Systems and methods for applying vibration stimuli in wearable devices

Abstract

In one embodiment, a wearable device for vibration stimulation is presented. The wearable device includes a sensor configured to receive data and generate a sensor output. The wearable device includes a processor in communication with the sensor and a memory communicatively connected to the processor. The memory includes instructions configuring the processor to receive the sensor output from the sensor. The processor is configured to determine a symptom of a movement disorder of a user based on the sensor output. The processor is configured to calculate a waveform output based on the symptom of the movement disorder. The processor is configured to direct a transducer in communication with the processor to apply the waveform output to the user to alleviate the symptom of the movement disorder.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to and the benefit of U.S. Provisional Application No. 63 / 371,145, entitled "Systems and Methods for Applying Vibratory Stimulus in a Wearable Device," filed August 11, 2022, which is incorporated herein in its entirety.

[0002] FIELD OF THE DISCLOSURE The present disclosure relates to systems and methods for applying stimulation, particularly in wearable devices. [Background technology]

[0003] Currently, there are approximately 10 million people living with movement disorders worldwide. Many of the current therapies for movement disorders can be invasive and expensive. Current movement disorder treatments and / or therapies can be improved. Summary of the Invention

[0004] This Summary is provided to introduce a selection of concepts in a simplified form that are further described in the Detailed Description. This Summary is not intended to identify key features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0005] In one embodiment, a wearable device for vibration stimulation is presented. The wearable device includes a sensor configured to receive data and generate a sensor output. The wearable device includes a processor in communication with the sensor and a memory communicatively connected to the processor. The memory includes instructions configuring the processor to receive the sensor output from the sensor. The processor is configured to determine a symptom of a movement disorder of a user based on the sensor output. The processor is configured to calculate a waveform output based on the symptom of the movement disorder. The processor is configured to direct a transducer in communication with the processor to apply the waveform output to the user to alleviate the symptom of the movement disorder.

[0006] In another embodiment, a method of providing vibration stimulation through a wearable device is presented. The method includes receiving user data through a sensor of the wearable device. The method includes generating a sensor output based on the user data through the sensor. The method includes communicating the sensor output to a processor of the wearable device. The method includes determining, by the processor, a symptom of the movement disorder based on the sensor output. The method includes calculating, by the processor, a waveform output based on the symptom of the movement disorder. The method includes instructing a transducer in communication with the processor to apply the waveform output to the user.

[0007] The foregoing aspects and many of the attendant advantages of the presently disclosed embodiments will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein: [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 illustrates a system for alleviating movement disorders. [Figure 2] FIG. 1 illustrates the flexor muscles and tendons of the wrist, fingers, and thumb. [Figure 3] FIG. 1 illustrates the extensor muscles and tendons of the wrist, fingers, and thumb. [Figure 4]FIG. 1 shows targeted somatosensory afferents, which are a subset of cutaneous mechanoreceptors. [Figure 5] FIG. 1 illustrates the location of upper extremity dermatomes innervated by the C5, C6, C7, C8, and T1 spinal nerves. [Figure 6] FIG. 10 illustrates a waveform parameter selection process. [Figure 7] FIG. 10 illustrates a user input process for waveform parameter selection. [Figure 8] 1 is a flow diagram of a method for alleviating movement disorders. [Figure 9] FIG. 1 is a diagram of a wearable device. [Figure 10] FIG. 2 is an exploded side view of the wearable device. [Figure 11] FIG. 1 illustrates a machine learning module. [Figure 12] FIG. 1 is a block diagram of a computing system that may implement any of the systems, processes, or methods described throughout this disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0009] The drawings are not necessarily to scale and may be illustrated by phantom lines, schematic representations, and fragmentary views. In some cases, details that are not necessary for an understanding of the embodiments or that make other details difficult to perceive may be omitted.

[0010] Aspects of the present disclosure can be used to achieve symptomatic relief of movement disorders through a wearable medical device. In one embodiment, the wearable medical device may provide vibration stimulation to a body part of a user. Another aspect of the present disclosure can be used to apply stimulation around a user's wrist through a wristband, which may allow stimulation of five separate somatosensory channels through the C5-T1 dermatomes and an additional 15 proprioceptive channels through the tendons passing through the wrist. This may allow stimulation of a total of 20 separate channels in a wristband form factor that is much less cumbersome than electric gloves.

[0011] FIG. 1 illustrates a system 100 for alleviating movement disorders according to one embodiment of the present invention. The system 100 may include a wearable device 100. The wearable device 100 may include a processor, such as a processing unit 104, and memory communicatively connected to the processing unit 104. The memory of the wearable device 100 may include instructions that configure the processing unit 104 of the wearable device 100 to perform various tasks. The wearable device 100 may include a communications module 108. As used throughout this disclosure, a "communications module" refers to any form of software and / or hardware capable of transmitting electromagnetic energy. For example, the communications module 108 may be configured to transmit and receive radio signals, Wi-Fi signals, Bluetooth® signals, cellular signals, etc. The communications module 108 may include a transmitter, a receiver, and / or other components. The transmitter of the communications module 108 may include, but is not limited to, an antenna. The antenna of the communications module 108 may include, but is not limited to, a dipole, monopole, array, loop, and / or other antenna types. The receiver of the communications module 108 may include, but is not limited to, an antenna as previously described. The communications module 108 may communicate with the processing unit 104. For example, the processing unit 104 may be physically connected to the communications module 108 through one or more wires, circuits, etc. The processing unit 104 may instruct the communications module 108 to send and / or receive data transmissions to one or more other devices. For example, but not limited to, the communications module 108 may transmit vibration stimulation data, movement data of the user's body 150, electrical activity of the user's muscles 164, etc. In some embodiments, the communications module 108 may transmit therapy data. The therapy data may include, but is not limited to, symptom severity, symptom type, frequency of the vibration stimulation 13, data from the sensor suite 112, etc.The communication module 108 may communicate with one or more external computing devices, such as, but not limited to, a smartphone, a tablet, a laptop, a desktop, a server, a cloud computing device, etc. The wearable device 100 is as further described below with reference to FIG.

[0012] With continued reference to FIG. 1 , the wearable device 100 may include one or more sensors. As used throughout this disclosure, a “sensor” is an element capable of detecting a physical property. The physical property may include, but is not limited to, kinetics, electricity, magnetism, radiation, thermal energy, and the like. In some embodiments, the wearable device 100 may include a sensor suite 112. As used throughout this disclosure, a “sensor suite” is a combination of two or more sensors. The sensor suite 112 may include multiple sensors, such as, but not limited to, two or more sensors. The sensor suite 112 may include two or more sensors of the same type. In other embodiments, the sensor suite 112 may include two or more sensors of different types. For example, the sensor suite 112 may include an electromyography sensor (EMG) 116 and an inertial measurement unit (IMU) 120. The IMU 120 may be configured to detect and / or measure specific forces, angular velocity, and / or orientation of the body. Other sensors in the sensor suite 112 may include, but are not limited to, an accelerometer, a gyroscope, an impedance sensor, a temperature sensor, and / or other types of sensors. The sensor suite 112 may be in communication with the processing unit 104. The communication between the sensor suite 112 and the processing unit 104 may be an electrical connection over which data may be shared between the sensor suite 112 and the processing unit 104. In some embodiments, the sensor suite 112 may be wirelessly connected to the processing unit 104, such as through Wi-Fi, Bluetooth, or other connection.In some embodiments, one or more components of the wearable device 100 may be the same as those described in U.S. Patent Application No. 16 / 563,087, entitled "Apparatus and Method for Reduction of Neurological Movement Disorder Symptoms Using Wearable Device," filed September 6, 2019, which is incorporated herein by reference in its entirety.

