U.S. patent application number 15/882213 was filed with the patent office on 2018-06-07 for non-invasive nerve stimulation.
The applicant listed for this patent is NeurostimOAB, Inc.. Invention is credited to Graham Harold Creasey, Michael Bernard Druke, Alan E. Loh, Robert W. Scott, Hoo-min D. Toong, Anthony Wei.
Application Number | 20180154146 15/882213 |
Document ID | / |
Family ID | 62239932 |
Filed Date | 2018-06-07 |
United States Patent
Application |
20180154146 |
Kind Code |
A1 |
Druke; Michael Bernard ; et
al. |
June 7, 2018 |
NON-INVASIVE NERVE STIMULATION
Abstract
One example provides a nerve stimulation treatment using
electrodes coupled to a user. Examples determine a target charge
level and output a series of pulses from the electrodes. For each
pulse outputted, examples measure a charge value of the pulse and
compare the charge value to the target charge level. If the charge
value is greater than the target charge level, examples reduce a
strength level of a subsequent outputted pulse. If the charge value
is less than the target charge level, examples increase the
strength level of a subsequent outputted pulse.
Inventors: |
Druke; Michael Bernard;
(Half Moon Bay, CA) ; Loh; Alan E.; (Los Altos,
CA) ; Scott; Robert W.; (El Granada, CA) ;
Wei; Anthony; (Palo Alto, CA) ; Creasey; Graham
Harold; (Menlo Park, CA) ; Toong; Hoo-min D.;
(Cambridge, MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
NeurostimOAB, Inc. |
Waltham |
MA |
US |
|
|
Family ID: |
62239932 |
Appl. No.: |
15/882213 |
Filed: |
January 29, 2018 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15040856 |
Feb 10, 2016 |
|
|
|
15882213 |
|
|
|
|
14893946 |
Nov 25, 2015 |
|
|
|
PCT/US2014/040240 |
May 30, 2014 |
|
|
|
15040856 |
|
|
|
|
62582634 |
Nov 7, 2017 |
|
|
|
62115607 |
Feb 12, 2015 |
|
|
|
61828981 |
May 30, 2013 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61N 1/3603 20170801;
A61N 1/08 20130101; A61N 1/36014 20130101 |
International
Class: |
A61N 1/36 20060101
A61N001/36; A61N 1/08 20060101 A61N001/08 |
Claims
1. A method of providing a nerve stimulation treatment using
electrodes coupled to a user, the method comprising: determining a
target charge level; outputting a series of pulses from the
electrodes; for each pulse outputted, measuring a charge value of
the pulse and compare the charge value to the target charge level;
if the charge value is greater than the target charge level,
reducing a strength level of a subsequent outputted pulse; and if
the charge value is less than the target charge level, increasing
the strength level of a subsequent outputted pulse.
2. The method of claim 1, in which the series of pulses are defined
based on a frequency and a duration.
3. The method of claim 1, in which determining the target charge
level Q.sub.target comprises generating an acquisition series of
pulses and Q target = i = 1 T * f Q pulse ( i ) , ##EQU00002##
where T is a duration of the acquisition series of pulses, f is a
frequency of the acquisition series of pulses and Q.sub.pulse(i) is
a measured charge of each of the acquisition series of pulses.
4. The method of claim 1, in which the measuring the charge value
of the pulse comprises determining an output of a differential
integrator.
5. The method of claim 1, in which the series of pulses comprises
at least 100 pulses.
6. The method of claim 1, in which an amount of the reducing the
strength and increasing the strength is limited by a predefined
step value.
7. The method of claim 1, in which the determining the target
charge level occurs after a manual adjustment of a voltage output
level.
8. A nerve stimulation device comprising: one or more electrodes;
one or more sensors; a processor coupled to the electrodes and the
sensors, the processor executing instructions to implement nerve
stimulation comprising: determining a target charge level;
outputting a series of pulses from the electrodes; for each pulse
outputted, measuring at the sensors a charge value of the pulse and
compare the charge value to the target charge level; if the charge
value is greater than the target charge level, reducing a strength
level of a subsequent outputted pulse; and if the charge value is
less than the target charge level, increasing the strength level of
a subsequent outputted pulse.
