movement of the body part in response to the electrical stimulus, wherein the physiotherapy regimen includes a pattern of electrical signals applied through the plurality of electrodes.

13. The second article further comprises a plurality of switches, Each switch is independently addressable and has a first terminal and a second terminal. For each of the switches in the aforementioned plurality of switches, The first terminal of each switch is electrically coupled to the power supply. The second terminal of each switch is electrically coupled to the corresponding connector of the pair of second connectors. The system according to claim 7, wherein each switch is configured to supply power to the corresponding connector when power is supplied to the first terminal of each switch by the power supply.

14. The system according to claim 13, wherein in the first article, the plurality of electrodes are connected in a daisy-chain configuration, and when the first article and the second article are operably connected, the plurality of switches are configured to be independently addressable to selectively supply power to any of the plurality of electrodes.

15. When executed, at least, a) Receiving input indicating user gestures, b) Generating patterns of electrical stimulation for human body parts based on the user's gestures, c) Generating a control signal based on the pattern of electrical stimulation, d) Providing the pattern of electrical stimulation to the human body part by supplying current to at least one of the plurality of electrodes based on the control signal. The system according to claim 1, further comprising non-temporary computer executable code configured to cause the system to perform the above.

16. The system according to claim 10, wherein the analysis determines that the human subject is unable to effectively grasp the object.

17. The wearable is configured to fit the forearm, The stimulator is configured to be worn on the wrist, The motion sensor of the stimulator is configured to detect the movement of the wrist. The system according to claim 12, wherein the non-temporary computer-executable code, when executed, is configured to track the movement of the wrist in response to electrical stimulation of the forearm and to design a physiotherapy regimen at least in part based on the movement of the wrist in response to electrical stimulation of the forearm.

18. The system according to claim 1, wherein the first article further comprises a return electrode configured as a frame around the plurality of electrodes.

Images