[0013] One or more sensors in the sensor suite 112 may be configured to receive data from a user, such as the user's body 150. The data received by the one or more sensors in the sensor suite 112 may include, but is not limited to, motion data, electrical data, etc. The motion data may include, but is not limited to, acceleration, velocity, angular velocity, and / or other types of dynamics. In some embodiments, the IMU 120 may be configured to receive motion 15 from the user's body 150. The motion 15 may include, but is not limited to, vibration, acceleration, muscle contraction, and / or other aspects of motion. The motion 15 may be generated from one or more muscles 164 of the user's body 150. The muscles 164 may include, but are not limited to, wrist muscles, hand muscles, forearm muscles, etc. In one embodiment, the motion 15 generated from the muscles 164 of the user's body 150 may be involuntarily generated by one or more symptoms of a movement disorder of the user's body 150. Movement disorders may include, but are not limited to, Parkinson's disease (PD), post-stroke recovery, etc. Symptoms of impaired movement include, but are not limited to, stiffness, freezing of gait, tremor, shaking, involuntary muscle contractions, and / or other symptoms. In other embodiments, movements 15 generated from muscles 164 of user's body 150 may be voluntary. For example, a user may actively control one or more of their muscles 164, thereby generating movements 15 that can be detected and / or received by sensors in sensor suite 112.

[0014] With further reference to FIG. 1 , one or more sensors of the sensor suite 112 may be configured to receive electrical data, such as electrical activity 14, which may be generated by one or more of the muscles 164. The electrical data may include, but is not limited to, voltage, impedance, current, resistance, reactance, waveform, etc. For example, the electrical activity 14 may include an increase in current and / or voltage in one or more of the muscles 164 during contraction of one or more of the muscles 164. The EMG 116 of the sensor suite 112 may be configured to receive and / or detect the electrical activity 14 generated by the muscles 164. In some embodiments, one or more sensors of the wearable device 11 may be configured to generate a sensor output. As used in this disclosure, “sensor output” refers to information generated by one or more sensing devices. The sensor output may include, but is not limited to, voltage, current, acceleration, velocity, and / or other outputs. The sensor output generated from one or more sensors of the sensor suite 112 may be communicated to the processing unit 104 via a wired connection, a wireless connection, other connection, or the like. The processing unit 104 may be configured to determine symptoms of a movement disorder based on sensor output received from one or more sensors. The processing unit 104 may be configured to determine symptoms such as, but not limited to, rigidity, tremor, and freezing of gait. Freezing of gait is a symptom of Parkinson's disease in which a Parkinson's patient suddenly and temporarily loses the ability to take a step forward despite the patient's intention to walk. Abnormal walking patterns can range from merely inconvenient to potentially dangerous because they may increase the risk of falls. Rigidity may refer to involuntary contraction and stiffening of the muscles of a Parkinson's patient. The processing unit 104 may compare one or more values ​​of sensor output from the sensor suite 112 to one or more values ​​associated with one or more symptoms of a movement disorder.For example, processing unit 104 may compare the sensor output of one or more sensors in sensor suite 112 to one or more stored values ​​that may already be associated with one or more symptoms of a movement disorder. As a non-limiting example, an acceleration of a user's arm of about 1 inch / second to about 3 inches / second may correspond to a symptom of a moderate tremor.

[0015] In some embodiments, the processing unit 104 may utilize a classifier or other machine learning model capable of categorizing the sensor output into movement disorder symptom categories. A “classifier,” as used in this disclosure, is a machine learning model, such as a mathematical model, neural net, or program, generated by a machine learning algorithm known as a “classification algorithm,” as described in more detail below, that sorts inputs into categories or bins of data and outputs the categories or bins of data and / or their associated labels. The classifier may be configured to output a datum that labels or otherwise identifies clustered data sets that are found to be close according to a distance metric, as described below. The processor 104 and / or another device may use a classification algorithm to generate a classifier, which is defined as the process by which a processor derives a classifier from training data. Classification may be performed using linear classifiers such as, but not limited to, logistic regression and / or naive Bayes classifiers, nearest neighbor classifiers such as k-nearest neighbor classifiers, support vector machines, least squares support vector machines, Fisher's linear discriminant, quadratic classifiers, decision trees, boosted trees, random forest classifiers, kernel estimation, learning vector quantization, and / or neural network-based classifiers.

[0016] Continuing with reference to Figure 1, classifiers may be generated using, as a non-limiting example, a naive Bayes classification algorithm. A naive Bayes classification algorithm generates classifiers by assigning class labels to problem instances, which are represented as vectors of element values. The class labels are drawn from a finite set. A naive Bayes classification algorithm may include generating a family of algorithms that assume that, given a class variable, the value of a particular element is independent of the value of any other element. A naive Bayes classification algorithm may be based on Bayes' theorem, expressed as P(A / B) = P(B / A) P(A) ÷ P(B), where P(AB) is the probability of hypothesis A given data B, also known as the posterior probability; P(B / A) is the probability of data B given hypothesis A being true; P(A) is the probability that hypothesis A is true regardless of the data, also known as the prior probability of A; and P(B) is the probability of the data regardless of the hypothesis. A Naive Bayes algorithm may be generated by first converting the training data into a frequency table.

[0017] The processor 104 may calculate a likelihood table by calculating the probabilities of different data entries and classification labels. The processor 104 may utilize a naive Bayes equation to calculate the posterior probability of each class. The class with the highest posterior probability is the predicted result. The naive Bayes classification algorithm may include a Gaussian model that follows a normal distribution. The naive Bayes classification algorithm may include a polynomial model that is used for discrete counts. The naive Bayes classification algorithm may include a Bernoulli model that may be utilized when the vector is binary.

[0018] With continued reference to FIG. 1 , the classifier may be generated using a K-nearest neighbor (KNN) algorithm. As used in this disclosure, a "K-nearest neighbor algorithm" includes a classification method that utilizes feature similarity to analyze how similar out-of-sample features are to training data and classify input data into one or more clusters and / or categories of features represented in the training data. This may be performed by representing both the training data and the input data in vector form, using one or more measures of vector similarity to identify classifications in the training data, and determining the classification of the input data. The K-nearest neighbor algorithm may include specifying a K value, or a numerical value that instructs the classifier to select the k training data most similar to a given sample, determining the most common classifier of entries in a database, and classifying the known sample. This may be performed recursively and / or iteratively to generate a classifier that can be used to classify the input data as further samples. For example, an initial set of samples may be run to target initial heuristics and / or "first guesses" at outputs and / or relationships, which may be seeded using, without limitation, expert input received according to any process described herein. As a non-limiting example, the initial heuristics may include ranking associations between inputs and elements of the training data. The heuristics may include selecting some of the highest-ranked associations and / or elements of the training data.

[0019] The classifier may be trained using training data that correlates the movement data and / or electrical data to symptoms of a movement disorder. The training data may be received through user input, an external computing device, and / or previous training iterations. As a non-limiting example, the IMU 120 may receive the movements 15 produced by the muscles 164 and generate sensor output including acceleration values, which may be communicated to the processing unit 104. The processing unit 104 may classify and / or categorize the sensor output as a symptom of freezing of gait.

[0020] With further reference to FIG. 1 , the processing unit 104 may train a classifier using training data that correlates the movement data and / or electrical data with symptoms of a movement disorder. In other embodiments, training of the classifier and / or other machine learning model may occur remotely from the processor 104, and one or more trained models, such as classifiers, machine learning models, weights, etc., may be sent to the processor 104. The training data may be received by user input, through one or more external computing devices, and / or through previous processing iterations. The classifier may input sensor output, such as the output of the sensor suite 112, and be configured to categorize the output into one or more groups, such as, but not limited to, tremor, rigidity, freezing of gait, etc. The processing unit 104 may calculate a waveform output based on the sensor output generated by one or more sensors of the wearable device 100. As used in this disclosure, a "waveform output" is a signal having a frequency. The waveform output may be generated as a vibratory, electrical, auditory, and / or other waveform. The waveform output may include one or more parameters, such as frequency, phase, amplitude, channel index, etc. The channel index may include the channel of the mechanoreceptor and / or the channel of the actuator to be used. For example, the channel index may include one or more channels of the mechanoreceptor, the actuator stimulating the mechanoreceptor, and / or a combination thereof. The processing unit 104 may select one or more parameters of the waveform output based on sensor output received from one or more sensors of the wearable device 100. In other embodiments, the waveform parameters may be selected by the user. As a non-limiting example, the user may use buttons on the wearable device 100 to select waveform parameters from a predefined waveform list. The predefined waveform list may include, but is not limited to, one or more waveforms having various frequencies, amplitudes, etc. The predefined waveform list may be generated through previous iterations of waveform generation. In other embodiments, the predefined waveform list may be input by one or more users.In some embodiments, the predefined waveform list may include waveforms for specific symptoms, such as, but not limited to, freezing of gait, tremor, rigidity, etc. In some embodiments, a user may select specific waveform parameters using an external computing device, such as, but not limited to, a smartphone, laptop, tablet, desktop, smartwatch, etc., that can communicate with the processing unit 104 through the communications module 108. Waveform output generation may be described in further detail below with reference to FIGS.