9. The nerve stimulation device of claim 8, in which the series of
pulses are defined based on a frequency and a duration.
10. The nerve stimulation device of claim 8, in which determining
the target charge level Q.sub.target comprises generating an
acquisition series of pulses and Q target = i = 1 T * f Q pulse ( i
) , ##EQU00003## where T is a duration of the acquisition series of
pulses, f is a frequency of the acquisition series of pulses and
Q.sub.pulse(i) is a measured charge of each of the acquisition
series of pulses.
11. The nerve stimulation device of claim 8, in which the measuring
the charge value of the pulse comprises determining an output of a
differential integrator.
12. The nerve stimulation device of claim 8, in which the series of
pulses comprises at least 100 pulses.
13. The nerve stimulation device of claim 8, in which an amount of
the reducing the strength and increasing the strength is limited by
a predefined step value.
14. The nerve stimulation device of claim 8, in which the
determining the target charge level occurs after a manual
adjustment of a voltage output level.
15. A non-transitory computer-readable medium having instructions
stored thereon that, when executed by a processor, cause the
processor to provide a nerve stimulation treatment using electrodes
coupled to a user, the nerve stimulation treatment comprising:
determining a target charge level; outputting a series of pulses
from the electrodes; for each pulse outputted, measuring a charge
value of the pulse and compare the charge value to the target
charge level; if the charge value is greater than the target charge
level, reducing a strength level of a subsequent outputted pulse;
and if the charge value is less than the target charge level,
increasing the strength level of a subsequent outputted pulse.
16. The non-transitory computer-readable medium of claim 15, in
which the series of pulses are defined based on a frequency and a
duration.
17. The non-transitory computer-readable medium of claim 15, in
which determining the target charge level Q.sub.target comprises
generating an acquisition series of pulses and Q target = i = 1 T *
f Q pulse ( i ) , ##EQU00004## where T is a duration of the
acquisition series of pulses, f is a frequency of the acquisition
series of pulses and Q.sub.pulse(i) is a measured charge of each of
the acquisition series of pulses.
18. The non-transitory computer-readable medium of claim 15, in
which the measuring the charge value of the pulse comprises
determining an output of a differential integrator.
19. The non-transitory computer-readable medium of claim 15, in
which the series of pulses comprises at least 100 pulses.
20. The non-transitory computer-readable medium of claim 15, in
which an amount of the reducing the strength and increasing the
strength is limited by a predefined step value.
21. The non-transitory computer-readable medium of claim 15, in
which the determining the target charge level occurs after a manual
adjustment of a voltage output level.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] [0001]This application claims priority of U.S. Provisional
Patent Application Ser. No. 62/582,634, filed on Nov. 7, 2017, and
claims priority as a continuation-in-part application of U.S.
patent application Ser. No. 15/040,856, filed on Feb. 10, 2016,
which claims priority to U.S. Provisional Patent Application Ser.
No. 62/115,607, filed Feb. 12, 2015 and claims priority as a
continuation-in-part application of U.S. patent application Ser.
No. 14/893,946, filed on Nov. 25, 2015, which claims priority to
PCT Patent Application Serial No. PCT/US14/40240, filed May 30,
2014, which claims priority to U.S. Provisional Patent Application
Ser. No. 61/828,981, filed May 30, 2013. The disclosure of each of
these applications is hereby incorporated by reference.
FIELD
[0002] One example is directed generally to a nerve stimulation,
and in particular to nerve stimulation using electrical signals
with computer control.
BACKGROUND INFORMATION
[0003] Mammalian and human nerves control organs and muscles.
Artificially stimulating the nerves elicit desired organ and muscle
responses. Accessing the nerves to selectively control these
responses from outside the body, without invasive implants or
needles penetrating the dermis, muscle or fat tissue, is
desired.
BRIEF DESCRIPTION OF THE DRAWINGS
[0004] FIG. 1 is a depiction of a neuron activating a muscle by
electrical impulse.
[0005] FIG. 2 is a representation of the electrical potential
activation time of an electrical impulse in a nerve.
[0006] FIG. 3 is a graph showing pulses applied to the skin.
[0007] FIG. 4 is a graph showing symmetrical and asymmetrical
pulses applied to the skin.
