Control system for movement reconstruction and / or restoration of a patient

Abstract

The present application relates to a control system (10) for a movement reconstruction and / or recovery system for a patient (P), comprising at least one sensor (12), at least one controller (14), at least one programmer (16), and at least one stimulation system (18), wherein the controller (14) is connected to the sensor (12), the programmer (16), and the stimulation system (18), wherein the sensor (12) is part of a training entity (22) or is connected to the training entity (22) in order to create and / or guide a movement model of the patient (P) and / or adjust stimulation settings based on input from the sensor (12).

Description

Technical Field

[0001] The present application relates to a system for controlling a motor reconstruction and / or recovery system, for example in a patient, for example in the field of improving recovery of neurological diseases, such as post-traumatic spinal cord injury (SCI). Background Art

[0002] Decades of physiological research have shown that the mammalian spinal cord is embedded with sensorimotor circuits that generate movement primitives (see Bizzi E. et al., Modular organization of motor behavior in the frog's spinal cord, Trends in neurosciences 18, 442-446 (1995); Levine AJ et al., Identification of a cellular node for motor control pathways, Nature neuroscience 17, 586-593 (2014)). These circuits process sensory information generated by moving limbs, as well as descending inputs from various brain regions, to produce adaptive motor behaviors.

[0003] Spinal cord injury (SCI) interrupts communication between the spinal cord and supraspinal centers, depriving these sensorimotor circuits of the excitatory and regulatory drive necessary to produce movement.

[0004] A series of studies in animal models and humans have shown that electrical neuromodulation of the lumbar spinal cord using epidural electrical stimulation (EES) can (re)activate these circuits. For example, EES has restored coordinated movement in animal models of SCI and isolated leg movement in individuals with motor paralysis (see van den Brand R. et al., Restoring Voluntary Control of Locomotion after Paralyzing Spinal Cord Injury, Science 336, 1182-1185 (2012); Angeli C. A. et al., Altering spinal cord excitability enables voluntary movements after chronic complete paralysis in humans. Brain: a journal of neurology 137, 1394-1409 (2014); Harkema S. et al., Effect of epidural stimulation of the lumbosacral spinal cord on voluntary movement, standing, and assisted stepping after motor complete paralysis: a case study). Movement, standing, and assisted stepping after motor complete paraplegia: a case study, The Lancet 377, 1938-1947 (2011); Danner SM et al., Human spinal locomotor control is based on flexibly organized burst generators, Brain: a journal of neurology 138, 577-588 (2015); Courtine G.et al., Transformation of nonfunctional spinal circuits into functional states after the loss of brain input, Nature neuroscience 12, 1333-1342, (2009); Capogrosso M et al., A brain-spine interface alleviating gait deficits after spinal cord injury in primates, Nature 539, 284-288 (2016).

[0005] EP 2 868 343 A1 discloses a system for delivering adaptive spinal cord electrical stimulation to promote and restore movement after neurological movement disorders. In particular, a closed-loop system for real-time control of epidural electrical stimulation is disclosed, the system comprising a device for applying neuromodulation with adjustable stimulation parameters to a subject, the device being operably connected to a real-time monitoring component comprising a sensor, the sensor continuously obtaining a feedback signal from the subject. The feedback signal provides the characteristics of the subject's movement, wherein the real-time monitoring component is operably connected to a signal processing device that receives the feedback signal and operates a real-time automatic control algorithm. The known system improves the consistency of walking in patients with neurological movement disorders. A real-time automatic control algorithm is used, the algorithm comprising a feedforward component using a single input single output model (single input-singleoutput model, SISO) or a multiple input single output (multiple input-single output, MISO) model. See also Wenger N. et al., Closed-loop neuromodulation of spinal sensorimotor circuits controls refined locomotion after complete spinal cord injury, Science Translational Medicine, 6, 255 (2014).

[0006] WO 2002 / 034331 A2 discloses a non-closed-loop implantable medical device system, which includes an implantable medical device and a transceiver device, wherein the transceiver device exchanges data with the patient between the patient and the implantable medical device and between a remote location and the implantable medical device. A communication device coupled to the transceiver device exchanges data with the transceiver device and the implantable medical device through a receiver device, and exchanges data between the transceiver device and the remote location to achieve bidirectional data transmission between the patient, the implantable medical device, the transceiver device, and the remote location. A converter unit converts the transmission of data from a first telemetry format to a second telemetry format, and a user interface enables information to be exchanged between the transceiver device and the patient, between the implantable medical device and the patient, and between the patient and the remote location through the transceiver device.

[0007] EP 3 184 145 A1 discloses a system for selective spatiotemporal electrical neural stimulation of the spinal cord. A signal processing device that receives signals from a subject and operates a signal processing algorithm to perform detailed stimulation parameter settings is operably connected to an implantable pulse generator (IPG), which receives stimulation parameter settings from the signal processing device and is capable of simultaneously transmitting independent current or voltage pulses to one or more electrode arrays. The electrode array is operably connected to one or more multi-electrode arrays suitable for covering at least a portion of the spinal cord of the subject to apply selective spatiotemporal stimulation to the spinal cord circuit and / or dorsal root, wherein the IPG is operably connected to one or more multi-electrode arrays to provide multipolar stimulation. Such a system allows effective control of the motor function of a subject in need thereof by selectively stimulating the spinal cord, particularly the dorsal root, in spatiotemporal manner.

[0008] EP 2 652 676 A1 relates to a posture control for monitoring vital signs and reusing an accelerometer (or more precisely, the acceleration sensed by a body sensor) for user control of the body sensor. This is achieved by detecting predefined patterns in the acceleration signal that are independent of other movements of the patient. These include tapping / tap sensors, shaking and rotation sensors. A new process is described that makes it possible to reuse acceleration sensing for reliable posture detection without introducing many false alarms due to non-posture movements such as breathing, heartbeat, walking, etc. Similar solutions for user tapping detection are known from US 8,326,569 and US 7,742,037.

[0009] WO 2007 / 047852 A2 discloses systems and methods for interactive neural stimulation and / or chemical substance delivery to a patient. A method according to one embodiment of the present application includes influencing a target neural population of a patient by providing at least one of an electromagnetic signal and a chemical substance to the patient. The method may further include detecting at least one characteristic of the patient that is associated with the performance of an auxiliary therapeutic task for the patient, the performance of which is associated with influencing the target neural population. In addition, the method may include controlling at least one parameter based on the affected target neural population at least in part on the detected characteristic.

[0010] WO 2017 / 062508 A1 discloses a system for controlling a therapeutic device and / or an environmental parameter, the system comprising one or more wearable sensor devices that detect and report one or more physical, physiological or biological parameters of a person in an environment. The sensor device can transmit sensor data indicating a physical, physiological or biological parameter of a person to an external hub, which processes the data and, based on the sensor data, communicates with the therapeutic device to provide therapy (e.g., neuromodulation, neurostimulation or drug delivery). Summary of the invention

[0011] According to the current state of the art, subjects still cannot achieve autonomous control of movements. It is important to remember that patients are not robots and cannot and should not be stimulated and controlled like robots. Therefore, there is a lack of a system that can overcome the shortcomings of the prior art. In particular, a system is needed that does not stimulate patients like a robot. The purpose of applying stimulation is not to control the patient, but to support the patient in training and daily life activities. Therefore, the control system should support the patient's own natural control circuit composed of the brain, nervous system and sensory organs. This means that the control system should not, for example, adjust the stimulation parameters to force the patient's lower body movement to reach a given reference trajectory. On the contrary, the patient should be able to determine, for example, the walking cadence.

[0012] The object of the present application is to improve neurostimulation systems, for example in the field of improving recovery from neurological diseases (such as spinal cord injuries), for example after trauma, and in particular in the field of increasing control systems for motor reconstruction and / or recovery systems for patients.

[0013] According to the present application, this object is solved by a control system for a system for reconstruction and / or restoration of movement of a patient having the features of claim 1. According to this, a system for a system for reconstruction and / or restoration of movement of a patient comprises:

[0014] - at least one sensor,

[0015] - at least one controller,

[0016] - at least one programmer,

[0017] - at least one stimulation system,

[0018] wherein the controller is connected to the sensor, the programmer and the stimulation system, wherein the sensor is part of or connected to a training entity so that

[0019] - Create and / or coach movement models for patients, and / or

[0020] -Adjust stimulus settings based on sensor input.

[0021] The present application is based on the following basic idea: in the case of neuromodulation, especially neurostimulation, the system can be used to control the electrical stimulation parameters that define stimulation in the movement reconstruction and / or recovery system for the patient, wherein a training entity is used, which is not the patient itself, but another entity. In this way, more explicit and even remote training and rehabilitation can be performed. The use of general hardware concepts and sensors that are part of or attached to the training entity are combined into a strategy and can be used for patients equipped with the system, thereby allowing support of limbs, such as lower limb motor function of patients with complete or incomplete SCI can be rehabilitated and promote daily living activities. The training entity defines movement, including but not limited to gait phases, including the kinematics of the body and / or parts of the body, such as the lower body (legs and feet), upper body (trunk, head, arms, hands). Therefore, it is necessary to determine the body kinematics to estimate the movement.

[0022] In order to estimate the gait phase, it is particularly necessary to determine the kinematics of the lower body. This can be measured directly by connecting sensors to the body and / or parts of the body (including but not limited to the trunk and / or abdomen and / or limbs and / or parts of limbs), or by indirectly measuring the activation of muscles or by measuring the interaction between the body and / or parts of the body, such as limbs and / or parts of limbs and their surroundings (such as ground reaction forces or upper body movements). Based on this, the system can stimulate the spinal cord at the right position and the right time when the patient performs different tasks. This means that the control system will not, for example, adjust the stimulation parameters to force the patient's body and / or limbs (one or more) to move to a given reference trajectory.

[0023] Furthermore, the patient's overall feeling and well-being may be enhanced (eg pain management).

[0024] The programmer is an application installed on a mobile device that communicates with the controller. The therapist, physiotherapist, or patient uses the programmer to provide input to the controller, such as selecting, starting, and stopping tasks or configuring stimulation parameters.

[0025] The programmer should allow adjustment of the stimulation parameters of the task while the task is running. When all stimulation sites are developed and tuned, this enables the user to tune the stimulation without having to start and stop the task (which can be very tedious at the beginning of rehabilitation training).

[0026] Programmers include, but are not limited to, a physiotherapist programmer (PTP) and a patient programmer (PP), which are applications installed on a mobile device that communicates with the controller.