[0021] In some embodiments, the processing unit 104 may transmit the waveform output to one or more transducers of the wearable device 100. As used in this disclosure, a "transducer" is a device that converts energy from one form to another. For example, a transducer may include, but is not limited to, an electrical transducer, a mechanical transducer, a thermal transducer, an acoustic transducer, and / or other types of transducers. The wearable device 100 may include one or more transducers. For example, the wearable device 100 may include two or more transducers. In some embodiments, the wearable device 100 may include two or more transducers of different types, such as a mechanical transducer and an electrical transducer, or an electrical transducer and an acoustic transducer. The transducers of the wearable device 100 may be positioned to provide stimulation to specific parts of the user's body 150, such as through the waveform output. The wearable device 100 may include one or more mechanical transducers 124, which may be positioned to stimulate one or more mechanoreceptors 154 of the user's body 150. For example, the mechanical transducers 124 may be positioned along the wristband of the wearable device 100. The wearable device 100, in one embodiment, may include four mechanical transducers 124 that are equidistant from one another and may be located within a wristband of the wearable device 100. In other embodiments, the mechanical transducers 124 may be located on a surface of the housing of the wearable device 100, as described in further detail below with reference to FIG. 10 . The mechanical transducers 124 may include, but are not limited to, piezoelectric motors, electromagnetic motors, linear resonant actuators (LRAs), eccentric rotating mass motors (ERMs), and the like. The mechanical transducers 124, in one embodiment, may be configured to vibrate at or above 200 kHz. The mechanical transducers 124 may draw energy from one or more batteries from the wearable device 100. For example, the mechanical transducers 124 may draw approximately 5 W of power from the battery of the wearable device 100.In some embodiments, the mechanical transducer 124 may have, but is not limited to, a maximum current draw of approximately 90 mA, a current draw of approximately 68 mA, a current draw of 34 mA at a 50% duty cycle, and a voltage of approximately 0 V to approximately 5 V. As used throughout this disclosure, "mechanoreceptor" refers to a cell in the human body that responds to mechanical stimuli. The mechanoreceptor 154 may include proprioceptors 158 and / or somatosensory organs 160. The proprioceptors 158 may include the sem muscles of the head innervated by the trigeminal nerve. The proprioceptors 158 may be part of one or more regions of the user's extremities, such as, but not limited to, the wrist, hand, leg, foot, or arm. The somatosensory organs 160 may include cells with receptor neurons located in the dorsal root ganglion. The mechanoreceptors 154 may be described in further detail below with reference to FIG. 4.

[0022] Continuing with reference to FIG. 1 , the processing unit 104 may be configured to instruct the mechanical transducer 124 to apply a vibration stimulus 13 to one or more mechanoreceptors 154 of the user's body 150. The vibration stimulus 13 may include a waveform output that is calculated by the processing unit 104 and applied to the user's body 150 through the mechanical transducer 124. It should be noted that while a mechanical transducer 124 is illustrated in FIG. 1 , other transducers, such as those described above, may also be used without limitation. The vibration stimulus 13 may be applied to the mechanoreceptors 154 through the mechanical transducer 124, thereby causing the mechanoreceptors 154 to generate one or more afferent signals 154. As used in this disclosure, an "afferent signal" is a neuronal signal in the form of an action potential that travels toward a target neuron. The afferent signal 154 may be transmitted to a peripheral nervous system (PNS) 172 of the user's body 150. As used in this disclosure, the term "peripheral nervous system" refers to the division of the nervous system that includes all nerves outside the central nervous system. The central nervous system (CNS) 1204 may include the spinal cord 184 and / or the brain 188 of the user's body 150. The brain 188 may transmit efferent signals 176 to the PNS 172 through the spinal cord 184. As used in this disclosure, an "efferent signal" is a signal that carries motive information for a muscle to perform an action. The efferent signal 176 may include one or more electrical signals that can cause a muscle 164 to contract or otherwise move. For example, the PNS 172 may input an afferent signal 168 and transmit the afferent signal 168 through the spinal cord 184 to the brain 188. The brain 188 may generate one or more efferent signals 176 and transmit the efferent signals to the PNS 172 through the spinal cord 184. The PNS 172 may transmit an efferent signal 176 to the muscle 164 .

[0023] The processing unit 104 may operate within a closed-loop system. For example, the processing unit 104 may operate within a feedback loop between data generated from the muscles 164 and the vibration stimuli 13 generated by the mechanical transducer 124. Furthermore, the closed-loop system may extend through and / or to the PNS 172, CNS 180, brain 188, etc. of the user's body 150 based on the afferent signals 168 and the efferent signals 176. In some embodiments, the processing unit 104 may be configured to operate in one or more modes. For example, the processing unit 104 may operate in a first mode and a second mode. The first mode may include passively monitoring the movement of the user's body 150 to detect symptoms of a movement disorder that exceed a threshold. The threshold may include a root-mean-square acceleration of 100 mG or 500 mG. The threshold may be set by the user and / or determined by the processing unit 104 based on historical data. The historical data may include, but is not limited to, the user's sensor data and / or waveform output data over a period of time, such as minutes, hours, weeks, months, or years. The threshold may include, but is not limited to, one or more acceleration, pressure, current, and / or voltage values. In some embodiments, when the threshold is reached, the processing unit 104 may be configured to operate in a second mode instructing the mechanical transducer 124 to provide vibration stimuli 13 to the mechanoreceptors 154.

[0024] Figure 2 shows the flexor muscles and tendons of the wrist, fingers, and thumb. The flexor muscles of the wrist are selected from the group consisting of the flexor carpi radialis (FCR) 21, flexor carpi ulnaris (FCU) 22, and palmaris longus (PL) 23. The flexor muscles of the fingers are selected from the group consisting of the flexor digitorum profundus (FDP) 24 and flexor digitorum superficialis (FDS) 25. The flexor muscles of the thumb are selected from the group consisting of the flexor pollicis longus (FPL) 26, flexor pollicis brevis (FPB) 27, and abductor pollicis brevis (APB) 28.

[0025] Figure 3 shows the extensor muscles and tendons of the wrist, fingers, and thumb. The wrist extensor muscles are selected from the group consisting of the extensor carpi radialis brevis (ECRB) 31, the extensor carpi radialis longus (ECRL) 32, and the extensor carpi ulnaris (ECU) 33. The finger extensor muscles are selected from the group consisting of the extensor digitorum communis (EDC) 34, the extensor digitorum minimi (EDM) or extensor digitorum quinti proprius (EDQP) 35, and the extensor digitorum indicis proprius (EIP) 36. The extensor muscles of the thumb are selected from the group consisting of the abductor pollicis longus (APL) 37, the extensor pollicis longus (EPL) 38, and the extensor pollicis brevis (EPB) 39.

[0026] Figure 4 illustrates various somatosensory afferents that can be targeted. Somatosensory afferents may be a subset of cutaneous mechanoreceptors. The set of cutaneous mechanoreceptors includes Pacinian corpuscles 41, Meissner corpuscles 42, Merkel complexes 43, Ruffini corpuscles 44, and C-fiber low-threshold mechanoreceptors (C-LTMRs). Pacinian corpuscles (PCs) 41 are cutaneous mechanoreceptors that respond primarily to vibration stimuli in the frequency range of 20–1000 Hz. Meissner corpuscles 42 are most sensitive to low-frequency vibrations between 10 and 50 Hz and can respond to skin indentations of less than 10 micrometers. Merkel nerve endings 43 are the most sensitive of the four main types of mechanoreceptors to low-frequency vibrations between approximately 5 and 15 Hz. Ruffini corpuscles44 are found in the superficial dermis of both hairy and glabrous skin and register low-frequency vibrations or pressure below 40 Hz. C-LTMRs45 are present in 99% of hair follicles and transmit input signals from the periphery to the central nervous system. The present invention focuses on stimulation of cutaneous mechanoreceptors in upper extremity dermatomes innervated by the C5, C6, C7, C8, and T1 spinal nerves, labeled according to the corresponding spinal nerves, as illustrated in Figure 5.