[0008] FIG. 5 is a cross-sectional diagram showing a field in
underlying tissue produced by application of two electrodes to the
skin.
[0009] FIG. 6 is a cross-sectional diagram showing a field in
underlying tissue produced by application of two electrodes to the
skin, with two layers of tissue of different electrical
resistivity.
[0010] FIG. 7 is a cross-sectional diagram showing a field in
underlying tissue when the stimulating pulse is turned off.
[0011] FIG. 8 shows potential applications of electrical
stimulation to the body.
[0012] FIG. 9A is a system diagram of an example software and
hardware components showing an example of a Topical Nerve
Stimulator/Sensor (TNSS) interpreting a data stream from a control
device in accordance with one example.
[0013] FIG. 9B is a flow chart showing an example of a function of
a master control program in accordance with one example.
[0014] FIG. 10 is a block diagram of an example TNSS component
configuration including a system on a chip (SOC) in accordance with
one example.
[0015] FIG. 11 is a flow diagram of the protocol for adaptive
current control in accordance with one example.
[0016] FIG. 12 is a Differential Integrator Circuit used in the
Adaptive Current Protocol in accordance with one example.
[0017] FIG. 13 is a table relating charge duration vs. frequency to
provide feedback to the Adaptive Current Protocol in accordance
with one example.
DETAILED DESCRIPTION
[0018] One example is a non-invasive nerve stimulator that uses
feedback to adapt the nerve stimulation output. One example is
implemented using a Topical Nerve Stimulator and Sensor ("TNSS")
device such as disclosed in U.S. patent application Ser. No.
14/893,946, hereby incorporated by reference, and that is used to
stimulate nerves. A TNSS may apply electrode-generated electric
field(s) in a low frequency to dermis in the proximity of a nerve,
and typically applies such a field in the magnitude of up to 100s
of hertz ("Hz"). The TNSS also includes hardware and logic for high
frequency ("GHz") communication to mobile devices.
[0019] A wireless system including a TNSS device is disclosed
herein. Its components, features and performance characteristics
are set forth in the following technical description. Advantages of
a wireless TNSS system over existing transcutaneous electrical
nerve stimulation devices include: (1) fine control of all
stimulation parameters from a remote device such as a smartphone,
either directly by the user or by stored programs; (2) multiple
electrodes of a TNSS can form an array to shape an electric field
in the tissues; (3) multiple TNSS devices can form an array to
shape an electric field in the tissues; (4) multiple TNSS devices
can stimulate multiple structures, coordinated by a smartphone; (5)
selective stimulation of nerves and other structures at different
locations and depths in a volume of tissue; (6) mechanical,
acoustic or optical stimulation in addition to electrical
stimulation; (7) the transmitting antenna of TNSS device can focus
a beam of electromagnetic energy within tissues in short bursts to
activate nerves directly without implanted devices; (8) inclusion
of multiple sensors of multiple modalities, including but not
limited to position, orientation, force, distance, acceleration,
pressure, temperature, voltage, light and other electromagnetic
radiation, sound, ions or chemical compounds, making it possible to
sense electrical activities of muscles (EMG, EKG), mechanical
effects of muscle contraction, chemical composition of body fluids,
location or dimensions or shape of an organ or tissue by
transmission and receiving of ultrasound.
[0020] Further advantages of the wireless TNSS system include: (1)
TNSS devices are expected to have service lifetimes of days to
weeks and their disposability places less demand on power sources
and battery requirements; (2) the combination of stimulation with
feedback from artificial or natural sensors for closed loop control
of muscle contraction and force, position or orientation of parts
of the body, pressure within organs, and concentrations of ions and
chemical compounds in the tissues; (3) multiple TNSS devices can
form a network with each other, with remote controllers, with other
devices, with the Internet and with other users; (4) a collection
of large amounts of data from one or many TNSS devices and one or
many users regarding sensing and stimulation, collected and stored
locally or through the internet; (5) analysis of large amounts of
data to detect patterns of sensing and stimulation, apply machine
learning, and improve algorithms and functions; (6) creation of
databases and knowledge bases of value; (7) convenience, including
the absence of wires to become entangled in clothing, showerproof
and sweat proof, low profile, flexible, camouflaged or skin
colored, (8) integrated power, communications, sensing and
stimulating inexpensive disposable TNSS, consumable electronics;
(9) power management that utilizes both hardware and software
functions will be critical to the convenience factor and widespread
deployment of TNSS device.