[0027] Neural stimulation can be achieved through electrical stimulation, optogenetics (optical neural stimulation), chemical stimulation (implantable drug pumps), ultrasound stimulation, magnetic field stimulation, mechanical stimulation, etc.

[0028] Known electrical stimulation systems use central nervous system (CNS) stimulation, in particular epidural electrical stimulation (EES), or peripheral nervous system (PNS) stimulation, in particular functional electrical stimulation (FES).

[0029] Epidural electrical stimulation (EES) is known to restore motor control in animal and human models, and more specifically has been shown to restore movement after spinal cord injury by artificially activating the neural network responsible for movement below the spinal cord injury. Capogrosso, M. et al., A Computational Model for Epidural Electrical Stimulation of Spinal Sensorimotor Circuits, Journal of Neuroscience, 33(49), 19326-19340 (2013); Courtine G. et al., Transformation of nonfunctional spinal circuits into functional states after the loss of brain input, Nat Neurosci. 12(10), 1333–1342 (2009); Moraud EM. et al., Mechanisms Underlying the Neuromodulation of Spinal Circuits for Correcting Gait and Balance Deficits after Spinal Cord Injury. Injury, Neuron, 89(4), 814–828 (2016). EES does not directly stimulate motor neurons, but rather stimulates afferent sensory neurons before entering the spinal cord. In this way, the spinal network responsible for movement is indirectly recruited through these afferents, restoring movement overall by activating the required muscle synergies.

[0030] To date, the peripheral nervous system (PNS) stimulation system used clinically is called functional electrical stimulation (FES), which provides electrical stimulation to the target muscle directly by stimulating its motor fibers (neuro-muscle stimulation) through surface electrodes, or through limited fixation reflexes (practically limited to withdrawal reflexes) or through transcutaneous stimulation of peripheral nerves. The resulting muscle fatigue makes FES unsuitable for use in daily life. In addition, the possibility of success is still limited by the cumbersome setup when using surface muscle stimulation, the needs in terms of selectivity are not met (when using transcutaneous nerve stimulation) and stability is lacking (when stimulating muscles, the placement of electrodes cannot be accurately reproduced every day due to the movement of electrodes due to clothing, sweating).

[0031] Neuromodulation and / or neurostimulation can be provided to a CNS with a CNS stimulation system and / or a PNS with a PNS stimulation system by the system. The CNS and PNS can be stimulated simultaneously, intermittently or as needed. These two complementary stimulation paradigms can be combined into a strategy and can be used by patients equipped with the system. For example, neuromodulation and / or neurostimulation of the CNS can be used to enhance and / or the patient's motor ability, particularly in a manner that supports the existing transmission mode of physiological signals in the patient's body, so that command signals for body movements, etc. are still provided by the patient's nervous system, and are only supported and / or enhanced or converted by the CNS stimulation system. The stimulation provided by the PNS system can be used to specifically bring and guide the stimulation signal to a specific peripheral nerve structure in order to trigger a specific movement and / or improve an existing movement. Such PNS stimulation can be used to improve and / or complete movement and / or the patient's motor ability. This can be used to complete, for example, but not limited to, bending or extending, lifting, rotating, etc., of the patient's toes, fingers, arms, feet, legs or any limbs. This can be done, for example, when it is realized that the neuromodulation and / or neurostimulation provided by the CNS stimulation system is insufficient to complete the patient's movement or desired state. This movement or desired state can then be completed or supported by the stimulation provided by the PNS stimulation system. PNS stimulation can also be used to reduce side effects or compensate for the inaccuracy of CNS stimulation.

[0032] EES can be phasic or tonic, selective PNS is always phasic. Phasic is defined as locking to a defined event in the sensed signal (decoded intention, continuous decoding, onset of muscle activity, onset of movement, events during a defined movement (e.g., foot lift-off or foot strike in gait)).

[0033] Through PNS stimulation, stimulation of the upper extremity nerves, namely the radial nerve, ulnar nerve and / or median nerve, can be provided. Similarly, lower extremity nerves such as the sciatic nerve and / or femoral nerve can be provided through PNS stimulation. All PNS stimulation can be accomplished by targeting one of the above nerves using either intraneural electrodes (lateral or longitudinal) or supraneurial (cuff) electrodes.

[0034] With CNS stimulation, the following neural structures can be stimulated: For upper extremity movement, the cervical spinal cord or hand / arm motor cortex can be stimulated with a CNS stimulation system. For lower extremity movement, the lumbar sacral spinal cord can be stimulated. All of these nerves can be targeted with epidural, subdural, or intraspinal / intracortical stimulation.

[0035] Both PNS and CNS stimulation systems can include implantable pulse generators (IPGs). IPGs can be used to provide the necessary stimulation currents and signals for the CNS stimulation system and the PNS stimulation system.

[0036] The IPG generates stimulation pulses that are delivered by a lead with multiple electrodes to the stimulation side, such as the spinal cord. For EES, the lead is located in the epidural space (i.e., outside of the dural sac, which surrounds the spinal cord and the cerebrospinal fluid in which the spinal cord "floats"), the top of the spinal cord (including, but not limited to, bidirectional T12, L1, L2, L3, L4, L5, and S1 segments).

[0037] It is also possible to provide two separate IPGs, one for the PNS stimulation system and one for the CNS stimulation system.

[0038] It is also possible to use an external stimulator, in particular for PNS stimulation. The stimulation parameters for PNS stimulation and EES stimulation may be frequency, amplitude, pulse width, etc.

[0039] The system can support open-loop or closed-loop stimulation modes. Open-loop stimulation can be performed while performing a predefined fixed stimulation without adapting to, for example, the patient's movements. The stimulation settings (i.e., electrode configuration - meaning, at what time, frequency and amplitude, through what electrodes to deliver the stimulation, to stimulate what functional muscle blocks) are determined entirely by the therapist or physiotherapist and / or field engineer and / or by an algorithm. The control system may interfere with the patient's natural feedback loop to achieve smooth movements, for example: a normal gait cycle comparable to that of a healthy subject. Closed-loop walking can be performed, in which feedback is used to adjust the stimulation to the patient's gait. Closed-loop cycling can be performed, in which feedback is used to adjust the stimulation to the patient's cycling phase.

[0040] The interfaces Wireless Sensor Network WSN (wireless link between sensor and controller), Communication COM (wireless link between programmer and controller) and Telemetry TEL (wireless link between stimulation system and controller) connect the various subsystems in the control loop.

[0041] The programmer is a mobile device, the stimulation system is implanted, the controller is wearable, and the sensors are attached to the patient's body and / or one or more parts of the patient's body and / or the patient's limbs / feet, or to a bicycle crank and / or any other training apparatus for any other type of exercise, including but not limited to rowing, stepping, and / or swimming. Therefore, these interfaces all need to be wireless.

[0042] The training entity may be a trainer and / or a physical therapist.

[0043] In particular, the patient's and / or physiotherapist's shoes can be equipped with sensors. The sensors can be placed on top of the instep of the patient's and / or physiotherapist's shoes, and / or behind the heel and / or under the heel of the shoe (e.g., in a pocket in the sole or as an inlay sole).

[0044] Furthermore, the training entity may be or may include a training device, wherein the device is at least one of an exoskeleton, a robot, a treadmill, a bicycle exercise machine and / or a body weight support system. In particular, the patient and / or the subject's trainer may be equipped with at least one sensor or a plurality of sensors.

[0045] The controller may be configured and arranged for tracking and estimating the motion of the training entity and for converting it into stimulation data based on the estimated motion, which stimulation data is provided to the patient by the stimulation system for motion reconstruction and / or rehabilitation training of the patient.

[0046] Specifically, the controller is a wearable platform that executes control software. The controller processes data acquired from sensors, stimulation systems, and programmers, and programs the stimulation system to deliver the correct stimulation.

[0047] Furthermore, the controller may be configured and arranged to perform tracking and estimation of motion online and / or offline.

[0048] Online tracking and estimation helps to achieve direct transmission of the movement of the training entity and is used to convert it into stimulation data provided to the patient by the stimulation system for movement reconstruction and / or rehabilitation training of the patient. This is helpful for achieving real-time solutions and real-time data transmission.

[0049] Here, real-time is defined as end-to-end latency of less than 100 ms, preferably less than 50 ms.

[0050] Offline configuration by performing the tracking and estimation process offline may allow the controller to program the stimulation system over a period of at least one complete motion (e.g., a gait cycle) based on recorded sensor data. Performing offline tracking and estimation may allow for use of conditions that cannot be used in real time.

[0051] The motion model can be created so that the motion phase always takes the same value in the same event. This can allow offline determination of different motion events of the recorded data using robust criteria that are common to all types of healthy or pathological motion.

[0052] At the beginning of the rehabilitation course, the motion model used is a general motion model trained on a different set of subjects, so the motion model (e.g. gait model) is not perfect but sufficient to take some steps. All analyses are recorded in parallel in the sensor buffer. Once a whole motion cycle (e.g. gait cycle) has been detected, the online expert system determines past motion events (e.g. gait events) and trains the motion model to adapt to the new data. Then, the motion model used online is updated.

[0053] When the movement model is sufficiently refined, the learning process can be stopped and it can be stored for further sessions with the same patient.

[0054] Tracking and estimating the patient's movements on-site may allow tracking and estimating the patient itself. Tracking and estimating the movements remotely may allow the movements of the training entity and / or the patient and / or the physical therapist to be replicated to more patients simultaneously.

[0055] In particular, the controller may allow tracking and estimation to be performed from one patient to another. This may be achieved, for example, by transferring settings from one patient to another. In particular, settings used for one patient may be used for the treatment of another patient. In particular, settings from a healthy person may be used for the treatment of a patient.

[0056] In particular, the controller may allow tracking and estimation of the steps performed and / or transfer from a patient who is progressing in the rehabilitation process to a less progressed patient.

[0057] Furthermore, the controller may be configured and arranged to track and estimate online and / or in real time and / or with a time delay.

[0058] In addition to applying the correct electric field at the right location, the stimulation also needs to be applied at the right time and sequenced correctly. For a patient equipped with the system, it can be very helpful to stimulate at or close to the moment of performing, for example, a desired movement. The patient needs to be able to predict when the stimulation will occur in order to take full advantage of the stimulation. Similarly, suppressing movement while providing stimulation also requires the patient to know when to expect stimulation. When the stimulation is not synchronized with the patient's (expected) movement, the patient will not be able to perform the appropriate movement. This means that the stimulation needs to be predictable to the patient because the patient needs to synchronize with the stimulation.

