A conversion device includes: a unit attribute update unit that specifies a unit associated with a representative value approximating a feature vector for a biological signal, and updates a representative value and energy of the specified unit; a unit update unit that, when a plurality of feature vectors outside a predetermined range of a representative value of a unit of the unit data are acquired within a predetermined time, adds data for a new unit identifier to unit data; a class update unit that, when there is no class including a unit having energy within a predetermined range of energy of the new unit, adds data for a new class identifier to class data; and a motion update unit that updates motion data by associating, with an identifier of a class acquired after updating the class data, an identifier of a motion corresponding to the class.

Systems and methods for postural control of a multi-function prosthesis are provided. Various embodiments provide for a postural controller that use EMG signals to drive a point in a posture space and outputs continuously varying joint angles for a powered prosthetic hand. The postural controller can include an EMG signal processing unit to receive signals from electrodes for processing (e.g., band pass filtering, rectification, root mean square averaging, dynamic tuning, etc.). The processed EMG signals can then be combined or converted to produce a point in the postural control domain. The PC domain map defines the posture that corresponds to each PC cursor coordinate. This map can have limitless possible postures and limitless possible positions of the postures. The Joint Angle Transform converts the PC cursor coordinate into the joint angle array which is sent to the prosthetic hand thereby creating more natural movements.

Systems and methods for postural control of a multi-function prosthesis are provided. Various embodiments provide for a postural controller that use EMG signals to drive a point in a posture space and outputs continuously varying joint angles for a powered prosthetic hand. The postural controller can include an EMG signal processing unit to receive signals from electrodes for processing (e.g., band pass filtering, rectification, root mean square averaging, dynamic tuning, etc.). The processed EMG signals can then be combined or converted to produce a point in the postural control domain. The PC domain map defines the posture that corresponds to each PC cursor coordinate. This map can have limitless possible postures and limitless possible positions of the postures. The Joint Angle Transform converts the PC cursor coordinate into the joint angle array which is sent to the prosthetic hand thereby creating more natural movements.

One aspect of the present disclosure can include an intrafascicular neural electrode. The intrafascicular neural electrode can include a microwire body having a proximal end, a distal anchoring end, and a middle portion extending between the proximal end and the distal anchoring end. The distal anchoring end can substantially match the mechanical and biological properties of the target nerve. The microwire body can have a middle anchoring portion extending between the proximal end and the distal end, wherein at least a portion of the distal end and/or the middle anchoring portion substantially match(es) the mechanical and biological properties of the target nerve. The electrode can be made of graphene. The microwire body, except for the distal anchoring end, can be coated with an insulation material, preferably with a biocompatible agent adsorbed onto the insulation material.

A technology is described for determining an intended movement from neuromuscular signals. An example method (800) includes receiving electromyography (EMG) data corresponding to single-ended channels of an electrode array (810), where EMG signals are detected by electrodes comprising the single-ended channels of the electrode array and the EMG signals are converted to the EMG data. Determining differential channel pairs for the single-ended channels of the electrode array (820) and extracting feature data from the EMG data of the differential channel pairs (830). Thereafter a feature data set is selected from the feature data of the differential channel pairs (840) and the feature data set is input to a decode model configured to correlate the feature data set to an intended movement (850). Decode output is received from the decode model indicating the intended movement (860) and the decode output is provided to a device (870).

A technology is described for determining an intended movement from neuromuscular signals. An example method (800) includes receiving electromyography (EMG) data corresponding to single-ended channels of an electrode array (810), where EMG signals are detected by electrodes comprising the single-ended channels of the electrode array and the EMG signals are converted to the EMG data. Determining differential channel pairs for the single-ended channels of the electrode array (820) and extracting feature data from the EMG data of the differential channel pairs (830). Thereafter a feature data set is selected from the feature data of the differential channel pairs (840) and the feature data set is input to a decode model configured to correlate the feature data set to an intended movement (850). Decode output is received from the decode model indicating the intended movement (860) and the decode output is provided to a device (870).

Stimulation of nervous system components by electrodes can be used in many applications, including in the operation of brain-machine interfaces, bidirectional neural interfaces, and neuroprosthetics. The optimal operation of such systems requires a means of accurately measuring neural responses to such stimulations. However, currently the measurement of neural responses is difficult due to heavy stimulation artifacts arising from stimulatory pulses. The invention encompasses novel methods of estimating stimulation artifacts in measurements attained by recording electrodes and the effective removal of these artifacts. This provides improved neural recording systems and enables the deployment of closed-loop neural stimulation systems.

Stimulation of nervous system components by electrodes can be used in many applications, including in the operation of brain-machine interfaces, bidirectional neural interfaces, and neuroprosthetics. The optimal operation of such systems requires a means of accurately measuring neural responses to such stimulations. However, currently the measurement of neural responses is difficult due to heavy stimulation artifacts arising from stimulatory pulses. The invention encompasses novel methods of estimating stimulation artifacts in measurements attained by recording electrodes and the effective removal of these artifacts. This provides improved neural recording systems and enables the deployment of closed-loop neural stimulation systems.

Method(s) and apparatus are provided for interfacing with a nervous system of a subject. In response to receiving a plurality of neurological signals associated with the neural activity of the first portion of nervous system: processing neural sample data representative of the received plurality of neurological signals using a first one or more machine learning (ML) technique(s) trained for generating estimates of neural data representative of the neural activity of the first portion of nervous system; and transmitting data representative of the neural data estimates to a first device associated with the first portion of nervous system; and in response to receiving device data from a second device associated with a second portion of the nervous system: generating one or more neurological stimulus signal(s) by inputting the received device data to a second one or more ML technique(s) trained for estimating one or more neurological stimulus signal(s) associated with the device data for input to the second portion of nervous system; and transmitting the one or more estimated neurological stimulus signal(s) towards the second portion of nervous system of the subject.

A user-specific model of muscular activity can be used to control an external device based on muscular activity within a limb of a user. The user-specific model of muscular activity can include single movements and corresponding one or more primary muscle patterns. New single movements can be added to the user-specific model of muscular activity can be by a system that includes a processor by receiving user-specific EMG signals (including one or more EMG patterns that indicate a single movement); decomposing the user-specific EMG signals into the one or more EMG patterns in EMG feature space that indicate the single movement; and updating the user-specific model of muscular activity to include the single movement and corresponding one or more primary muscle patterns based on the one or more EMG patterns in EMG feature space.