[0021] Referring to FIG. 1, a nerve cell normally has a voltage
across the cell membrane of 70 millivolts with the interior of the
cell at a negative voltage with respect to the exterior of the
cell. This is known as the resting potential and it is normally
maintained by metabolic reactions which maintain different
concentrations of electrical ions in the inside of the cell
compared to the outside. Ions can be actively "pumped" across the
cell membrane through ion channels in the membrane that are
selective for different types of ion, such as sodium and potassium.
The channels are voltage sensitive and can be opened or closed
depending on the voltage across the membrane. An electric field
produced within the tissues by a stimulator can change the normal
resting voltage across the membrane, either increasing or
decreasing the voltage from its resting voltage.
[0022] Referring to FIG. 2, a decrease in voltage across the cell
membrane to about 55 millivolts opens certain ion channels,
allowing ions to flow through the membrane in a self-catalyzing but
self-limited process which results in a transient decrease of the
trans membrane potential to zero, and even positive, known as
depolarization followed by a rapid restoration of the resting
potential as a result of active pumping of ions across the membrane
to restore the resting situation which is known as repolarization.
This transient change of voltage is known as an action potential
and it typically spreads over the entire surface of the cell. If
the shape of the cell is such that it has a long extension known as
an axon, the action potential spreads along the length of the axon.
Axons that have insulating myelin sheaths propagate action
potentials at much higher speeds than those axons without myelin
sheaths or with damaged myelin sheaths.
[0023] If the action potential reaches a junction, known as a
synapse, with another nerve cell, the transient change in membrane
voltage results in the release of chemicals known as
neuro-transmitters that can initiate an action potential in the
other cell. This provides a means of rapid electrical communication
between cells, analogous to passing a digital pulse from one cell
to another.
[0024] If the action potential reaches a synapse with a muscle cell
it can initiate an action potential that spreads over the surface
of the muscle cell. This voltage change across the membrane of the
muscle cell opens ion channels in the membrane that allow ions such
as sodium, potassium and calcium to flow across the membrane, and
can result in contraction of the muscle cell.
[0025] Increasing the voltage across the membrane of a cell below
-70 millivolts is known as hyper-polarization and reduces the
probability of an action potential being generated in the cell.
This can be useful for reducing nerve activity and thereby reducing
unwanted symptoms such as pain and spasticity
[0026] The voltage across the membrane of a cell can be changed by
creating an electric field in the tissues with a stimulator. It is
important to note that action potentials are created within the
mammalian nervous system by the brain, the sensory nervous system
or other internal means. These action potentials travel along the
body's nerve "highways". The TNSS creates an action potential
through an externally applied electric field from outside the body.
This is very different than how action potentials are naturally
created within the body.
[0027] Electric Fields that can Cause Action Potentials
[0028] Referring to FIG. 2, electric fields capable of causing
action potentials can be generated by electronic stimulators
connected to electrodes that are implanted surgically in close
proximity to the target nerves. To avoid the many issues associated
with implanted devices, it is desirable to generate the required
electric fields by electronic devices located on the surface of the
skin. Such devices typically use square wave pulse trains of the
form shown in FIG. 3. Such devices may be used instead of implants
and/or with implants such as reflectors, conductors, refractors, or
markers for tagging target nerves and the like, so as to shape
electric fields to enhance nerve targeting and/or selectivity.
[0029] Referring to FIG. 3, the amplitude of the pulses "A",
applied to the skin, may vary between 1 and 100 Volts, pulse width
"t", between 100 microseconds and 10 milliseconds, duty cycle (t/T)
between 0.1% and 50%, the frequency of the pulses within a group
between 1 and 100/sec, and the number of pulses per group "n",
between 1 and several hundred. Typically, pulses applied to the
skin will have an amplitude of up to 60 volts, a pulse width of 250
microseconds and a frequency of 20 per second, resulting in a duty
cycle of 0.5%. In some cases balanced-charge biphasic pulses will
be used to avoid net current flow. Referring to FIG. 4, these
pulses may be symmetrical, with the shape of the first part of the
pulse similar to that of the second part of the pulse, or
asymmetrical, in which the second part of the pulse has lower
amplitude and a longer pulse width in order to avoid canceling the
stimulatory effect of the first part of the pulse.