[0059] In particular, real-time can be understood in such a way that the delay between the sensing signal and the provided stimulation signal does not exceed 30 ms (see also WO 2016 / 0279418 A1). For the present application and its preferred embodiments, real-time control, i.e., in particular, the delay between the sensing signal and the provided stimulation signal should not be greater than 30 ms, is beneficial for both open-loop and closed-loop methods.

[0060] There is a delay from nerve stimulation (e.g., spinal cord) to muscle activation. In particular, the delay value varies depending on the type of muscle. The controller can adapt to this different signal delay.

[0061] There may be a sensor network formed of at least two or more sensors, wherein at least one of the two or more sensors is connected to a controller.

[0062] Using a sensor network of two or more sensors, limb position estimates (e.g. lower limb position estimates) can be obtained by integrating the measured acceleration twice in combination with drift correction. However, position estimates of the trunk and / or head can also be obtained. In this way, for healthy subjects, non-real-time reconstruction of the trajectories of limbs and / or parts of limbs and / or trunk and / or head can be completed with an accuracy of several centimeters. In particular, movements (e.g. gait phase and cadence) can be estimated using two sensors.

[0063] In particular, two or more sensors may be placed on a foot and / or another part of a leg, including but not limited to the patient's calf and / or thigh and / or buttocks and / or other parts of the body, including but not limited to the torso and / or one or both arms and / or one or both hands and / or another part of an arm and / or the head and / or neck to provide an accurate description of movement.

[0064] More sensors can show different topologies, including but not limited to star networks, body networks, and chain networks. Using more sensors located on a chain, for example, the relative positions of all leg joints from the hip to the foot via the thigh, knee, and calf, the complete kinematics of the lower body can be reconstructed, including the trajectory of the foot, and the angles of the knee and hip joints. In general, more sensors can be placed on a chain from head to toe to determine human kinematics.

[0065] The relative position estimate can be drift-free.

[0066] The control system may further comprise an augmented reality and / or virtual reality module, which is configured and arranged to provide information related to motion reconstruction and / or restoration, in particular information related to training of motion reconstruction and / or restoration to be performed or being performed.

[0067] By simulating real-life activities, patients may be able to perform rehabilitation exercises in settings that are not normally possible to create in a hospital setting.

[0068] In particular, augmented reality and / or virtual reality modules can be designed to assist the rehabilitation process and track the patient's progress. Augmented reality and / or virtual reality modules can improve, for example, coordination, balance, muscle strength, range of motion, reaction time, memory.

[0069] The augmented reality and / or virtual reality modules can use different technologies, including virtual reality headsets (including, for example, gyroscopes, accelerometers, structured light systems, eye tracking sensors, etc.), glasses, heads-up displays, bionic contact lenses, virtual retinal displays, head-mounted displays, eye traps (EyeTrap), handheld displays, spatial augmented reality, etc.

[0070] Additionally, the augmented reality and / or virtual reality modules may be configured and arranged to provide gamified information related to motion reconstruction and / or recovery.

[0071] In particular, during rehabilitation sessions, the patient's body movements can be transferred to the game world.

[0072] In certain cases, this feature can facilitate and positively influence the progress of recovery through the patient's control over his or her body and / or certain parts of his or her body (e.g., limbs). Synchronizing all the expressive capabilities of the body with, for example, a digital avatar can allow the patient to analyze his / her movements in an individual training environment.

[0073] At least one sensor may be or may include at least one of an inertial measurement unit (IMU), an optical sensor, a camera, a piezoelectric component, a velocity sensor, an accelerometer, a magnetic sensor, a torque sensor, a pressure sensor, a displacement sensor, a contact sensor, an EMG measurement unit, a goniometer, a Hall sensor and / or a gyroscope and / or a motion tracking camera or an infrared camera.

[0074] Some sensors may require fixed base stations in the environment, including but not limited to magnetic sensors or infrared sensors.

[0075] Electromagnetic position sensors, optical sensors, and cameras can estimate 3D position and orientation.

[0076] In particular, magnetic sensors and magnetic field sensors can be combined in shoes to walk on magnetic sensor plates, or inserted into treadmills or gait phase detection devices. Magnetic force can be detected and acquired by magnetic sensors under gait training.

[0077] Torque sensors can be placed on bicycle cranks to assess torque during cycling.

[0078] Some sensors may be worn by the patient without a fixed base station, including but not limited to piezoelectric components, pressure sensors, and / or torque sensors.

[0079] The velocity sensor can monitor linear and angular velocity and detect motion. The 3-axis gyroscope can estimate 3D angular velocity.

[0080] The IMU can measure and report 3D acceleration, 3D angular velocity, and 3D direction using a combination of one or more of the accelerometers, one or more of the gyroscopes, and one or more of the magnetometers. Optionally, a temperature sensor can also be included to compensate for the effects of temperature on the sensor readings. By integrating the angular velocity evaluated by the one or more gyroscopes and fusing with the data from the one or more accelerometers (Kalman filter), the movement and / or angle of, for example, the calf, thigh, foot, arm and / or hand and / or torso and / or head can be accurately measured. For healthy subjects, the movement and / or angle may have a regular and characteristic pattern, but not for injured patients. Based on these measurements, the direction of the IMU relative to the fixed world can be accurately estimated using standard sensor fusion algorithms.

[0081] In order to estimate movement, for example gait phase, it is necessary to determine (lower) body kinematics. The sensor collects motion data, based on which the motion phase, for example gait phase or foot phase, is determined in real time. This operation can be done directly by connecting the sensor to the training entity and / or by the sensor as part of the training entity, or indirectly by measuring muscle activation or by measuring the interaction between the body and / or body parts and its / their surroundings and / or by connecting the sensor to the patient's body (for example, connected to the head and / or neck and / or trunk and / or one or more limbs and / or one or more parts of the limbs). Therefore, the sensor is able to determine motion events, for example gait events with a standard that is universal for various healthy or pathological gaits.

[0082] The acceleration and orientation of the body and / or parts of the body, such as the hips, thighs, calves, feet, arms, hands, torso, head, can be sampled at a high enough rate and low enough latency so that the sampled acceleration and direction known to the system closely match the true acceleration and direction of the foot.

[0083] In some embodiments, the sensor setup can operate anywhere in daily life, is not limited to a specific location (such as when using a camera), and can be independent of an assistive device (e.g., a body weight support system, a walker, a cane, etc.).

[0084] The use of pressure or contact sensors may allow direct measurement of the essence of stance, the weight-bearing phase of the gait cycle.

[0085] The EMG measurement unit can sense muscle activity with the aid of surface or intramuscular EMG electrodes for flexor and extensor muscles.

[0086] The patient's kinematics may be sensed directly or indirectly with the aid of variables such as kinematic markers.

[0087] In the case of closed-loop walking, the sensors may be worn on the legs and / or feet and / or torso and / or head and / or arms, or in the case of closed-loop cycling, a single sensor may be attached to a bicycle crank or to one or both feet of the patient.

[0088] Thus, for closed-loop riding, the stimulus may be determined by crank angle and / or foot angle.

[0089] During the stance phase, pressure sensors and piezoelectric components can sense the pressure distribution on the sole of the foot, the force on the ground, and the torque on the bicycle crank. The swing and stance phases can be estimated, as well as events such as foot strike, heel-off, toe-off, and applied force.

[0090] Furthermore, the training entity may be the patient himself.

[0091] Often, the controller may be integrated into the stimulation system. In addition, the programmer may be integrated into the controller or vice versa.

[0092] Additionally, the control system may include an early warning module configured and arranged to provide an early warning signal indicative of an impending stimulation event.

[0093] Adjusting the gait to a predetermined reference value interferes with the patient's autonomous movements. In particular, the patient's autonomous movements may have a significant impact on movement, because the patient's autonomous control can regulate the activation of muscles. Therefore, despite the same stimulus, the movement pattern of an injured or impaired subject may still be different compared to a healthy subject. Early warning signals can help patients adjust their autonomous control to the planned individual movements, so that conventional movements can be performed. Early warning signals can be, for example, audio and / or visual and / or sensory and / or tactile signals. Early warning signals can include, but are not limited to, sound signals, vibrations, light signals, smells, tastes, pain, temperature (warm, cold), humidity, dryness, etc.

[0094] In particular, the warning signal may act in the sub-motor threshold region, which elicits a sensory rather than a motor response.

[0095] In the following, it is determined which control output parameters exist and their influence on the afferent nerves and their ultimate influence on muscle activation are described. Based on this, it can be selected which output parameters the control system will control. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Further details and advantages of the present application will now be disclosed in conjunction with the accompanying drawings.

[0097] It is displayed in:

[0098] Figure 1 is a schematic, greatly simplified illustration of a stimulation pulse delivered by a system according to the present application;

[0099] Figure 2A is the variation of the current required to trigger an action potential in a nerve fiber and the required charge with pulse width (using a square wave);

[0100] Figure 2B is the variation of the current required to trigger an action potential in a nerve fiber and the required charge with pulse width (using a square wave);

[0101] Figure 3 is a table that describes the fiber type, diameter, and function;

[0102] Figure 4 It is the relationship between the reaction delay and the internal muscle reaction delay;

[0103] Figure 5 is a table detailing the intended movement as well as the agonist muscles involved and the antagonist muscles involved;

[0104] Figure 6 It is a discrete set of functional muscle blocks (FMB) and custom muscle blocks (CMB);

[0105] Figure 7 is the general layout of an embodiment of a control system for a patient's motion reconstruction and / or restoration system according to the present application;

[0106] Fig. 8A is a schematic diagram of a patient equipped with an exoskeleton according to the present application, the skeleton and Figure 7 Related to the embodiments disclosed in;

[0107] Figure 8B Is equipped with Figure 7 A stereoscopic view of a patient of a control system according to the present application disclosed in , the control system comprising two sensors;

[0108] Figure 8C Is equipped with Figure 7 A stereoscopic view of a patient with a control system disclosed in the disclosure, the control system comprising seven sensors;

[0109] Fig.8D is a perspective view of a sensor insole according to the present application; and

[0110] Fig. 8E Is equipped with Figure 7 A stereoscopic view of a patient with a control system according to the present application disclosed in , the control system comprising one IMU and one pressure insole for each foot.

[0111] Fig. 9 is a schematic diagram of a patient and a trainer (rehabilitation specialist) according to the present application;

[0112] Fig.10 is a schematic diagram of a patient and a remote trainer according to the present application;

[0113] Fig.11 is a flow chart of an offline workflow of a control system according to the present application;

[0114] Fig.12 is a flow chart of the online workflow of the control system according to the present application; and

[0115] Fig.13 Is equipped with Figure 7Schematic diagram of the control system for the patient's foot inclination / forward acceleration is shown.