[0027] Figure 5A shows the location of the upper extremity dermatomes innervated by the C5, C6, C7, C8, and T1 spinal nerves in a frontal view.

[0028] Figure 5B shows the location of the upper extremity dermatomes innervated by the C5, C6, C7, C8, and T1 spinal nerves in a dorsal view.

[0029] Referring to FIG. 6 , a waveform parameter selection system 600 is presented. The system 600 may be local to the wearable device 100, such as processed by the processor 104. In other embodiments, the system 600 may be run through an external computing device, such as, but not limited to, a smartphone, tablet, desktop, laptop, server, or cloud computing device. The processing unit 104 may be configured to process raw sensor input 604 received from the sensor suite 112 based on activity of the muscles 164. The raw sensor input 604 may include unprocessed and / or unfiltered sensor data collected and / or generated by the sensor suite 112. In some embodiments, the processing unit 104 may pass the raw sensor input 111 through one or more filters. The filters may include, but are not limited to, a noise filter. The filters may include nonlinear filters, linear filters, time-varying filters, time-invariant filters, causal filters, non-causal filters, discrete-time filters, continuous-time filters, passive filters, active filters, infinite impulse response (IIR) filters, finite impulse response (FIR) filters, etc. The processing unit 104 may use one or more filters, such as noise filter 608, to remove noise from the sensor output. Noise may include unwanted modifications to the signal, such as irrelevant sensor outputs of one or more sensors in the sensor suite 112. The noise filter 608 may use knowledge of the output waveform to subtract from the sensed waveform or may use knowledge of the timing of the output waveform to limit sensing to the "off" phase of the pulse stimulation. In some embodiments, the processing unit 104 may remove all information unrelated to the movement disorder through a filter, such as movement disorder filter 612. Information unrelated to the movement disorder may include specific frequencies and / or frequency ranges that may be outside the range of movement disorder indications.As a non-limiting example, a tremor may have a range of approximately 3 Hz to approximately 15 Hz; frequencies outside this range are not associated with tremor and may subsequently be filtered out through one or more filters. As another non-limiting example, a typical resting tremor, isolated postural tremor, and kinetic tremor during slow movement may be approximately 3 Hz to approximately 7 Hz, 4 Hz to approximately 9 Hz, and 7 Hz to approximately 12 Hz, respectively. The processing unit 104 may be configured to filter any frequencies outside any of the ranges described above. In some embodiments, the processing unit 104 may be configured to extract the fundamental tremor frequency through spectral analysis. The fundamental tremor frequency may be used in one or more filters, such as a digital band-pass filter with cutoff frequencies near and below the fundamental frequency. The processing unit 104 may be configured to implement and / or generate one or more filters based on the patient's specific fundamental tremor frequency. The movement disorder filter 612 may be any type of filter. In some embodiments, the movement impairment filter 612 may include a 0-15 Hz bandpass filter configured to reject any other signal components not caused by the movement impairment. In other embodiments, the movement impairment filter 612 may include, without limitation, a bandpass filter with an upper limit greater than 15 Hz. The processing unit 104 may use the movement impairment filter 612 to determine the user's extraneous movements by removing noise unrelated to the user's extraneous movements. In one embodiment, the processing unit 104 may utilize three or more filters. The processing unit 104 may first use the noise filter 608 to remove noise from the raw sensor input 604, and then use a second filter, such as the movement impairment filter 612, to remove all information unrelated to the movement impairment. In some embodiments, after processing the sensor output through one or more filters, filtered sensor data 616 may be generated. In some embodiments, one or more features may be extracted from the filtered sensor data 616.The extraction may include obtaining temporal, spectral, or other features of the filtered sensor data 616. Temporal features may include, but are not limited to, minimum, maximum, first three standard deviation values, signal energy, root-mean-square (RMS) amplitude, zero-crossing rate, principal component analysis (PCA), kernel or wavelet convolution, or autoconvolution. Spectral features may include, but are not limited to, Fourier transform, fundamental frequency, (Mel-frequency) cepstral coefficients, spectral centroid, and bandwidth. The processing unit 104 may input the filtered sensor data 616 and / or the extracted features of the filtered sensor data 616 to a waveform parameter algorithm 115.

[0030] The waveform parameter selection 620 may be a parameter selection algorithm. The parameter selection algorithm may include an algorithm that determines one or more parameters of an output. The waveform parameter selection 620 may include a classification algorithm, such as, but not limited to, logistic regression, naive Bayes, decision tree, support vector machine, neural network, random forest, and / or other algorithms. In some embodiments, the waveform parameter selection 620 may be an argmax (FFT) algorithm. The waveform parameter selection 620 may include calculation of the mean, median, interquartile range, X-percentile signal frequency, root-mean-square amplitude, power, log(power), and / or linear or nonlinear combinations thereof. For example, but not limited to, the waveform parameter selection 620 may modify the frequency, amplitude, peak-to-peak value, etc. of one or more waveforms. The waveform parameter algorithm 620 may modify one or more parameters of a waveform output applied to a mechanoreceptor 154, such as a vibration stimulus 13. In some embodiments, the waveform parameter algorithm 620 may be configured and / or programmed to determine the waveform parameter set based on the current waveform parameter set and / or the filtered sensor data 616. As a non-limiting example, the filtered sensor data 616 may include tremor amplitude. The waveform parameter algorithm 620 may compare the tremor amplitude from the current waveform parameter set to the tremor amplitude observed from the previous waveform parameter set to determine which of the two waveform parameter sets resulted in the smallest tremor amplitude. The resulting set with the smallest tremor amplitude may be used as a baseline for the next iteration of the waveform parameter selection 620, and the waveform parameter selection 620 may compare this baseline to the new waveform parameter set. The waveform parameter selection 620 may utilize one or more of a Q-learning model, one or more neural networks, a genetic algorithm, differential dynamic programming, an iterative quadratic regulator, and / or a guided policy search.The waveform parameter selection 620 may determine one or more new waveform parameters from the applied current waveform parameter set based on an optimization model to best minimize the severity of the user's symptoms. The optimization model may include, but is not limited to, discrete optimization, continuous optimization, etc. For example, the waveform parameter selection 620 may utilize an optimization model that may be configured to input the filtered sensor data 616 and / or the current waveform parameters of the vibration stimulus 13 and output a selection of new waveform parameters that may minimize the severity of the user's symptoms. The severity of the symptoms may include, but is not limited to, freezing of gait, stiffness, tremors, etc.

[0031] In some embodiments, the vibration stimulus 13 may target afferents selected from the set consisting of, but not limited to, somatosensory cutaneous afferents in the C5-T1 dermatomes and proprioceptive afferents in the muscles and tendons of the wrist, fingers, and thumb. In one embodiment, the vibration stimulus 13 may be applied around the user's wrist, allowing stimulation of five separate somatosensory channels through the C5-T1 dermatomes and an additional 15 proprioceptive channels through the tendons passing through the wrist, for a total of 20 separate channels. The waveform parameter selection 620 may be configured to generate one or more waveform parameters specific to one or more proprioceptive and / or somatosensory channels. For example, but not limited to, the waveform parameter selection 620 may select a single proprioceptive channel through the C5 dermatome for application of the vibration stimulus 13. In another example, but not limited to, the waveform parameter selection 620 may select a combination of the C5 dermatome and the T1 dermatome channel. In some embodiments, the waveform parameter selection 620 may be configured to generate a multi-channel waveform by generating one or more waveform parameters for one or more proprioceptive and / or somatosensory channels. The channels of the multi-channel waveform may be specific to one or more proprioceptive and / or somatosensory channels. In some embodiments, each transducer of a plurality of transducers may generate a waveform output for a specific proprioceptive and / or somatosensory channel, and the channels may be different, the same, or a combination thereof. The waveform parameter selection 620 may select any combination of proprioceptive and / or somatosensory channels, without limitation. The waveform parameter selection 620 may select one or more proprioceptive channels to target based on one or more symptoms of a movement disorder. For example, without limitation, the waveform parameter selection 620 may select both the T1 and C5 channels for stimulation based on a symptom of muscle stiffness. In some embodiments, the waveform parameter selection 620 may include a stimulation machine learning model.The stimulus machine learning model may include, but is not limited to, any of the machine learning models described throughout this disclosure. In some embodiments, the stimulus machine learning model may be trained using training data that correlates sensor data and / or waveform parameters to optimal waveform parameters. The training data may be received through user input, an external computing device, and / or previous processing iterations. The stimulus machine learning model may be configured to input filtered sensor data 616 and / or a current waveform parameter set and output a new waveform parameter set. The stimulus machine learning model may be configured to output a specific target for vibration stimulation, such as, but not limited to, one or more proprioceptive and / or somatosensory channels described above. As a non-limiting example, the stimulus machine learning model may input filtered sensor data 616 and output a waveform parameter set specific to the C6 and C8 proprioceptive channels. The vibration stimulus 13 may be applied to one or more mechanoreceptors 154. In some embodiments, if process 600 is performed external to wearable device 100, the computing device running process 600 may communicate one or more waveform parameters to wearable device 100.