[0030] Formation of Electric Fields by Stimulators
[0031] The location and magnitude of the electric potential applied
to the tissues by electrodes provides a method of shaping the
electrical field. For example, applying two electrodes to the skin,
one at a positive electrical potential with respect to the other,
can produce a field in the underlying tissues such as that shown in
the cross-sectional diagram of FIG. 5.
[0032] The diagram in FIG. 5 assumes homogeneous tissue. The
voltage gradient is highest close to the electrodes and lower at a
distance from the electrodes. Nerves are more likely to be
activated close to the electrodes than at a distance. For a given
voltage gradient, nerves of large diameter are more likely to be
activated than nerves of smaller diameter. Nerves whose long axis
is aligned with the voltage gradient are more likely to be
activated than nerves whose long axis is at right angles to the
voltage gradient.
[0033] Applying similar electrodes to a part of the body in which
there are two layers of tissue of different electrical resistivity,
such as fat and muscle, can produce a field such as that shown in
FIG. 6. Layers of different tissue may act to refract and direct
energy waves and be used for beam aiming and steering. An
individual's tissue parameters may be measured and used to
characterize the appropriate energy stimulation for a selected
nerve.
[0034] Referring to FIG. 7, when the stimulating pulse is turned
off the electric field will collapse and the fields will be absent
as shown. It is the change in electric field that will cause an
action potential to be created in a nerve cell, provided sufficient
voltage and the correct orientation of the electric field occurs.
More complex three-dimensional arrangements of tissues with
different electrical properties can result in more complex
three-dimensional electric fields, particularly since tissues have
different electrical properties and these properties are different
along the length of a tissue and across it, as shown in Table
1.
TABLE-US-00001 TABLE 1 Electrical Conductivity (siemens/m)
Direction Average Blood .65 Bone Along .17 Bone Mixed .095 Fat .05
Muscle Along .127 Muscle Across .45 Muscle Mixed .286 Skin (Dry)
.000125 Skin (Wet) .00121
[0035] Modification of Electric Fields by Tissue
[0036] An important factor in the formation of electric fields used
to create action potentials in nerve cells is the medium through
which the electric fields must penetrate. For the human body this
medium includes various types of tissue including bone, fat,
muscle, and skin. Each of these tissues possesses different
electrical resistivity or conductivity and different capacitance
and these properties are anisotropic. They are not uniform in all
directions within the tissues. For example, an axon has lower
electrical resistivity along its axis than perpendicular to its
axis. The wide range of conductivities is shown in Table 1. The
three-dimensional structure and resistivity of the tissues will
therefore affect the orientation and magnitude of the electric
field at any given point in the body.
[0037] Modification of Electric Fields by Multiple Electrodes
[0038] Applying a larger number of electrodes to the skin can also
produce more complex three-dimensional electrical fields that can
be shaped by the location of the electrodes and the potential
applied to each of them. Referring to FIG. 3, the pulse trains can
differ from one another indicated by A, t/T, n, and f as well as
have different phase relationships between the pulse trains. For
example with an 8.times.8 array of electrodes, combinations of
electrodes can be utilized ranging from simple dipoles, to
quadripoles, to linear arrangements, to approximately circular
configurations, to produce desired electric fields within
tissues.
[0039] Applying multiple electrodes to a part of the body with
complex tissue geometry will thus result in an electric field of a
complex shape. The interaction of electrode arrangement and tissue
geometry can be modeled using Finite Element Modeling, which is a
mathematical method of dividing the tissues into many small
elements in order to calculate the shape of a complex electric
field. This can be used to design an electric field of a desired
shape and orientation to a particular nerve.
[0040] High frequency techniques known for modifying an electric
field, such as the relation between phases of a beam, cancelling
and reinforcing by using phase shifts, may not apply to application
of electric fields by TNSSs because they use low frequencies.