[0116] [List of Reference Numerals]

[0117] 10 Control System

[0118] 12 Sensors

[0119] 12a IMU

[0120] 12b Pressure sensor

[0121] 12c Sensor Network

[0122] 14 Controller

[0123] 16 Programmer

[0124] 18 Stimulation System

[0125] 20 Wire

[0126] 22 Training Entity

[0127] 22a Exoskeleton

[0128] 24 Electrode Module

[0129] 24a Electrode

[0130] 24b Electrode

[0131] 100 Offline Workflow

[0132] 200 Online Workflow

[0133] 300 Sensor Insole

[0134] di Remote

[0135] m Motion Model

[0136] pr proximal

[0137] L position

[0138] L1 Position

[0139] Patient P

[0140] S Shoes

[0141] T Trainer

[0142] T1 Remote Trainer

[0143] CMB Custom Muscle Blocks

[0144] COM communication

[0145] EES Epidural Electrical Stimulation

[0146] FMB Functional Muscle Block

[0147] WL Wireless Link

[0148] WSN interface Wireless Sensor Network

[0149] STP Space-Time Programmer

[0150] TEL Telemetry

[0151] LVLat Left vastus lateralis

[0152] RVLat right vastus lateralis

[0153] Lll Left iliopsoas

[0154] Rll right iliopsoas

[0155] LRF Left rectus femoris

[0156] RRF Right rectus femoris

[0157] LST Left semitendinosus

[0158] RST right semitendinosus

[0159] LTA Left tibialis anterior

[0160] RTA Right tibialis anterior

[0161] LMG left medial gastrocnemius muscle

[0162] RMG right medial gastrocnemius muscle

[0163] LSol left soleus muscle

[0164] RSol Right soleus

[0165] LFHL Left flexor hallucis longus

[0166] RFHL Right flexor hallucis longus DETAILED DESCRIPTION

[0167] Note that in the following we mainly refer to CNS / EES stimulation. A person skilled in the art can convert the stimulation parameters to PNS / FES stimulation.

[0168] The control system can provide stimulation data for motor reconstruction and / or restoration using current pulses for stimulating afferent nerve fibers. Based on this starting point, the following stimulation parameters can be determined:

[0169] -Electrode configuration (electrodes to use, polarity)

[0170] - Stimulus (pulse) amplitude

[0171] - Stimulus (pulse) width

[0172] - Stimulation (pulse) frequency

[0173] Figure 1 A very simplified schematic diagram of a stimulation pulse is shown, illustrating pulse amplitude, pulse width and pulse frequency. Each stimulation pulse is followed by a neutralization pulse or neutralization period (not shown) to remove electrical charge from the tissue to avoid tissue damage.

[0174] The role of each stimulation parameter is described below.

[0175] Electrode Configuration: Stimulating a specific muscle group requires the application of a specific electric field at a specific location in the spinal cord, either directly by stimulating the muscle fibers (neuromuscular stimulation), or by limited fixed reflexes or by percutaneous stimulation of peripheral nerves. Thus, in the present control system, the electrical stimulation can be delivered, for example, via a wire with multiple electrodes connected to the spinal cord. The location, shape and direction of the generated electric field can be altered by selecting different electrode configurations for transmitting the current (which electrodes to use, which polarity and potential to use). Thus, the electrode configuration can determine, for example, which spinal root the stimulation is delivered to, and subsequently which muscle or muscle group activity will be enhanced as a result.

[0176] Pulse amplitude and pulse width: Figure 2A and Figure 2B In, the current and charge required to trigger an action potential in a nerve fiber are shown as a function of pulse width (using a square wave) (see: Merrill DR. et al., Electrical Stimulation of excitable tissue: design of efficient and safe protocols, J Neurosci Methods 141(2): 171-98 (2005)). Figure 2A and Figure 2B Also shown are the rheobase current I_rh, ie the current required if an infinite pulse width were used, and the chronaxiie time t_c, which is the pulse width required at a current of 2I_rh.

[0177] See also Figure 2B , although a larger current may be required at a smaller pulse width, the total required charge may decrease as the pulse width decreases. Therefore, shorter pulses with higher current amplitudes may be energetically beneficial.

[0178] Since smaller diameter fibers may require higher currents, for smaller diameter nerves, Figure 2A The current-pulse width curve will be shifted. Therefore, since nerve fibers with smaller diameters can also be activated (until saturation), higher currents can activate more nerve fibers. However, cross-talk will also increase, since more neurons in neighboring roots may also be activated. Fortunately, as Figure 3 As shown, afferent fibers involved in motor control (fiber types Ia and Ib) may all be relatively large (12-20 μm), while fibers involved in touch, temperature, and pain feedback (which should not be triggered) may be relatively small (0.5-12 μm). Therefore, as pulse width and / or current amplitude increases, type Ia and Ib fibers may be recruited first. This may allow (most) relevant fibers to be recruited while keeping crosstalk and patient discomfort to a minimum.

[0179] Pulse frequency: Assuming that each pulse delivers enough charge to trigger an action potential, the pulse frequency determines the frequency of action potentials generated in the afferent nerve. Since no new action potentials will appear in the refractory nerve, the frequency of triggering action potentials will saturate at high pulse frequencies. The saturation point of afferent fibers is usually around 200Hz (Miller JP. et al., Parameters of Spinal Cord Stimulation and Their Role in Electrical Charge Delivery: A Review, Neuromodulation: Technology at the Neural Interface 19, 373-384, (2016)). However, because the total charge delivered per unit time (i.e., charge per second) can be increased by increasing the frequency without changing the current amplitude or pulse width, stimulation at frequencies above the saturation point may still be beneficial (Miller JP. et al., Parameters of Spinal Cord Stimulation and Their Role in Electrical Charge Delivery: A Review, Neuromodulation: Technology at the Neural Interface 19, 373-384, (2016)).

[0180] Pulse Positioning: Many tasks, including walking, require the simultaneous activation of multiple muscle groups. Therefore, to support these tasks, multiple muscle groups may need to be stimulated simultaneously, each requiring a specific electric field and pulse frequency. When applied simultaneously, these different electric fields may affect each other, which may lead to unexpected and uncontrolled effects. Therefore, to avoid this, care should be taken not to apply individual stimulation pulses targeting different muscle groups and their neutralization periods simultaneously. This may not be considered a stimulation parameter, but it does identify a desired system function: the pulse positioning algorithm.

[0181] The previous section described the effects of stimulation parameters on the triggering of action potentials in afferent nerve fibers. Although triggering these action potentials is an essential step in treatment, ultimately, stimulation should enable or support the patient to perform specific lower body movements, which may require the activation of specific muscles or muscle groups. The effects of triggering action potentials in afferent nerve fibers on muscle activation may be filtered inside the spinal cord through spinal reflex circuits and regulated by the patient's voluntary control. Therefore, the effects of stimulation parameters on muscle activation may not be well understood and may be affected by intra- and inter-patient variability. The following aspects may be relevant in this regard:

[0182] - The level of voluntary control of the lower body may vary between patients, depending on the type and severity of the SCI lesion level and the (spontaneous) status of recovery.

[0183] - Stimulating afferent nerve fibers can assist or activate the corresponding muscles, but not necessarily strengthen the movement. Patients can modulate activation (e.g., take larger or smaller steps without changing the stimulation) or even completely resist the movement of the leg. This can vary from patient to patient and may change as recovery increases.

[0184] - Speculation: Since the spinal cord floats in the cerebrospinal fluid, the distance between the spinal cord and the wire electrode may vary (mainly depending on the patient's posture: prone - large distance, supine - small distance). Another hypothesis could be that due to changes in posture, the thickness of the low-conducting epidural fat layer between the wire electrode and the dura mater / cerebrospinal fluid changes, leading to changes in the impedance seen by the electrode and, therefore, to changes in the current / voltage of the stimulus delivered electronically. Therefore, the effect of the applied stimulation (both muscle action and saturation) may also vary with the patient's posture. Although this speculation has not been confirmed, patients can successfully exploit the described effect to modulate the intensity of the stimulation by changing their posture: bending forward decreases the intensity, bending backwards increases it.

[0185] - Although stimulation up to 500Hz is theoretically possible, pulse frequencies between 40 and 120Hz are likely to be used in most cases, as this may have benefits in terms of selective muscle activation and improved voluntary control in the patient.

[0186] In general, increasing pulse amplitude may not lead to increased recruitment of muscle fibers (with a corresponding increase in crosstalk), and increasing stimulation frequency may lead to increased activation of muscles without affecting crosstalk. However, increasing stimulation frequency may reduce the strength of natural proprioception and lead to decreased sensation in the patient's legs. This may be due to the collision of natural sensory input with the antidromic action potentials generated by electrical stimulation. At high frequencies (above 100 Hz), patients may even report a complete loss of sensation in their legs and "feel as if walking without legs". This is an uncomfortable situation for a patient who wants to take a confident step with each step and will not be able to trust that the legs he / she no longer feels will support him / her in the next stance phase. Therefore, it may be necessary to adjust the balance between stimulation amplitude and frequency to find the best compromise between crosstalk limitation and sensory loss. Simulations suggest that a possible solution may be to shift the stimulation to lower amplitude or even higher frequency, so as to trigger the same amount of activity in the spinal cord with the minimum number of stimulated fibers. This hypothesis needs to be verified by additional clinical data. Finally, it can also be identified that different patients require different stimulations, i.e., the optimal frequency and amplitude settings may vary greatly between patients. Therefore, the relationship between stimulation amplitude and frequency in muscle activation remains largely unclear. Furthermore, optimal stimulation settings may vary throughout the day and depend on the assistive device used (crutches, walkers, etc.), the improvement of recovery over time, and the purpose of the training or activity.

[0187] Timing: In addition to applying the correct electric field at the correct location in the spinal cord, the electric field may need to be applied correctly at the correct time and sequenced correctly. The relevant timing aspects that are determined are listed below.

[0188] - There is a certain delay from spinal cord stimulation to muscle activation (typical values ​​are around 0-30ms, depending on the muscle, see Figure 4 =, LVLat=left vastus lateralis, RLVat=right vastus lateralis, Lll=left iliopsoas, Rll=right iliopsoas, LRF=left rectus femoris, RRF=right rectus femoris, LST=left semitendinosus, RST=right semitendinosus, LTA=left tibialis anterior, RTA=right tibialis anterior, LMG=left medial gastrocnemius, RMG=right medial gastrocnemius, LSol=left soleus, RSol=right soleus, LFHL=left flexor hallucis longus, RFHL=right flexor hallucis longus).