[0032] The waveform parameter selection 620 may generate a train of waveform outputs. The waveform output train may include two or more waveform outputs that may be applied sequentially to the user. The time between two or more waveform outputs of the waveform output train may be, but is not limited to, milliseconds, seconds, minutes, etc. Each waveform output of the waveform output train may have various parameters, such as, but not limited to, amplitude, frequency, peak-to-peak value, etc. In some embodiments, the waveform output train may include multiple waveform outputs, each waveform output having a higher frequency than the previous waveform output. In some embodiments, each waveform output may have a lower frequency than or the same frequency as the previous waveform output. The waveform parameter selection 620 may provide a train of waveform outputs until the waveform output reaches a frequency that will inhibit the user's extraneous movement output.

[0033] 6, the wearable device 100 may be configured to operate in one or more settings. The settings of the wearable device 100 may include one or more operating modes. A user may be configured to select one or more settings of the wearable device 100 through interactive elements such as, but not limited to, buttons, a touchscreen, and / or through a remote computing device, such as via an application. Interactive elements and applications may be described in further detail below with reference to FIG. 7.

[0034] The settings of the wearable device 100 may include an automatic setting, a tremor reduction setting, a freezing setting, a stiffness setting, and / or an adaptive mode setting.

[0035] The automatic configuration of the wearable device 100 may include the processing unit 104 automatically selecting the best waveform output based on data generated from one or more sensors in the sensor suite 112. For example, the waveform parameter selection 620 may select one or more waveform parameters that are generally best suited to the current sensor data, such as the filtered sensor data 616. The automatic mode of the wearable device 100 may find one or more mean values, standard deviation values, etc. of the therapeutic vibration stimuli 13 based on multiple data generated from multiple users using the wearable device 100. In some embodiments, generating the automatic mode of the wearable device 100 may include crowdsourcing from one or more users. A cloud computing system may be implemented to collect data from one or more users.

[0036] With further reference to FIG. 6 , the wearable device 100 may be configured to operate in a tremor reduction setting. The tremor reduction setting may include waveform parameter selection 620 assigning a high weight or value to the filtered sensor data 616 corresponding to tremor while reducing the weight or value of other symptoms. The waveform parameter selection 620 may be configured to generate one or more waveform parameters that optimize tremor reduction for the user's tremor. Optimizing tremor reduction for the user may include minimizing weights, values, and / or waveform parameters for other symptoms, such as freezing of gait and stiffness. Similarly, a freezing of gait setting may optimize reduction of a user's freezing of gait, and a stiffness setting may optimize reduction of a user's stiffness. Each setting may be iteratively updated based on data received from crowdsourcing, user history data, etc. For example, each setting may be continuously updated to optimize symptom reduction for a majority of users from a population of multiple users. In some embodiments, the settings of the wearable device 100 may include an adaptive mode. The adaptive mode may include waveform parameter selection 620 continually searching for the highest weight and / or most severe symptom in sensor data 616 and generating one or more waveform parameters to reduce that symptom and / or weight. The adaptive mode of wearable device 100 may utilize a machine learning model, such as that described below with reference to FIG. 11 . The adaptive mode machine learning model may be trained using training data that correlates sensor data and / or weights of the sensor data to one or more waveform parameters. The training data may be received through user input, an external computing device, and / or previous processing iterations. The adaptive mode machine learning model may be configured to input filtered sensor data 616 and one or more optimal waveform parameters 620 to reduce the symptom with the highest severity. In some embodiments, the adaptive mode machine learning model may be trained remotely, and the trained model weights may be communicated to wearable device 100, thereby reducing the processing load on wearable device 100.

[0037] FIG. 7 illustrates a process of waveform parameter selection by a mobile device. Process 700 may be performed by a processor, such as, but not limited to, processing unit 104 described above with reference to FIG. 1 . Process 700 may include waveform parameter selection 704. Waveform parameter selection 704 may be the same as waveform parameter selection 620 described above with reference to FIG. 6 . In some embodiments, an application 708 may be configured to run. Application 708 may run on, but is not limited to, a laptop, desktop, tablet, smartphone, etc. In some embodiments, application 708 may take the form of a web application. Application 708 may be configured to display data to a user through a graphical user interface (GUI). The GUI may include one or more text, graphics, or other icons. The GUI generated by application 708 may include one or more windows capable of displaying data, such as images, text, etc. The GUI generated by application 708 may be configured to display sensor data, stimulus data, etc. In some embodiments, the GUI generated by application 708 may be configured to receive user input 712. User input 712 may include, but is not limited to, keystrokes, mouse input, touch input, etc. For example, but not limited to, a user may click on an icon in a GUI generated by application 708, which may trigger an event handler in application 708 to perform one or more actions, such as displaying data through a window, communicating data to another device, etc. In some embodiments, user input 712 received through application 708 may generate smartphone application data 716. Smartphone application data 716 includes one or more selections of one or more waveform parameters. Waveform parameters may include, but are not limited to, amplitude, frequency, etc.The waveform parameters may be as described above with reference to Figures 1 and 6. As a non-limiting example, the smartphone application data 716 may include a selection of a higher frequency for the waveform output, the selection being generated by user input through the application 708.

[0038] Additionally and / or alternatively, a user may generate user input 712 through one or more interactive elements of the wearable device. The wearable device may be as, but is not limited to, those described above in FIG. 1 . The wearable device may include one or more interactive elements such as, but not limited to, knobs, switches, buttons, sliders, etc. Each interactive element of the wearable device may correspond to a function. For example, one button on the wearable device may correspond to increasing the frequency of the waveform output, while another button on the wearable device may correspond to decreasing the frequency of the waveform output. A user may generate device button data 720 through user input 712 on the wearable device. In some embodiments, the wearable device may include a touchscreen or other interactive display through which the user may generate device button data 720. In one embodiment, the wearable device may be configured to run application 708 locally and receive smartphone application data 716 through a touchscreen or other input device that may be part of the wearable device. The waveform parameter selection 704 may be run locally on the wearable device and / or offloaded to one or more computing devices. In some embodiments, the waveform parameter selection 704 may be configured to receive smartphone application data 716 and / or device button data 720. The waveform parameter selection 704 may be configured to generate a waveform output, such as a vibration stimulus 13, based on the smartphone application data 716 and / or device button data 720. A user may adjust the vibration stimulus 13 through generating the smartphone application data 716 and / or device button data 720. The vibration stimulus 13 may be transmitted to one or more mechanoreceptors 154 through one or more transducers, as described above with reference to, but not limited to, FIGS. 1 and 6 .

[0039] Referring now to FIG. 8 , an example of a method 800 for alleviating symptoms of a movement disorder is shown. Method 800 may be applied and / or implemented in any of the processes described. Method 800 may be based on Hebbian learning. Hebbian learning refers to the neuropsychological theory that repeated and sustained stimulation of postsynaptic cells by presynaptic cells results in increased synaptic efficacy. For example, in some embodiments, a user may perform a predefined series of movements during stimulation, such as those described above with reference to FIGS. 1 and 5 . Performing the predefined series of movements during stimulation may induce neuroplastic changes that may persist even after stimulation has ceased. These movements can be performed under the guidance of a physical therapist, occupational therapist, other caregiver, or on the user's own initiative. Stimulating neuronal pathways during movement strengthens shared synapses between neurons associated with the movement over time, allowing for sustained therapeutic effects even in the absence of stimulation.