Instead, examples use beam selection to move or shift or shape an
electric field, also described as field steering or field shaping,
by activating different electrodes, such as from an array of
electrodes, to move the field. Selecting different combinations of
electrodes from an array may result in beam or field steering. A
particular combination of electrodes may shape a beam and/or change
the direction of a beam by steering. This may shape the electric
field to reach a target nerve selected for stimulation.
[0041] Activation of Nerves by Electric Fields
[0042] Typically, selectivity in activating nerves has required
electrodes to be implanted surgically on or near nerves. Using
electrodes on the surface of the skin to focus activation
selectively on nerves deep in the tissues, as with examples of the
invention, has many advantages. These include avoidance of surgery,
avoidance of the cost of developing complex implants and gaining
regulatory approval for them, and avoidance of the risks of
long-term implants.
[0043] The features of the electric field that determine whether a
nerve will be activated to produce an action potential can be
modeled mathematically by the "Activating Function" disclosed in
Rattay F., "The basic mechanism for the electrical stimulation of
the nervous system", Neuroscience Vol. 89, No. 2, pp. 335-346
(1999). The electric field can produce a voltage, or extracellular
potential, within the tissues that varies along the length of a
nerve. If the voltage is proportional to distance along the nerve,
the first order spatial derivative will be constant and the second
order spatial derivative will be zero. If the voltage is not
proportional to distance along the nerve, the first order spatial
derivative will not be constant and the second order spatial
derivative will not be zero. The Activating Function is
proportional to the second-order spatial derivative of the
extracellular potential along the nerve. If it is sufficiently
greater than zero at a given point it predicts whether the electric
field will produce an action potential in the nerve at that point.
This prediction may be input to a nerve signature.
[0044] In practice, this means that electric fields that are
varying sufficiently greatly in space or time can produce action
potentials in nerves. These action potentials are also most likely
to be produced where the orientation of the nerves to the fields
change, either because the nerve or the field changes direction.
The direction of the nerve can be determined from anatomical
studies and imaging studies such as MRI scans. The direction of the
field can be determined by the positions and configurations of
electrodes and the voltages applied to them, together with the
electrical properties of the tissues. As a result, it is possible
to activate certain nerves at certain tissue locations selectively
while not activating others.
[0045] To accurately control an organ or muscle, the nerve to be
activated must be accurately selected. This selectivity may be
improved by using examples disclosed herein as a nerve signature,
in several ways, as follows: [0046] (1) Improved algorithms to
control the effects when a nerve is stimulated, for example, by
measuring the resulting electrical or mechanical activity of
muscles and feeding back this information to modify the stimulation
and measuring the effects again. Repeated iterations of this
process can result in optimizing the selectivity of the
stimulation, either by classical closed loop control or by machine
learning techniques such as pattern recognition and artificial
intelligence; [0047] (2) Improving nerve selectivity by labeling or
tagging nerves chemically; for example, introduction of genes into
some nerves to render them responsive to light or other
electromagnetic radiation can result in the ability to activate
these nerves and not others when light or electromagnetic radiation
is applied from outside the body; [0048] (3) Improving nerve
selectivity by the use of electrical conductors to focus an
electric field on a nerve; these conductors might be implanted, but
could be passive and much simpler than the active implantable
medical devices currently used; [0049] (4) The use of reflectors or
refractors, either outside or inside the body, is used to focus a
beam of electromagnetic radiation on a nerve to improve nerve
selectivity. If these reflectors or refractors are implanted, they
may be passive and much simpler than the active implantable medical
devices currently used; [0050] (5) Improving nerve selectivity by
the use of feedback from the person upon whom the stimulation is
being performed; this may be an action taken by the person in
response to a physical indication such as a muscle activation or a
feeling from one or more nerve activations; [0051] (6) Improving
nerve selectivity by the use of feedback from sensors associated
with the TNSS, or separately from other sensors, that monitor
electrical activity associated with the stimulation; and [0052] (7)
Improving nerve selectivity by the combination of feedback from
both the person or sensors and the TNSS that may be used to create
a unique profile of the user's nerve physiology for selected nerve
stimulation.
[0053] Potential applications of electrical stimulation to the body
are shown in FIG. 8.