[0189] - Although EES enables patients to perform movements, patients may need to be able to predict when stimulation will occur in order to take full advantage of it. Similarly, inhibiting movement while delivering stimulation also requires patients to know when to expect stimulation. Therefore, predictability of stimulation timing is essential.

[0190] - When the stimulation is not synchronized with the patient's (expected) movement, the patient may not be able to perform the appropriate movement. Here, this may mean that the stimulation needs to be predicted by the patient, because the patient needs to synchronize with the stimulation.

[0191] -The duration of the stimulation for leg swing during walking may need to be fine-tuned. For some patients, increasing the duration of this stimulation by 100ms can cause the patient to jump instead of taking appropriate steps.

[0192] 20ms may be sufficient resolution for adjusting stimulation timing (i.e., stimulation on / off timing for a particular muscle group may not need to be controlled with an accuracy of less than 20ms). Based on the currently available data, controlling timing with a resolution less than 20ms does not appear to improve stimulation effectiveness.

[0193] Based on the previous section, the stimulus parameters can be selected in the control system. This determines the control output space used and therefore the complexity of the control problem and the potential effectiveness of the control system.

[0194] We first discuss which parameter spaces can be reduced or eliminated. The remaining control output space is summarized as follows.

[0195] Electrode Configuration: Walking, as well as other movements of the lower limbs, may consist of well-coordinated flexion and extension of the lower limb joints (through contraction of the agonist muscles and relaxation of the antagonist muscles). Specific agonist and antagonist muscles for specific flexion and extension of joints can be grouped, and due to the limited number of joints, this means that only a small number of discrete muscle groups need to be stimulated. For each joint flexion and extension, a Space Time Programmer (STP) for programming the space and time of stimulation will enable the creation of the optimal electrode configuration to activate the agonist muscles while avoiding activation of the antagonist muscles (as well as avoiding activation of muscles on the opposite side). This can be done through a process called functional imaging. We define Functional Muscle Blocks (FMBs) as the final stimulation configuration for each specific muscle group. At least 12 specific FMBs for controlling the system have been identified and are listed together with the corresponding agonist and antagonist muscles in Figure 5 middle.

[0196] Because both knee flexion and hip extension involve the semitendinosus, it is physiologically impossible to target knee flexion or hip extension separately. Figure 5 This does not include knee flexion (which could be considered redundant with hip extension).

[0197] exist Figure 5Next to the 12 FMBs listed in the , it is also foreseen that a trainer / therapist / physiotherapist can create Custom Muscle Blocks (CMBs). Creating CMBs may be useful in cases where the trainer / therapist / physiotherapist wants to apply a stimulation that does not specifically target any of the 12 muscle groups targeted by the FMBs, or in cases where the trainer / therapist / physiotherapist wants to use a variation of one of them.

[0198] Therefore, by restricting the electrode configurations to a discrete set of FMBs and CMBs (as opposed to an infinite number of possible electrode configurations), the complexity of the control problem can be significantly reduced without significantly affecting the potential effectiveness of the control system. Figure 6 , the stimulus for the task is then reduced to stimulus for (a subset of) the predefined FMB and CMB. For this example, right trunk stability was used in both Task 1 and Task 2.

[0199] Functional imaging may require measurement using EMG sensors Figure 5 The functional imaging process is performed in separate sessions using STP for programming the space and time of stimulation, rather than, for example, performing it adaptively in the control system. Therefore, it is possible to consider providing the configuration of the FMB and CMB to the control system.

[0200] Pulse Width: From the perspective of triggering action potentials in afferent nerve fibers, the pulse width parameter and the pulse amplitude parameter may be closely related and together may determine which afferent nerve fibers are recruited. Increasing the pulse width can reduce the amplitude, and decreasing the pulse width can reduce the energy consumption (because the total charge required to trigger an action potential decreases as the pulse width decreases, see Figure 2B ), and stimulate more FMBs simultaneously or at a higher frequency. However, from a control perspective, these two parameters may be (almost) redundant, as increasing them may lead to the recruitment of more afferent nerve fibers over a larger area.

[0201] Pulse widths below the time axis time t_c may quickly require high currents (and therefore high voltages), which are difficult to generate and may cause patient discomfort. Figure 2AThe intensity-duration curve of is almost flat, so increasing the pulse width above t_c has little effect on the required amplitude while increasing the total power consumption. Also considering that having a fixed pulse width simplifies pulse positioning, a fixed pulse width is chosen (fixed at a value close to the time axis time t_c to keep both energy consumption and the required current amplitude low, where t_c ≈ 200 μs for human afferent dorsal root nerve fibers). This reduces the complexity of the control problem by reducing the number of output parameters.

[0202] This may result in the following stimulation parameters being controlled over time by the control system:

[0203] - How exciting is FMB

[0204] - Stimulation amplitude for each FMB

[0205] - Stimulation frequency for each FMB

[0206] Pulse locating may be considered a lower level issue and therefore may not be a direct output of the control system (system function). Pulse locating will be performed by the IPG.

[0207] Although it has been considered to combine amplitude and frequency into a single "intensity" parameter, this may not be ideal for a control system, as these parameters may have very different effects. Amplitude and frequency can be independent parameters when triggering action potentials in afferent nerve fibers: amplitude determines at which afferent nerve fibers action potentials are triggered, and frequency determines the rate at which they are triggered. Therefore, while it is not clear whether the independence of the two parameters also holds true for muscle activation due to the signal processing that occurs in the spinal cord, amplitude in principle determines which muscle fibers are activated and frequency determines the intensity. Moreover, it is clear that for some patients, varying amplitude may give the best results, while for others, frequency may be a more useful parameter.

[0208] Since the precise relationship between frequency and amplitude is unknown in clinical situations, it may not be advisable to combine frequency and amplitude into a single parameter. Therefore, stimulation frequency and amplitude may be controlled independently of each other.

[0209] In the following, the sensor, controller, programmer and stimulation system (eg, IPG) of the present application will be described in more detail.

[0210] Sensor: A battery-powered, wearable sensor (directly or indirectly, and / or placed on and / or integrated into one or more training entities) that collects motion data and sends it to the controller. Its intended use is to capture body motion parameters.

[0211] Controller: A battery-powered, wearable device that receives data (directly or indirectly) from the sensor(s) and is capable of sending stimulation commands to the IPG to accomplish a specific task (i.e., activity / training exercise). Its intended use is to determine the optimal stimulation settings for any given task and provide this information to the IPG. In addition, the device can take the IPG out of shelf mode, transcutaneously charge the IPG battery, and initiate IPG lead integrity testing.

[0212] Programmer: A programmer, or also called a clinical programmer, can be used to receive, inter alia, stimulation parameters, patient data, physiological data, training data, etc.

[0213] It may include, for example, a Space Time Programmer (STP) for programming the space and time of stimulation, a Physiotherapist Programmer (PTP) allowing a physiotherapist to adjust the stimulation, and a Patient Programmer (PP) allowing a patient to select a specific stimulation program.

[0214] The Space Time Programmer (STP), Physiotherapist Programmer (PTP) and Patient Programmer (PP) can be embodied as applications installed on a mobile device that communicates with the controller. They are used by a treating physician (TP), physiotherapist (PT) or patient to provide input to the controller, such as selecting, starting and stopping tasks or configuring stimulation parameters.

[0215] The programmer can adjust the stimulation parameters of a task while it is running. When all stimulation sites are developed and tuned, this enables the user to tune stimulation without having to start and stop the task, which is very tedious at the beginning of rehabilitation training.

[0216] Generally speaking, a programmer may have the following structure:

[0217] In a first embodiment, the programmer may be embodied such that it may in particular, but not limited to, receive stimulation parameters, patient data, etc., check and / or reprogram the stimulation data and send it back to, for example, a controller.

[0218] In this first embodiment, the programmer is able to receive data from an implanted (part of) system (e.g. a controller), display the data, receive input from the user, and then send it back to the controller. In other words: the programmer can receive, process and resend data.

[0219] In a second embodiment, the programmer may receive data from a remote database. The database may be linked to the stimulation system, for example, via a separate interface configured only for data transfer from the system to the database.

[0220] In a second embodiment, the programmer is able to receive data from a remote database, display the data, receive input from the user, and then send it to the controller. In other words: the programmer only connects to the controller to send data, it does not receive data from the controller or any implanted system components.

[0221] Stimulation system, here IPG: Implantable Pulse Generator, a battery-powered device implanted subcutaneously that can generate electrical stimulation. Its intended use is to deliver electrical stimulation to the leads based on commands received from the controller.

[0222] Figure 7 The general layout of an embodiment of a control system 10 for a motion reconstruction and / or restoration system for a patient P according to the present application is shown.

[0223] The control system 10 includes one or more sensors 12 , and two or more sensors may form a sensor network.

[0224] Furthermore, in the illustrated embodiment, the control system 10 includes a controller 14 .

[0225] Additionally, the control system 10 includes a programmer 16 .

[0226] There is also an implantable pulse generator (IPG) 18.

[0227] In an alternative embodiment, the pulse generator may also be a non-implantable pulse generator.

[0228] The control system may further include a wire 20 .

[0229] The lead 20 may be a connecting cable (ie, a lead cable) having one or more electrodes.

[0230] One or more electrodes may be arranged on a lead plate connected to a lead.

[0231] In one embodiment, controller 14 is wearable, programmer 16 is a mobile device, IPG 18 is implanted in the body, and one or more sensors 12 are attached to a limb / foot or trainer of the patient.

[0232] The training entity 22 may be a bicycle.

[0233] In one embodiment, the training entity 22 may be a trainer T and / or a physical therapist.

[0234] One or more sensors 12 are connected to a controller 14 .

[0235] The connection between the one or more sensors 12 and the controller 14 is a two-way connection.

[0236] In the illustrated embodiment, the connection between the one or more sensors 12 and the controller 14 is a direct connection.

[0237] However, an indirect connection (ie with another component in between) to the control system 10 is generally possible.

[0238] In the illustrated embodiment, the connection between the one or more sensors 12 and the controller 14 is established via a wireless network WSN.

[0239] However, cable bond connections are often also possible.

[0240] Furthermore, in the illustrated embodiment, controller 14 is connected to programmer 16 via a direct connection COM (also referred to as a “communication line”).

[0241] However, indirect connections are generally possible.

[0242] In the illustrated embodiment, the connection between controller 14 and programmer 16 is established via a wireless link.

[0243] However, cable bond connections are often also possible.

[0244] In the illustrated embodiment, the controller 14 is connected to the IPG 18 via a direct connection.