[0040] In step 805, the method includes orienting mechanical transducers in the wearable device to target mechanoreceptors in the affected region. For example, one or more mechanical transducers of the wearable device may be oriented around the user's wrist, arm, leg, etc.

[0041] In step 810, a wearable device may be placed on the subject's extremity. The wearable device may be worn during flexion and / or extension of one or more affected muscles of the user. In some embodiments, the user may perform one or more predefined movements such as, but not limited to, walking, clenching a fist, writing, or raising an arm.

[0042] Stimulation may be provided to the user through the wearable medical device while the user is exercising in step 815. For example, the user may be performing one or more predefined exercises described in step 810, while the wearable device simultaneously stimulates a part of the user's body. The mechanical transducer delivers vibration stimulation with a frequency between 1 Hz and 300 Hz.

[0043] At step 820, a determination is made that the therapy is complete. The determination may be made by a user, a professional, an application, a timer, and / or combinations thereof. In some embodiments, the wearable device may be configured to apply stimulation for a predetermined period of time. The predetermined period of time may be selected by the user, selected by the professional, and / or calculated by the wearable device through historical data. If it is determined at step 820 that the therapy is complete, the method proceeds to step 825, where stimulation is stopped. If it is determined at step 820 that the therapy is not complete, the method loops back 830 to step 815 for providing stimulation to the subject through the wearable device. Any one of the steps of method 800 may be performed as, but not limited to, described above with reference to FIGS. 1-7.

[0044] Referring now to FIG. 9 , a diagram of a wearable device 900 is presented. In some embodiments, the wearable device 900 may include a housing 904 that may be configured to house one or more components of the wearable device 900. For example, the housing 904 of the wearable device 900 may comprise a circular, oval, rectangular, square, or other shaped material. In some embodiments, the housing 904 may be, but is not limited to, approximately 5 inches long, approximately 5 inches wide, and approximately 5 inches long. In some embodiments, the housing 904 may be, but is not limited to, approximately 1.5 inches long, approximately 1.5 inches wide, and approximately 0.5 inches high. The housing 904 of the wearable device 900 may have an interior and an exterior. The interior of the housing 904 of the wearable device 100 may include, but is not limited to, one or more sensors, transducers, energy sources, processors, memory, etc., such as those described above with reference to FIG. 1 . In some embodiments, the exterior of the housing 904 of the wearable device 900 may include one or more interactive elements 916. As used in this disclosure, an "interactive element" is a component configured to respond to user input. Interactive elements 916 may include, but are not limited to, buttons, switches, etc. In some embodiments, the wearable device 900 may have a single interactive element 916. In other embodiments, the wearable device 900 may have two or more interactive elements 916. In embodiments in which the wearable device 900 has multiple interactive elements 916, each interactive element 916 corresponds to a different function. For example, a first interactive element 916 may correspond to a power function, a second interactive element 916 may correspond to a waveform adjustment, a third interactive element 916 may correspond to a mode of the wearable device 900, and so on. In some embodiments, the wearable device 900 may include a touchscreen display.

[0045] In some embodiments, the wearable device 900 may include one or more batteries. For example, without limitation, the wearable device 900 may include one or more replaceable batteries, such as lead-acid, nickel-cadmium, nickel-metal hydride, lithium-ion, and / or other types of batteries. The housing 904 of the wearable device 900 may include a charging port that allows access to the rechargeable battery of the wearable device 900. For example, without limitation, the wearable device 900 may include one or more rechargeable lithium-ion batteries, and the charging port of the housing 904 of the wearable device 900 may be a USB-C, micro-USB, and / or other type of port. The battery of the wearable device 900 may be configured to charge at a rate of approximately 10 W / hour. The battery of the wearable device 900 may be configured to charge with a current consumption of approximately 630 mA at approximately 3.7 V. The battery of the wearable device 900 may have a capacity of, without limitation, approximately 2.5 Wh, greater than 2.5 Wh, or less than 2.5 Wh. In some embodiments, the wearable device 900 may include one or more wireless charging circuits that may be configured to receive power via electromagnetic waves. The wearable device 900 may be configured to be wirelessly charged at a rate of about 5 W / hr via a charging pad or other wireless power transmission system. In some embodiments, the battery of the wearable device 900 may be configured to be charged at about, greater than, or less than 460 mA.

[0046] With further reference to FIG. 9 , the wearable device 900 may include an attachment system. The attachment system may include any component configured to secure two or more elements together. For example, without limitation, the wearable device 900 may include a wristband 908. The wristband 908 may include one or more layers of material. For example, without limitation, the wristband 908 may include multiple layers of a polymer, such as rubber. The wristband 908 may have an interior and an exterior. The interior and exterior of the wristband 908 may be the same material, texture, etc. In other embodiments, the interior of the wristband 908 may be softer and / or smoother than the exterior of the wristband 908. As a non-limiting example, the interior of the wristband 908 may be a smooth rubber material, and the exterior of the wristband 908 may be a Velcro material. The wristband 908 may be approximately 2 mm thick. In other embodiments, the wristband 908 may be greater than or less than approximately 2 mm thick. The wristband 908 may be an elastic band, a Velcro strap, or the like. In some embodiments, the wristband 908 may be adjustable. For example, the wristband 908 may be a flexible loop that attaches to itself via a Velcro attachment system. In some embodiments, the wristband 908 may be attached to one or more hooks 912 on the exterior of the housing 904 of the wearable device 900. In some embodiments, the wristband 908 may be magnetic. In other embodiments, the wristband 908 may include a row, grid, or other arrangement of holes that can accept a latch from the hook 912.

[0047] Referring now to FIG. 10 , an exploded side view of a wearable device 900 is shown. The wearable device 900 may include a mechanical transducer 1000. The mechanical transducer 1000 may be housed within a wristband 908. The wristband 908 may be configured to interface with a user's wrist. The wearable device 900 may have an upper housing half 1024 and a lower housing half 1020. In some embodiments, a printed circuit board 1004 (PCB) may be disposed between the upper half 1024 and the lower half 1020. Additionally, a silicone square may be disposed to insulate the bottom of the PCB 1004, which may be disposed over a battery 1016. The battery 1016 may include protection circuitry to protect against overcharging and unwanted discharge. In some embodiments, the wearable device 900 may include a magnetic connector 1008. The magnetic connector 1008 may be configured to align the wearable device 900 with a charging pad, station, or the like. The magnetic connector 1008 may be configured to receive power wirelessly to charge the battery 1016. The magnetic connector 1008 may be coupled to the battery 1016 and mounted to the housing 1020 and / or 1024. In some embodiments, the magnetic connector 1008 may be inserted into the PCB 1004. The magnetic connector 1008 may be configured to mate with a connector from an external charger.

[0048] Referring to FIG. 11, an example machine learning module 1100 may perform machine learning processes and may be configured to use the machine learning processes to perform various determinations, calculations, processes, etc. described in this disclosure.

[0049] With further reference to FIG. 11 , the machine learning module 1100 may utilize training data 1104. For example, but not limited to, the training data 1104 may include multiple data entries, each representing a set of data elements recorded, received, and / or generated together. The training data 1104 may include data elements that may be correlated by coexistence within a given data entry, proximity within a given data entry, etc. The multiple data entries in the training data 1104 may demonstrate one or more trends in correlations between categories of data elements. For example, but not limited to, higher values ​​of a first data element belonging to a first category of data elements may tend to correlate with higher values ​​of a second data element belonging to a second category of data elements, the trend indicating a possible proportionality or other mathematical relationship linking values ​​belonging to the two categories. In the training data 1104, multiple categories of data elements may be associated according to various correlations. Correlations may indicate causal and / or predictive connections between categories of data elements, which may be modeled as relationships, such as mathematical relationships, by a machine learning process, as described in further detail below. The training data 1104 may be formatted and / or organized by categories of data elements. The training data 1104 may be organized, for example, by associating data elements with one or more descriptors that correspond to the categories of the data elements. As a non-limiting example, the training data 1104 may include data entered by one or more individuals in a standardized form, such that the entry of a given data element in a given field in the form can be mapped to one or more descriptors of the category. Elements in the training data 1104 may be linked to the category descriptors by tags, tokens, or other data elements. The training data 1104 may be provided in a fixed-length format, a format that links data location to categories, such as comma-separated values ​​(CSV), and / or a self-describing format.Self-describing formats may include, but are not limited to, Extensible Markup Language (XML), JavaScript Object Notation (JSON), etc., which may allow a process or device to discover categories of data.

