PROSTHETIC DEVICES AND METHODS AND SYSTEMS RELATED THERETO
First Claim
Patent Images
1. A method of automatically positioning an electrode disposed inside a brain, the method comprising:
- moving an electrode along a predetermined path of travel until a neural signal having a current signal quality metric that is greater than or equal to a minimum threshold value and comprising action potentials is detected;
after the neural signal is detected, moving the electrode to a plurality of locations along the predetermined path and after each move, recording a sample of the neural signal;
detecting action potentials in the neural signal of each of the recorded samples;
classifying the action potentials of each of the recorded samples into classes;
calculating a first signal quality metric for each of the classes;
for each of the recorded samples, identifying a dominant class;
for each of the action potentials of the dominant class of each of the recorded samples, determining a second signal quality metric;
modeling an isolation curve as a function of the second signal quality metric for each of the action potentials of the dominant class of a first portion of the recorded samples and the locations of the plurality of locations where each of the recorded samples of the first portion of the recorded samples were recorded;
determining a move distance from a current electrode position to a location along the modeled isolation curve that the modeled isolation curve indicates will improve signal quality; and
as long as the move distance is greater than a predetermined minimum move distance threshold,moving the electrode the move distance to a new electrode position along the predetermined path and including the new electrode position in the plurality of locations,recording a new sample of the neural signal and including the new sample in the recorded samples,detecting action potentials in the new sample of the neural signal,classifying the action potentials of the new sample into classes,calculating the first signal quality metric for each of the classes,identifying the dominant class,for each of the action potentials of the dominant class, determining the second signal quality metric,modeling the isolation curve as a function of the second signal quality metric of each of the action potentials in the dominant class of a second portion of the recorded samples and the locations of the plurality of locations where each of the recorded samples of the second portion of the recorded samples were recorded, anddetermining the move distance from the new electrode position to a location along the modeled isolation curve that the modeled isolation curve indicates will improve signal quality.
2 Assignments
0 Petitions
Accused Products
Abstract
The present invention relates to systems and methods for controlling neural prosthetic devices and electrophysiological recording equipment, and for using the same in clinical operation. Various embodiments of the invention are directed to an algorithm for autonomously isolating and maintaining neural action potential recordings. The algorithm may be used in connection with a neural interface microdrive capable of positioning electrodes to record signals from active neurons.
65 Citations
33 Claims
-
1. A method of automatically positioning an electrode disposed inside a brain, the method comprising:
-
moving an electrode along a predetermined path of travel until a neural signal having a current signal quality metric that is greater than or equal to a minimum threshold value and comprising action potentials is detected; after the neural signal is detected, moving the electrode to a plurality of locations along the predetermined path and after each move, recording a sample of the neural signal; detecting action potentials in the neural signal of each of the recorded samples; classifying the action potentials of each of the recorded samples into classes; calculating a first signal quality metric for each of the classes; for each of the recorded samples, identifying a dominant class; for each of the action potentials of the dominant class of each of the recorded samples, determining a second signal quality metric; modeling an isolation curve as a function of the second signal quality metric for each of the action potentials of the dominant class of a first portion of the recorded samples and the locations of the plurality of locations where each of the recorded samples of the first portion of the recorded samples were recorded; determining a move distance from a current electrode position to a location along the modeled isolation curve that the modeled isolation curve indicates will improve signal quality; and as long as the move distance is greater than a predetermined minimum move distance threshold, moving the electrode the move distance to a new electrode position along the predetermined path and including the new electrode position in the plurality of locations, recording a new sample of the neural signal and including the new sample in the recorded samples, detecting action potentials in the new sample of the neural signal, classifying the action potentials of the new sample into classes, calculating the first signal quality metric for each of the classes, identifying the dominant class, for each of the action potentials of the dominant class, determining the second signal quality metric, modeling the isolation curve as a function of the second signal quality metric of each of the action potentials in the dominant class of a second portion of the recorded samples and the locations of the plurality of locations where each of the recorded samples of the second portion of the recorded samples were recorded, and determining the move distance from the new electrode position to a location along the modeled isolation curve that the modeled isolation curve indicates will improve signal quality. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A method of monitoring a neural signal using an electrode disposed inside a brain comprising:
-
moving the electrode inside the brain to a plurality of locations and at each location, collecting a sample of the neural signal to obtain a plurality of samples of the neural signal; collecting a sample of the neural signal to obtain a new sample of the neural signal; detecting action potentials in the new neural signal, each action potential having an amplitude; using unsupervised clustering to classify the action potentials into clusters; identifying a dominant cluster; calculating a current signal quality metric as a function of the amplitude of the action potentials of the dominant cluster; and as long as the current signal quality metric is less than a minimum signal quality threshold, modifying the plurality of samples to include the new sample, using a portion of the plurality of samples to create a model of an isolation curve of the neural signal, determining a move distance to a new location that the model indicates will improve signal quality, moving the electrode the move distance, after moving the electrode, re-sampling the neural signal to replace the new sample, detecting action potentials in the new neural signal, each action potential having an amplitude, using unsupervised clustering to classifying the action potentials into clusters, re-identifying the dominant cluster, and recalculating the current signal quality metric as a function of the amplitude of the action potentials of the dominant cluster. - View Dependent Claims (19, 20, 21, 22, 23, 24, 25, 26, 27)
-
-
28. A system comprising:
-
a microdrive comprising an electrode disposed within a brain, the electrode being configured to detect a neural signal of the brain, and a piezoelectric actuator configured to move the electrode along a predetermined path of travel in response to a command signal; a memory comprising instructions; and a processor coupled to the memory, coupled to the electrode, and configured to execute the instructions and provide the command signal to the electrode, wherein the instructions comprise instructions directing the processor to generate the command signal, the instructions comprising; instructions for generating the command signal to move the electrode a predetermined search distance, after the electrode is moved detect action potentials in the neural signal detected by the electrode, and until action potentials are not detected repeatedly generate the command signal to move the electrode the predetermined search distance, instructions for generating a command signal to move the electrode a second predetermined distance a predetermined number of times, and after each move, record the neural signal, and instructions for using the predetermined number of recorded neural signals detected by the electrode to model an isolation curve, use the isolation curve to determine a move distance, and if the move distance is greater than a minimum move threshold, generate a command signal to move the electrode the move distance.
-
-
29. A computer readable-medium having instructions executable by a processor coupled to a microdrive comprising a movable electrode, the instructions comprising:
-
instructions for detecting action potentials comprising; instructions for determining a search distance, instructions for moving the electrode of the microdrive the search distance, after the move, instructions for recording the neural signal, instructions for determining whether action potentials are present in the recorded neural signal, and if action potentials are not present in the recorded neural signal, instructions directing the processor to execute the instructions for detecting action potentials; instructions for moving the electrode to a plurality of electrode positions and after each move, recording the neural signal to create a plurality of neural signal recordings each having a corresponding recorded electrode position; instructions for modeling an isolation curve of a neuron having an order using the neural signal recordings and corresponding recorded electrode positions and if the order is greater than a predetermined order threshold, instructions for using the model of the isolation curve to determine a move distance to a new location that the model of the isolation curve indicates will improve signal quality; and instructions for iteratively moving the electrode comprising; comparing the move distance to a move threshold value, if the move distance is greater than the move threshold value, moving the electrode of the microdrive the move distance to a new electrode position, after the move, recording a new neural signal and the new electrode position, determining a signal quality metric using the new neural signal, and comparing the signal quality metric to a minimum signal quality threshold, if the signal quality metric is below the minimum signal quality threshold, modifying the plurality of neural signal recordings to include the new neural signal, modifying the electrode positions corresponding to the plurality of neural signal recordings to include the new electrode position, and executing the instructions for modeling the isolation curve of a neuron and the instructions for iteratively moving the electrode. - View Dependent Claims (30, 31, 32, 33)
-
Specification