[0245] However, an indirect connection (ie with another component in between) to the control system 10 is generally possible.

[0246] In the illustrated embodiment, the connection between the controller 14 and the IPG 18 is established via a wireless link TEL.

[0247] However, cable bond connections are often also possible.

[0248] The IPG 18 is connected to a conductor 20 .

[0249] By means of one or more sensors 12 , signals indicative of movement (eg, movement of a limb position, such as a foot or hand, or movement of the torso or head or other part of the body) may be sensed and used by the control system 10 .

[0250] The sensor signals are transmitted to the controller 14 and processed there.

[0251] Controller 14 processes data from, for example, sensors 12 , IPG 18 , and programmer 16 .

[0252] The control software is executed by means of the controller 14 .

[0253] Inputs (eg, selecting, starting and stopping tasks or configuring stimulation parameters) are provided to the controller 14 by means of the programmer 16 .

[0254] Typically, programmer 16 may allow adjustment of stimulation parameters of a task while the task is running.

[0255] Programmer 16 may typically be used by a therapist, physical therapist, or patient.

[0256] Controller 14 programs IPG 18 to deliver the correct stimulation via leads 20 .

[0257] Stimulation, here EES, may be provided via leads 20 and respective electrode(s).

[0258] Alternatively, other suitable stimulation signals may also be provided.

[0259] In particular, PNS stimulation may also be provided.

[0260] In particular, PNS stimulation may be provided by the IPG 18 .

[0261] In general, control system 10 creates and / or directs a movement model m of the patient and / or adjusts stimulation settings based on input from sensors 12 .

[0262] However, external stimulation systems are often also possible.

[0263] exist Figure 7 Not shown in greater detail is the fact that one or more sensors 12 are part of the training entity 22 or are attached to the training entity 22 .

[0264] In an alternative embodiment, the training entity 22 may also be the patient P himself.

[0265] The controller 14 may also track and / or estimate the motion of the training entity 22 to convert the motion of the training entity 22 into estimated motion-based stimulation data, which is provided to the patient by the stimulation system 18 for patient training.

[0266] exist Figure 7 Not shown, the control system 10 may include an early warning module configured and arranged to provide an early warning signal indicative of an impending stimulation event.

[0267] In particular, the warning signal may act in the submotor threshold region that elicits a sensory but not a motor response.

[0268] exist Figure 7 Not shown in the drawings, at least one sensor 12 is an inertial measurement unit (IMU).

[0269] In alternative embodiments, at least one sensor 12 may also be an optical sensor, a camera, a piezoelectric component, a velocity sensor, an accelerometer, a magnetic field sensor, a torque sensor, a pressure sensor, a displacement sensor, an EMG measurement unit, a goniometer, a magnetic position sensor, a Hall sensor, a gyroscope and / or a motion tracking camera or an infrared camera.

[0270] exist Figure 7 Not shown, the control system 10 may further include an augmented reality and / or virtual reality module, which is configured and arranged to provide information related to motion reconstruction and / or restoration, in particular information related to training for motion reconstruction and / or restoration to be performed or being performed.

[0271] Figure 7 Not further shown in the figures, the augmented reality and / or virtual reality module may be configured and arranged to provide gamified information related to motion reconstruction and / or restoration.

[0272] exist Figure 7 Not further shown in the drawings, controller 14 and / or programmer 16 may generally be integrated into IPG 18, or vice versa. Furthermore, programmer 16 may be integrated into controller 14, or vice versa.

[0273] In general, each individual component of the control system 10 may be integrated into any other component of the control system 10 .

[0274] According to the current state of the art, subjects still cannot achieve voluntary control of movements. It is important to remember that patients are not robots and cannot and should not be stimulated and controlled like robots. Therefore, there is a lack of a system that can overcome the shortcomings of the current state of the art. In particular, there is a need for a system that does not stimulate the patient like a robot. The purpose of applying stimulation is not to control the patient, but to support the patient in training and daily life activities.

[0275] Therefore, the control system 10 should support the patient's own natural control loop consisting of the brain, nervous system and sensory organs. This means that the control system should not, for example, adjust the stimulation parameters to force the patient's lower body movement to reach a given reference trajectory. Instead, the patient should be able to determine, for example, the walking cadence.

[0276] Fig. 8A It is shown that according to the present application Figure 7 The embodiments disclosed herein relate to a patient equipped with an exoskeleton.

[0277] In this embodiment, the patient P is equipped with Figure 7 The control system 10 and the training entity 22, in particular the exoskeleton 22a, disclosed in the disclosure.

[0278] The exoskeleton 22a in this embodiment is an external structure designed around the shape and function of the patient's P lower torso, particularly the patient's P legs.

[0279] However, in alternative embodiments, the exoskeleton 22a may also be designed around the patient's P torso and / or neck and / or head and / or arms.

[0280] In alternative embodiments, the exoskeleton 22a may also be designed around the entire body of the patient P.

[0281] One or more sensors 12 are placed on the exoskeleton 22a to assess leg kinematics.

[0282] In an alternative embodiment, the sensor 12 may also be integrated into the exoskeleton 22a.

[0283] according to Figure 7 With the aid of one or more sensors 12 attached to the exoskeleton 22a, information indicating movement, such as movement of the patient's P body and / or the position of one or more parts of the patient's P body (here the legs and / or feet), can be sensed and used by the control system 10.

[0284] The controller 14 tracks and / or estimates the motion of the exoskeleton 22a and converts the motion of the exoskeleton 22a into stimulation data based on the estimated motion, which is provided to the patient by the stimulation system 18 for patient training.

[0285] In an alternative embodiment, the system and exoskeleton 22a may be remotely controlled.

[0286] exist Fig. 8A Not shown in greater detail, other embodiments of the training entity 22 may generally include a robot, a treadmill, a bicycle machine, and / or a body weight support system, among others.

[0287] exist Fig. 8A What is not shown is that the training entity 22 can also be the patient P himself.

[0288] Figure 8B Shown is equipped with Figure 7 A perspective view of a patient P of the control system 10 disclosed in FIG. 1 , which control system 10 comprises two sensors 12 according to the present application.

[0289] In this embodiment, the patient P is equipped with Figure 7 The control system 10 disclosed in , the control system 10 includes two sensors 12, here two IMUs 12a attached to the shoes S of the patient P.

[0290] In particular, one IMU 12a is attached to the left shoe S of the patient P, and one IMU 12a is attached to the right shoe S of the patient P.

[0291] In this embodiment, the IMU 12a is placed on the heel area of ​​the shoe S of the patient P.

[0292] In this embodiment, the control system 10 also includes two electrodes 24a for FES.

[0293] Specifically, one electrode 24 a for FES is attached to the left leg of the patient P, and one electrode 24 a for FES is attached to the right leg of the patient P.

[0294] However, each leg of the patient P may generally be equipped with two or more electrodes 24a for FES.

[0295] Specifically, one electrode 24 a for FES is attached to the left thigh of the patient P, and one electrode 24 a for FES is attached to the right thigh of the patient P.

[0296] However, in general, one or more electrodes 24a for FES may be placed anywhere on the legs and / or buttocks and / or torso of the patient P.

[0297] Furthermore, in this embodiment, the control system 10 includes an electrode 24b for EES.

[0298] The electrodes 24b for EES are attached to the dorsal roots of the patient P.

[0299] However, it is also generally possible to position two or more electrodes 24b for EES in the epidural space or in the dorsal roots on top of the spinal cord.

[0300] according to Figure 7 , with the help of two IMUs 12a attached to each shoe S of the patient P, the control system 10 senses and uses each movement of the left and right foot of the patient P.

[0301] The controller 14 tracks and / or estimates the motion of the patient P's foot to convert it into stimulation data based on the estimated motion, which is provided to the patient P by the IPG 18 .

[0302] The IPG 18 provides one or more electrodes 24 a for FES via leads 20 and an electrode module 24 .

[0303] The IPG 18 provides one or more electrodes 24 b for EES via leads 20 and an electrode module 24 .

[0304] In alternative embodiments, the IMU 12a may be placed and / or inserted into, and / or integrated at different locations within the shoe S or the sole and / or insole.

[0305] In an alternative embodiment, the control system 10 may include only one IMU 12a, which is directly or indirectly positioned on the left or right foot of the patient P, or the left or right shoe S.

[0306] Alternatively, equipped with Figure 7 (See also Figure 8C ) can be equipped with two or more sensors 12 for each foot and / or leg.

[0307] In particular, equipped with Figure 7 A patient with the control system 10 disclosed in can be equipped with more sensors 12 located in a chain, for example from the hip to the foot via the thigh, knee and calf.

[0308] Typically, two or more sensors 12 may be located on a chain from head to toe.

[0309] In particular, the shoe S and / or the sole and / or the insole may be equipped with two or more sensors 12 .

[0310] The sensor 12 may be positioned anywhere from the distal to the proximal end of the foot, in particular in the heel region and / or the plantar region and / or the toe region and / or on the side of the foot.

[0311] In alternative embodiments, one or more sensors 12 may be part of, and / or inserted into and / or integrated into, the patient's exoskeleton, corset, belt, band, stretch strap, knee socks, athletic socks, and / or shoes S.

[0312] However, socks and tights can generally also consist of or include piezoelectric textile sensors integrated in the torso, waist, hips, knees, heel and / or toe areas.

[0313] An electrical response based on mechanical stretching, squeezing or pulling can be delivered.

[0314] In particular, socks or tights can be provided with electrodes and / or conductive yarns.

[0315] Alternatively, the magnetic sensor and magnetic field sensor may be incorporated into a shoe S for walking on a magnetic sensor plate, or inserted into a treadmill or gait phase detection device.

[0316] The magnetic force can be detected and acquired by a magnetic sensor under training (e.g. gait training).

[0317] exist Figure 8B Not shown is that in order to evaluate upper body movements and / or arm movements and / or hand movements, one or more sensors 12 can be inserted into and / or integrated into and / or integrated onto clothing, etc. of the upper body, torso and / or arms, and / or hands, including but not limited to tops, long-sleeved shirts, pullovers, jackets, one or more arm cuffs, gloves and / or one or more armbands.

[0318] exist Figure 8B What is not shown is that the electrodes 24a for FES may also be configured and arranged for stimulation of foot cramps and / or leg cramps to relieve and / or detect foot cramps and / or leg cramps.

[0319] exist Figure 8B What is not shown is that stimulating movement of one or more limbs and / or a part of one or more limbs does not necessarily require stimulation of the motor system of one or more limbs and / or a part of one or more limbs separately and directly.