[0050] Continuing with reference to FIG. 11 , the training data 1104 may include one or more uncategorized elements. The uncategorized data in the training data 1104 may include data that may be unformatted or that includes descriptors for some elements of the data. In some embodiments, machine learning algorithms and / or other processes may sort the training data 1104 according to one or more categorizations. The machine learning algorithm may sort the training data 1104 using, for example, natural language processing algorithms, tokenization, detecting correlation values ​​in the raw data, etc. In some embodiments, categories of the training data 1104 may be generated using correlation algorithms and / or other processing algorithms. As a non-limiting example, in a body of text, phrases that make up “n” compounds, such as nouns modified by other nouns, may be identified according to statistically significant occurrences of n-grams that include such words in a particular order. For example, n-grams may be categorized as linguistic elements, such as “words,” that are tracked similarly to single words, thereby generating new categories as a result of the statistical analysis. In data entries that include any text data, names of people may be identified by reference to a list, dictionary, or other glossary to allow for ad-hoc categorization by a machine learning algorithm and / or automated association of data in the data entry with a descriptor or given format. The ability to automatically categorize data entries may allow the same training data 1104 to be applied to two or more separate machine learning algorithms, as described in more detail below. The training data 1104 used by the machine learning module 1100 may associate, without limitation, any of the input data described in this disclosure with any of the output data described in this disclosure.

[0051] With further reference to FIG. 11 , the training data 1104 may be filtered, sorted, and / or selected using one or more supervised and / or unsupervised machine learning processes and / or models, as described in further detail below. In some embodiments, the training data 1104 may be classified using a training data classifier 1116. The training data classifier 1116 may include a classifier. The training data classifier 1116 may utilize a mathematical model, neural net, or program generated by a machine learning algorithm. The machine learning algorithm of the training data classifier 1116 may include a classification algorithm. As used in this disclosure, a "classification algorithm" is one or more computer processes that generate a classifier from training data. The classification algorithm may sort input into categories and / or bins of data. The classification algorithm may output data categories and / or labels associated with the data. The classifier may be configured to output datums that label or identify data sets that can be clustered together. The machine learning module 1100 may use a classification algorithm to generate a classifier, such as the training data classifier 1116. The classification may be performed using, but is not limited to, linear classifiers such as logistic regression and / or naive Bayes classifiers, nearest neighbor classifiers such as k-nearest neighbor classifiers, support vector machines, least squares support vector machines, Fisher's linear discriminant, quadratic classifiers, decision trees, boosted trees, random forest classifiers, learning vector quantization, and / or neural network-based classifiers. As a non-limiting example, the training data classifier 1116 may classify elements of the training data into one or more faces.

[0052] With further reference to FIG. 11 , the machine learning module 1100 may be configured to execute a lazy-learning process 1120, which may include a “lazy load” or “call on demand” process and / or protocol. A “lazy learning process” may include a process in which machine learning is performed by receiving an input to be converted into an output, on demand, by combining the input with a training set and deriving an algorithm used to create the output. For example, an initial series of simulations may be performed to target an initial heuristic and / or “first guess” at the output and / or relationships. As a non-limiting example, the initial heuristic may include ranking associations between the input and elements of the training data 1104. The heuristic may include selecting some of the highest-ranked associations and / or elements of the training data 1104. Lazy learning may implement any suitable lazy learning algorithm, including, but not limited to, a K-nearest neighbor algorithm, a lazy Naive Bayes algorithm, etc. Those skilled in the art, upon consideration of this disclosure as a whole, will recognize a variety of lazy learning algorithms that may be applied to generate outputs as described in this disclosure, including but not limited to lazy learning applications of machine learning algorithms, which are described in more detail below.

[0053] With further reference to FIG. 11 , the machine learning processes described in this disclosure may be used to generate a machine learning model 1124. As used in this disclosure, a “machine learning model” is a mathematical and / or algorithmic representation of a relationship between inputs and outputs, such as may be generated and stored in memory using any machine learning process, including, but not limited to, any of the processes described above. For example, inputs may be sent to the machine learning model 1124, which, once created, may generate an output as a function of the derived relationship. For example, but not limited to, a linear regression model generated using a linear regression algorithm may calculate a linear combination of the input data using coefficients derived during the machine learning process to generate an output. As a further non-limiting example, the machine learning model 1124 may be generated by creating an artificial neural network, such as a convolutional neural network, including an input layer of nodes, one or more hidden layers, and an output layer of nodes. Connections between nodes may be created through a process of "training" the network, in which elements from a training data 1104 set are applied to input nodes, and then a suitable training algorithm (such as Levenberg-Marquardt, conjugate gradient, simulated annealing, or other algorithm) is used to adjust the connections and weights between nodes in adjacent layers of the neural network to create desired values ​​at the output nodes. This process is sometimes referred to as deep learning.

[0054] With further reference to FIG. 11 , the machine learning algorithm may include a supervised machine learning process 1128. As used in this disclosure, a “supervised machine learning process” is one or more algorithms that receive labeled input data and generate an output according to the labeled input data. For example, the supervised machine learning process 1128 may include the above-described movement data as input, movement disorder symptoms as output, and a scoring function that represents a desired type of relationship to be detected between the input and the output. The scoring function may maximize the probability that a given input and / or combination of component inputs is associated with a given output and minimize the probability that a given input is not associated with a given output. The scoring function may be expressed as a risk function that represents the “expected loss” of the algorithm that associates inputs with outputs, where the loss is calculated as an error function that represents the degree to which a prediction generated by the relationship is inaccurate when compared to a given input-output pair provided in the training data 1104. By considering this entire disclosure, one skilled in the art will recognize various possible variations of at least the supervised machine learning process 1128 that may be used to determine a relationship between an input and an output. The supervised machine learning process may include the classification algorithm defined above.

[0055] With further reference to FIG. 11 , the machine learning process may include an unsupervised machine learning process 1132. As used in this disclosure, an “unsupervised machine learning process” is a process that calculates relationships between one or more data sets without labeled training data. The unsupervised machine learning process 1132 may be free to find any structure, relationships, and / or correlations provided in the training data 1104. The unsupervised machine learning process 1132 may not require a response variable. The unsupervised machine learning process 1132 may calculate patterns, inferences, correlations, etc. between two or more variables of the training data 1104. In some embodiments, the unsupervised machine learning process 1132 may determine the degree of correlation between two or more elements of the training data 1104.

[0056] With further reference to FIG. 11 , the machine learning module 1100 may be designed and configured to create the machine learning model 1124 using techniques for developing linear regression models. The linear regression model may include ordinary least squares regression, which aims to minimize the squared difference between predicted and actual results according to an appropriate norm for measuring such difference (e.g., a vector space distance norm). The coefficients of the resulting linear equation may be modified to improve the minimization. The linear regression model may include ridge regression, in which the function to be minimized includes a least-squares function and a term that penalizes large coefficients by multiplying the square of each coefficient by a scalar. The linear regression model may also include a least absolute shrinkage and selection operator (LASSO) model, in which ridge regression is combined with a least-squares term multiplied by a coefficient equal to I divided by twice the number of samples. The linear regression model may include a multitask lasso model, in which the norm applied to the least-squares terms of the lasso model is the Frobenius norm, which corresponds to the square root of the sum of the squares of all terms. The linear regression model may include an elastic net model, a multitask elastic net model, a least-angle regression model, a LARS lasso model, an orthogonal matching pursuit model, a Bayesian regression model, a logistic regression model, a stochastic gradient descent model, a perceptron model, a passive-aggressive algorithm, a robust regression model, a Huber regression model, or any other suitable model that may occur to one skilled in the art upon consideration of this disclosure in its entirety. In one embodiment, the linear regression model may be generalized to a polynomial regression model, thereby finding a polynomial equation (e.g., a quadratic, cubic, or higher-order equation) that provides the best fit between the predicted output and the actual output. As will be apparent to one skilled in the art upon consideration of this disclosure in its entirety, methods similar to those described above may be applied to minimize the error function.