[0320] As just one example, the spinal cord or thigh may be stimulated to induce reflexes and / or movements in the foot.

[0321] Figure 8C The device is shown as being equipped with seven sensors 12. Figure 7 A stereoscopic view of a patient using the control system 10 disclosed in FIG.

[0322] In this embodiment, the patient P is equipped with Figure 7 The control system 10 disclosed in , includes seven sensors 12, which are seven IMUs 12a.

[0323] The seven IMUs 12a establish a sensor network 12c.

[0324] In this embodiment, seven IMUs 12a are attached to the lower body of the patient P.

[0325] Specifically, one IMU 12a is placed in the center of the hip area, while the left leg is equipped with three IMUs 12a placed on the foot, calf and thigh, respectively, and the right leg is equipped with three IMUs 12a placed on the foot, calf and thigh, respectively.

[0326] However, alternative placements of the IMU 12a along the legs and / or feet and / or lower torso are generally possible.

[0327] In general, alternative arrangements of the IMU 12a or other sensor 12 types along the body and / or a portion of the body, such as the head, neck, torso, and / or one or both arms and / or one or both hands, are generally possible.

[0328] according to Figure 7 , with the help of seven IMUs 12a placed on the lower body of the patient P, the control system 10 senses and uses every movement of the patient's P legs and feet.

[0329] according to Figure 8B , both FES and EES may be provided by IPG 18, leads 20, and electrode module 24 having respective electrodes 24a and 24b.

[0330] Fig.8D A perspective view of a sensor insole according to the present application is shown.

[0331] In this embodiment, according to Figure 7 In the control system 10 disclosed in FIG. 1 , various sensors 12 are integrated into the sensor insole 300 of the shoe S of the patient P.

[0332] In this embodiment, the sensor 12 is a pressure sensor 12b.

[0333] In particular, eight pressure sensors 12 b are incorporated into the sensor insole 300 of the shoe S of the patient P.

[0334] Specifically, the eight pressure sensors 12 b are distributed from the distal end di of the sensor insole 300 to the proximal end pr of the sensor insole 300 of the shoe S of the patient.

[0335] Specifically, the eight pressure sensors 12 b are distributed along the heel region, metatarsal region, and toe region of the sensor insole 300 .

[0336] Specifically, two pressure sensors 12 b are placed in the heel region, two pressure sensors 12 b are placed in the toe region, and four pressure sensors 12 b are placed in the metatarsal region of the sensor insole 300 .

[0337] Typically, both shoes S of the patient P may be equipped with sensor insoles 300 .

[0338] The sensor insole 300 provides an accurate map of foot forces.

[0339] In particular, the pressure sensor 12b in the sensor insole 300 provides an accurate description of the gait phase and cadence. For example, by analyzing the sensor data obtained from a sensor insole 300 of the shoe S, the pre-swing, swing, weight-bearing reaction and / or stance (or another swing, stance, toe-off, mid-swing, heel strike, flat foot, mid-stance and / or heel-off) of a foot can be identified.

[0340] By using a second sensor insole 300 , the same events and parameters may be identified for the other foot of the patient P.

[0341] By combining the signals of the sensor insoles 300 of both feet of the patient P with the gait phase and cadence of the stimulus input, a reliable gait phase and cadence estimate can be provided.

[0342] according to Figure 7 As disclosed, the sensor stream is transmitted to the controller 14.

[0343] In one embodiment, alternative arrangements of the eight pressure sensors 12b in the sensor insole 300 are possible.

[0344] However, it is also possible to integrate 1-7 or more than 8 pressure sensors 12b into the sensor insole 300 of the shoe S of the patient P.

[0345] The sensor insole 300 itself may also be the pressure sensor 12b.

[0346] Fig. 8E Shown is equipped with Figure 7 A stereoscopic view of a patient with the control system disclosed in , which includes an IMU and a pressure insole for each foot of the patient according to the present application.

[0347] In this embodiment, if Figure 8B As shown, patient P is equipped with Figure 7 The control system 10 disclosed in the specification includes an IMU 12a placed on the left shoe S of the patient P and an IMU 12a placed on the right shoe S of the patient P, and Fig.8D A sensor insole 300 for a left shoe S of a patient P is shown, and Fig.8D A sensor insole 300 for a right shoe of a patient P is shown.

[0348] Thus, the sensor insoles 300 of the two pairs of shoes for the patient P include eight pressure sensors 12b (only exemplary in Fig.8D ).

[0349] Alternatively, patient P may be equipped with Figure 7The control system 10 described in , includes an IMU 12a and a separate sensor insole 300 for the left foot or the right foot.

[0350] In another embodiment, the IMU 12a and / or the sensor insole 300 may be replaced by another type of sensor 12, including but not limited to, for example, a piezoelectric component.

[0351] In this embodiment, the piezoelectric component may be integrated into, for example, a wearable device such as a sock, knee sock, tights, or shoe.

[0352] Fig. 9 A schematic diagram of a patient P and a trainer T (rehabilitation specialist) according to the present application is shown.

[0353] In this embodiment, both the patient P and the trainer T are equipped with Figure 7 A control system 10 is disclosed in.

[0354] In this embodiment, the control system 10 of the patient P and the control system 10 of the trainer T are interconnected.

[0355] The connection between the control system 10 of the patient P and the control system 10 of the trainer T is established via a wireless link WL.

[0356] However, cable bond connections are often also possible.

[0357] By means of the wireless link WL between the control system 10 of the patient P and the control system 10 of the trainer T, the reference data from the control system 10 of the trainer T are copied to the control system 10 of the patient P.

[0358] In particular, by means of a wireless link WL between the control system 10 of the patient P and the control system 10 of the trainer T, the stimulation timing of the patient P is synchronized with the movement(s) of the trainer T.

[0359] However, other reference data (including but not limited to step height or step length) may also generally be transmitted from the trainer T's control system 10 to the patient P's control system 10 .

[0360] In particular, the control system 10 of the trainer T serves as a reference for open-loop stimulation of the patient P via the control system 10 of the patient P.

[0361] Alternatively, the trainer T's control system 10 may be used as a reference for closed-loop stimulation of the patient P via the patient P's control system 10 .

[0362] Note that different gait events (toe-off, mid-swing, heel strike, flat foot, mid-stance and / or heel-off, or pre-swing, swing, weight-bearing reaction and / or stance) can be synchronized between trainer T and patient P.

[0363] However, for other movements besides gait, such as cycling, swimming, rowing, standing up, sitting down, different movement events can also be synchronized between the trainer T and the patient P.

[0364] However, in alternative embodiments it may be possible to transfer data offline and with a time delay, generally.

[0365] Note that the synchronization may enable identification and / or assessment and / or correction of the difference(s) between the healthy, regular and physiological movements of the trainer T and the impaired and irregular movements of the patient P.

[0366] Furthermore, it is generally possible to synchronize the control system 10 of a patient P who is more advanced in the rehabilitation process with the control system 10 of a patient P who is less advanced in the rehabilitation process.

[0367] However, a partial or complete tracking and estimation of the control algorithm and / or the motion model from the control system 10 of the patient P to the control system 10 of the trainer T is also generally possible.

[0368] Furthermore, synchronization and / or partial or full tracking and estimation of control algorithms and / or motion models from a control system 10 of a patient P who is more advanced in the rehabilitation process to a control system 10 of a patient P who is less advanced in the rehabilitation process is also generally possible.

[0369] Tracking and estimating may be performed online and / or in real time and / or with a time delay.

[0370] However, in alternative embodiments, offline tracking and estimation is also generally possible.

[0371] exist Fig. 9 Not shown in FIG. 1 is the fact that the patient P, the trainer T or one or more other patients can be equipped with training equipment, such as Fig. 8A The exoskeleton 22a shown is a robot, a treadmill, a bicycle and / or a body weight support system, etc.

[0372] Fig.10 A schematic diagram of a patient P and a remote trainer T1 according to the present application is shown.

[0373] In this embodiment, the patient P is equipped with Figure 7 The control system 10 disclosed in and located at position L.

[0374] In this embodiment, the remote trainer T1 is equipped with Figure 7The control system 10 disclosed in , and is located at a position L1 away from a position L of the patient P.

[0375] The control system 10 of the patient P and the control system 10 of the remote trainer T1 are connected to each other.

[0376] The connection between the control system 10 of the remote trainer T1 and the control system 10 of the patient P is established via a wireless link WL.

[0377] By means of the wireless link WL between the control system 10 of the patient P and the control system 10 of the remote trainer T1 , the control algorithm from the controller 14 of the control system 10 of the remote trainer T1 can be copied to the controller 14 of the patient P's control system 10 .

[0378] However, it is also possible to copy the control algorithm from the controller 14 of the control system 10 of one remote trainer T1 to the controllers 14 of the control systems 10 of two or more patients P located at different locations.

[0379] Two or more patients may also progress differently in the recovery process.

[0380] In this embodiment, tracking and estimation are performed online and in real time.

[0381] However, tracking and estimation can also be done offline and with a time delay.

[0382] exist Fig.10 What is not shown in FIG. 1 is that it is generally possible for one or more patients P to watch the remote trainer T1 via, for example, a standard screen, a projector, a computer, a laptop, a tablet or a mobile phone, and vice versa.

[0383] The patient P and the remote trainer T1 may also communicate with each other via common telecommunication equipment.

[0384] Fig.10 Not shown in FIG. 1 , the controller may generally allow tracking and estimation to be performed from one patient to another.

[0385] exist Fig.10 What is not shown is that it is generally possible that one or more patients P and a remote trainer T1 can communicate with each other in a manner similar to that shown in FIG. Figure 7 Embodiments described herein relate to communications within augmented reality and / or virtual reality modules.

[0386] Here too, gamification is possible.

[0387] Fig.11 A flow chart showing an offline workflow of a control system for a patient's motion reconstruction and / or restoration system according to the present application.

[0388] It particularly applies to all described embodiments of the present application described in this disclosure.

[0389] like Figure 7 and Fig. 8A The illustrated off-line workflow 100 of the control system 10 for the gait reconstruction and / or restoration system for the patient P comprises steps S101 - S106 .

[0390] In a starting step S101 , one or more sensors 12 or one or more sensor networks measure sensor data over a period of at least one complete movement cycle.

[0391] In one embodiment, the motion cycle may be a gait cycle.

[0392] In alternative embodiments, the exercise cycle may be a swimming or rowing cycle, or standing or sitting.

[0393] In alternative embodiments, the motion cycle may be any other motion.

[0394] In step S102 , the sensor data is transmitted to the controller 14 .