[0057] With continued reference to FIG. 11 , the machine learning algorithm may include, but is not limited to, linear discriminant analysis. The machine learning algorithm may include quadratic discriminant analysis. The machine learning algorithm may include kernel ridge regression. The machine learning algorithm may include support vector machines, including, but not limited to, support vector classification-based regression processes. The machine learning algorithm may include stochastic gradient descent algorithms, including classification and regression algorithms based on stochastic gradient descent. The machine learning algorithm may include nearest neighbor algorithms. The machine learning algorithm may include various forms of latent space regularization, such as variational regularization. The machine learning algorithm may include Gaussian processes, such as Gaussian process regression. The machine learning algorithm may include cross-decomposition algorithms, including partial least squares and / or canonical correlation analysis. The machine learning algorithm may include naive Bayes methods. The machine learning algorithm may include decision tree-based algorithms, such as decision tree classification or regression algorithms. The machine learning algorithms may include ensemble methods such as bagging meta-estimators, randomized tree forests, AdaBoost, gradient tree boosting, and / or voting classifier methods. The machine learning algorithms may include neural net algorithms, including convolutional neural net processes.

[0058] 12 illustrates an exemplary computer for implementing the systems and methods described herein. In some embodiments, the computing device includes at least one processor 1202 coupled to a chipset 1204. The chipset 1204 includes a memory controller hub 1220 and an input / output (I / O) controller hub 1222. The memory 1206 and the graphics adapter 1212 are coupled to the memory controller hub 1220, and the display 1218 is coupled to the graphics adapter 1212. The storage device 1208, the input interface 1214, and the network adapter 1216 are coupled to the I / O controller hub 1222. Other embodiments of the computing device have different architectures.

[0059] The storage device 1208 is a non-transitory computer-readable storage medium such as a hard drive, a compact disc read-only memory (CD-ROM), a DVD, or a solid-state memory device. The memory 1206 holds instructions and data used by the processor 1202. The input interface 1214 is a touchscreen interface, a mouse, a trackball, or other type of input interface, a keyboard, or a combination thereof, and is used to input data into the computing device. In some embodiments, the computing device may be configured to receive input (e.g., commands) from the input interface 1214 via gestures from a user. The graphics adapter 1212 displays images and other information on the display 1218. The network adapter 1216 couples the computing device to one or more computer networks.

[0060] Graphics adapter 1212 displays representations, graphs, tables, and other information on display 1218. In various embodiments, display 1218 is configured to allow a user (e.g., a data scientist, a data owner, a data partner) to input user selections on display 1218. In one embodiment, display 1218 may include a touch interface. In various embodiments, display 1218 may show one or more forecasted lead times for fulfilling customer orders.

[0061] The computing device 1200 is adapted to execute computer program modules to provide the functionality described herein. As used herein, the term "module" refers to computer program logic used to provide a specified functionality. Thus, a module may be implemented in hardware, firmware, and / or software. In one embodiment, the program module is stored on the storage device 1208, loaded into the memory 1206, and executed by the processor 1202.

[0062] The type of computing device 1200 may differ from the embodiments described herein. For example, the system may operate within a single computer 1200 or within multiple computers 1200 communicating with each other over a network, such as a server farm. In another example, the computing device 1200 may lack some of the components described above, such as the graphics adapter 1212, the input interface 1214, and the display 1218.

[0063] The foregoing has been a detailed description of exemplary embodiments of the present invention. Various modifications and additions may be made thereto without departing from the spirit and scope of the present invention. Each feature of the various embodiments described above may be combined, as appropriate, with features of other described embodiments to provide various combinations of features in related new embodiments. Furthermore, while the above describes several individual embodiments, what has been described herein is merely illustrative of the application of the principles of the present invention. Furthermore, although certain methods herein may be illustrated and / or described as being performed in a particular order, the order may vary considerably within the ordinary skill of the art to implement methods, systems, and software according to the present disclosure. Accordingly, this description should be construed as exemplary only, and not as limiting the scope of the present invention.

[0064] Exemplary embodiments have been disclosed above and illustrated in the accompanying drawings. It will be understood by those skilled in the art that various modifications, omissions, and additions may be made to what is specifically disclosed herein without departing from the spirit and scope of the present invention.

Claims

1. A wearable device for vibration stimulation, comprising: a sensor configured to receive data and generate a sensor output; a processor in communication with the sensor; a memory communicatively connected to the processor; wherein the memory comprises: receiving the sensor output; determining a symptom of a movement disorder of the user based on the sensor output; calculating a waveform output based on the symptoms of the movement disorder; directing a transducer in communication with the processor to apply the waveform output to the user to alleviate the symptoms of the movement disorder.

11. A wearable device comprising: instructions for configuring a processor to:

2. 10. The wearable device of claim 1, wherein the symptom of movement disorder is one of stiffness, rigidity, freezing, paresis, paresis, dyskinesia, or a combination thereof.

3. The wearable device of claim 1 , wherein the transducer is further configured to apply the waveform output to a proprioceptive nerve of the user.

4. 4. The wearable device of claim 3, wherein the proprioceptive nerve is within the proprioceptive tissue of one of the flexor carpi radialis, flexor carpi ulnaris, extensor carpi radialis, extensor carpi ulnaris, or combinations thereof.

5. 2. The wearable device of claim 1, wherein the transducer is positioned to apply the waveform output to one of the C5, C6, C7, C8, or T1 dermatomes.

6. 10. The wearable device of claim 1, wherein the processor is further configured to determine the extraneous movements of the user by processing the sensor output to remove noise unrelated to the extraneous movements of the user.

7. the processor: calculating a frequency of the waveform output that reduces the amplitude of the extraneous motion; applying the frequency of the waveform output to reduce the amplitude of the extraneous motion. The wearable device of claim 6 further configured to:

8. The wearable device of claim 1 , wherein the processor is further configured to generate a multi-channel waveform output and apply the multi-channel waveform output to the user through the transducer.

9. The wearable device of claim 1 , wherein each channel of the multi-channel waveform is calculated to target a specific proprioceptive channel of the user.

10. 2. The wearable device of claim 1, wherein the processor is further configured to provide a train of waveform outputs through the transducer, each waveform output in the train having a higher frequency than the previous waveform output until the waveform output reaches a frequency with an output to input ratio that will suppress the output of the extraneous movement.

11. 1. A method of reducing symptoms of a movement disorder through a wearable device, comprising: receiving user data via a sensor in a wearable device; generating a sensor output based on the data of the user; communicating the sensor output to a processor of the wearable device; determining, in the processor, a symptom of a movement disorder based on the sensor output; calculating, in the processor, a waveform output based on the symptoms of the movement disorder; instructing a transducer in communication with said processor to apply said waveform output to said user; A method comprising:

12. 12. The method of claim 11, wherein the symptom of movement disorder is any of stiffness, rigidity, freezing, paralysis, paresis, dyskinesia, or a combination thereof.

13. The method of claim 11 , further comprising applying the waveform output to a proprioceptive nerve of the user.

14. 14. The method of claim 13, wherein the proprioceptive nerve is within the proprioceptive tissue of one of the flexor carpi radialis, flexor carpi ulnaris, extensor carpi radialis, extensor carpi ulnaris, or combinations thereof.

15. 12. The method of claim 11, wherein the transducer is positioned within the wearable device to apply the waveform output to one of the user's C5, C6, C7, C8, or T1 dermatomes.

16. The method of claim 11 , further comprising the step of determining, by the processor, the extraneous motion of the user by processing the sensor output to remove noise unrelated to the extraneous motion of the user.

17. calculating, by the processor, a frequency of the waveform output that reduces the amplitude of the extraneous motion; applying the frequency of the waveform output through the transducer to reduce the amplitude of the extraneous motion; 17. The method of claim 16, further comprising:

18. generating, by said processor, a multi-channel waveform; applying the multi-channel waveform output to the user through the transducer; The method of claim 11 further comprising:

19. 20. The wearable device of claim 18, wherein generating the multi-channel waveform comprises generating multiple channels of a waveform, each channel of the multi-channel waveform calculated to target a specific proprioceptive channel of the user.

20. 12. The method of claim 11, further comprising providing a train of waveform outputs through the transducer, each waveform output in the train having a higher frequency than the previous waveform output until the waveform output reaches a frequency with an output to input ratio that will suppress the output of the extraneous movement.

Images