[0395] In step S103 , the sensor data accumulated during the recorded period of time is stored in a sensor data buffer of the controller 14 .

[0396] Thereafter, in step S104 , the controller 14 determines a list of different motion events and phases offline based on the sensor data accumulated over the time period recorded in the sensor data buffer.

[0397] For gait, possible gait events may include, but are not limited to, initial ground contact, heel strike, foot flat, weight-bearing reaction, mid-stance, final stance, heel-off, pre-swing, toe-off, initial swing, mid-swing, final swing and / or heel strike (or, for example, pre-swing, swing, weight-bearing reaction, stance).

[0398] However, it is possible that there are only two gait events, foot strike and foot lift-off.

[0399] After the list of different motion events and phases has been determined offline in step 104 , a motion model for the recorded motion events is created in step S105 .

[0400] Thereafter, in step S106, the patient is stimulated.

[0401] The created motion models can be used offline at any time.

[0402] Controller 14 programs IPG 18 to deliver the correct stimulation via leads 20 according to the movement model determined offline by controller 14 .

[0403] According to the motion model determined offline, the motion phase always takes the same value in the same event.

[0404] Execution tracking and offline estimation may allow the use of conditions that are not possible in real time.

[0405] Note that the sensor data buffer of the controller 14 may include accumulated sensor data from one patient P and / or from two or more patients P.

[0406] However, the sensor buffer may also include accumulated sensor data from one or more trainers T and / or one or more healthy subjects.

[0407] exist Fig.11 Not shown in FIG. 1 , the recorded sensor data may be pre-processed using a Kalman filter.

[0408] Fig.12 A flow chart showing an online workflow of a control system for a patient's motion reconstruction and / or restoration system according to the present application.

[0409] It particularly applies to all described embodiments of the present application described in this disclosure.

[0410] like Figure 7 and Fig. 8A The illustrated online workflow 200 of the control system 10 for the motion reconstruction and / or restoration system for the patient P comprises steps S201 - S206 .

[0411] In a starting step S201 , one or more sensors 12 or one or more sensor networks measure sensor data.

[0412] In step S202 , the sensor data is transmitted from the sensor 12 to the sensor data buffer of the controller 14 .

[0413] In step S203 , the sensor data of the recorded time period is stored online in the sensor data buffer of the controller 14 .

[0414] In other words, the sensor data buffer is always updated online with the latest sensor data from the sensor(s) 12 .

[0415] In step S204 , based on the conventional motion model and the sensor data accumulated in the sensor data buffer of the controller 14 , the controller 14 determines a list of different motion events and phases for all recorded motion events.

[0416] For gait, possible motion events, gait events may include, but are not limited to, initial ground contact, heel strike, foot flat, weight bearing reaction, mid-stance, final stance, heel-off, early swing, toe-off, initial swing, mid-swing, final swing and / or heel strike, respectively.

[0417] However, it is possible that there are only two gait events, foot strike and foot lift-off.

[0418] Various sensor data inputs from the sensor(s) 12 update the sensor data buffer, and once an entire motion (eg, a gait cycle) is detected, past motion events (eg, gait events) are determined online.

[0419] In step 205 , the controller 14 trains the motion model to adapt to the specific motion of the patient P using the most recently accumulated sensor data.

[0420] In step S206 , the patient P is stimulated according to the motion model.

[0421] Controller 14 programs IPG 18 to deliver the correct stimulation via leads 20 based on the most recent motion pattern determined online by controller 14 .

[0422] The online workflow 200 enables real-time solutions and real-time data transfer.

[0423] Note that the general motion model used for fusion with the most recent sensor data may be based on accumulated sensor data from one patient P and / or from two or more patients P.

[0424] However, the general motion model used for fusion with the nearest sensor may also be based on sensor data from one or more trainers T and / or one or more healthy subjects.

[0425] When the motion model is sufficiently refined, the online learning process can be stopped and it can be stored for further sessions with the same patient P.

[0426] exist Fig.12 Not shown in FIG. 1 , filters may be used to pre-process the recorded sensor data.

[0427] exist Fig.12 Not shown is that the filter may be a Kalman filter.

[0428] However, other types of filters may also be used to pre-process the recorded sensor data in general.

[0429] exist Fig.12Not shown, the online workflow 200 may also be used to control the system 10 for a motion reconstruction and / or restoration system for a variety of motions including, for example, walking, running, cycling, swimming, rowing, standing, sitting.

[0430] Fig.13 Shown is equipped with Figure 7 Schematic diagram of the foot inclination / forward acceleration of a patient P of the control system disclosed in FIG.

[0431] In this case, each foot of the patient P is equipped with an IMU 12 a .

[0432] Alternatively, patient P may be equipped with Figure 7 The control system 10 described in the specification includes an IMU 12a and a sensor insole 300 for the left foot or the right foot respectively.

[0433] In another embodiment, the patient P may be equipped with two or more IMUs 12a per foot.

[0434] Furthermore, the IMU 12a and / or sensor insole 300 may be replaced by another type of sensor 12, including but not limited to, for example, a piezoelectric assembly.

[0435] In this embodiment, the piezoelectric component can be integrated into a wearable device such as a sock, knee sock, tights, or shoe.

[0436] Shows the equipment Figure 7 The foot inclination (degrees) and forward acceleration (meters per s2) of the right foot of a patient P during walking according to the control system 10 disclosed in FIG.

[0437] From these signals, cadence, pre-swing, swing, load-bearing reaction, and stance can be clearly identified.

[0438] The same events and parameters can be identified for the left foot.

[0439] Since walking is a periodic motion, all measured signals are also periodic.

[0440] By combining the gait phase and cadence information of both feet of the patient P with the gait phase and cadence of the stimulation input, a reliable gait phase and cadence estimate can be provided.

[0441] Note that the gait of different patients P as well as a single patient P at different walking speeds and with different assistive devices (body weight support, walker, cane, etc.) can vary greatly.

[0442] Especially for gait disorders, not all gait events are present.

[0443] Therefore, it is always possible to estimate the cadence by extracting the fundamental frequency of the measured signal.

[0444] Furthermore, machine learning methods can be used to adapt the gait phase estimation to the specific gait of patient P.

[0445] The degree of consistency and difference between the movements of the left and right feet and the stimulation input can be used to indicate the reliability of the gait phase estimation, for example the measured left foot cadence should be equal to the measured right foot cadence and the cadence of the provided stimulation, and the left and right feet should be (roughly) in anti-phase.

[0446] It can also be achieved in a way that the feet do not move independently of each other in the control loop, but are connected mechanically via the hip joint and on a neural level via the spinal cord.

[0447] In particular, inhibitory reflex circuits in the spinal cord may modulate neural firing rate (and hence the recruitment of motor neurons via the EES).

[0448] Note that the example control and estimation routines included herein can be used with various system configurations. The control methods and routines disclosed herein can be stored as executable instructions in a non-transitory memory and can be executed by the control system 10, for example as part of a controller 14 in combination with a controller 12, a programmer 16, a stimulation system 18, a lead 20 and other system hardware. The specific routines described herein can represent one or more of any number of processing strategies, such as event-driven, interrupt-driven, multi-tasking, multi-threading, etc. In this way, the various actions, operations and / or functions shown can be performed in the order shown, in parallel, or omitted in some cases. Likewise, the processing order is not necessary to achieve the features and advantages of the example embodiments described herein, but is provided for ease of illustration and description. Depending on the specific strategy used, one or more of the illustrated actions, operations and / or functions can be repeatedly performed. In addition, the described actions, operations and / or functions can graphically represent the coding in the non-transitory memory of a computer-readable storage medium to be programmed into the controller 14, wherein the described actions are performed by executing instructions in the control system 10 including various hardware components.

Claims

1. A control system (10) for a patient (P) movement reconstruction and / or recovery system, characterized in that include: - at least one sensor (12), - at least one controller (14), - at least one programmer (16), - at least one stimulation system (18), wherein the controller (14) is connected to the sensor (12), the programmer (16) and the stimulation system (18), wherein the sensor (12) is part of a training entity (22) or is connected to the training entity (22) so that - Create and / or guide a movement model for the patient (P), and / or - adjusting stimulation settings based on input from the sensor (12), and wherein the controller (14) is configured and arranged to track and estimate the movement of the training entity (22) and to convert it into stimulation data based on the estimated movement, the stimulation data being provided to the patient (P) via the stimulation system (18) to train the patient in movement reconstruction and / or restoration, and wherein the controller (14) is also configured and arranged to program the stimulation system (18) based on sensor data measured during at least one complete movement cycle of the patient; The controller (14) is configured and arranged so that the tracking and estimating are performed offline; The control system (10) comprises an early warning module, which is configured and arranged to provide an early warning signal indicating an impending stimulation event.

2. The control system (10) according to claim 1, characterized in that: The training entity (22) is a trainer (T) and / or a physiotherapist.

3. The control system (10) according to claim 1 or claim 2, characterized in that: The training entity (22) is a training device or comprises a training device, wherein the device is an exoskeleton (22a), a robot, a treadmill, a bicycle and / or a body weight support system.

4. The control system (10) according to claim 1, characterized in that: The controller (14) is configured and arranged such that the tracking and estimating are performed online and / or in real time and / or with a time delay.

5. The control system (10) according to claim 1, characterized in that: The controller (14) is configured and arranged to perform the tracking and estimating from one patient (P) to another patient (P).

6. The control system (10) according to claim 1, characterized in that There is a sensor network (12c) formed by at least two or more sensors (12), wherein at least one of the two or more sensors (12) is connected to the controller (14).

7. The control system (10) according to claim 1, characterized in that The control system (10) further comprises an augmented reality and / or virtual reality module, which is configured and arranged to provide information related to motion reconstruction and / or restoration.

8. The control system (10) according to claim 7, characterized in that The augmented reality and / or virtual reality module is configured and arranged to provide gamified information related to motion reconstruction and / or restoration.

9. The control system (10) according to claim 1, characterized in that: The at least one sensor (12) is or includes an inertial measurement unit (12a), an optical sensor, a camera, a piezoelectric component, a velocity sensor, an accelerometer, a magnetic field sensor, a torque sensor, a pressure sensor (12b), a displacement sensor, an EMG measurement unit, a goniometer, a magnetic position sensor, a Hall sensor, a gyroscope and / or a motion tracking camera or an infrared camera.

10. The control system (10) according to claim 1, characterized in that The training entity (22) is the patient (P) itself.

11. The control system (10) according to claim 7, characterized in that The information is information related to the training for movement reconstruction and / or rehabilitation to be performed or being performed.

Images