[0054] Referring to FIG. 9A, a TNSS 934 human and mammalian
interface and its method of operation and supporting system are
managed by a Master Control Program ("MCP") 910 represented in
function format as block diagrams. It provides the logic for the
nerve stimulator system in accordance to one example.
[0055] In one example, MCP 910 and other components shown in FIG.
9A are implemented by one or more processors that are executing
instructions. The processor may be any type of general or specific
purpose processor. Memory is included for storing information and
instructions to be executed by the processor. The memory can be
comprised of any combination of random access memory ("RAM"), read
only memory ("ROM"), static storage such as a magnetic or optical
disk, or any other type of computer readable media.
[0056] Master Control Program
[0057] The primary responsibility of MCP 910 is to coordinate the
activities and communications among the various control programs, a
Data Manager 920, a User 932, and the external ecosystem and to
execute the appropriate response algorithms in each situation. The
MCP 910 accomplishes electric field shaping and/or beam steering by
providing an electrode activation pattern to TNSS device 934 to
selectively stimulate a target nerve. For example, upon
notification by a Communications Controller 930 of an external
event or request, the MCP 910 is responsible for executing the
appropriate response, and working with the Data Manager 920 to
formulate the correct response and actions. It integrates data from
various sources such as Sensors 938 and external inputs such as
TNSS devices 934, and applies the correct security and privacy
policies, such as encryption and HIPAA required protocols. It will
also manage the User Interface (UI) 912 and the various Application
Program Interfaces (APIs) 914 that provide access to external
programs.
[0058] MCP 910 is also responsible for effectively managing power
consumption by TNSS device 934 through a combination of software
algorithms and hardware components that may include, among other
things: computing, communications, and stimulating electronics,
antenna, electrodes, sensors, and power sources in the form of
conventional or printed batteries.
[0059] Communications Controller
[0060] Communications controller 930 is responsible for receiving
inputs from the User 932, from a plurality of TNSS devices 934, and
from 3rd party apps 936 via communications sources such as the
Internet or cellular networks. The format of such inputs will vary
by source and must be received, consolidated, possibly reformatted,
and packaged for the Data Manager 920.
[0061] User inputs may include simple requests for activation of
TNSS devices 934 to status and information concerning the User's
932 situation or needs. TNSS devices 934 will provide signaling
data that may include voltage readings, TNSS 934 status data,
responses to control program inquiries, and other signals.
Communications Controller 930 is also responsible for sending data
and control requests to the plurality of TNSS devices 934. 3rd
party applications 936 can send data, requests, or instructions for
the Master Control Program 910 or User 932 via the Internet or
cellular networks. Communications Controller 930 is also
responsible for communications via the cloud where various software
applications may reside.
[0062] In one example, a user can control one or more TNSS devices
using a remote fob or other type of remote device and a
communication protocol such as Bluetooth. In one example, a mobile
phone is also in communication and functions as a central device
while the fob and TNSS device function as peripheral devices. In
another example, the TNSS device functions as the central device
and the fob is a peripheral device that communicates directly with
the TNSS device (i.e., a mobile phone or other device is not
needed).
[0063] Data Manager
[0064] The Data Manager (DM) 920 has primary responsibility for the
storage and movement of data to and from the Communications
Controller 930, Sensors 938, Actuators 940, and the Master Control
Program 910. The DM 920 has the capability to analyze and correlate
any of the data under its control. It provides logic to select and
activate nerves. Examples of such operations upon the data include:
statistical analysis and trend identification; machine learning
algorithms; signature analysis and pattern recognition,
correlations among the data within the Data Warehouse 926, the
Therapy Library 922, the Tissue Models 924, and the Electrode
Placement Models 928, and other operations. There are several
components to the data that is under its control as disclosed
below.
[0065] The Data Warehouse (DW) 926 is where incoming data is
stored; examples of this data can be real-time measurements from
TNSS devices 934 or from Sensors (938), data streams from the
Internet, or control and instructional data from various sources.
The DM 920 will analyze data, as described above, that is held in
the DW 926 and cause actions, including the export of data, under
MCP 910 control. Certain decision making processes implemented by
the DM 920 will identify data patterns both in time, frequency, and
spatial domains and store them as signatures for reference by other
programs. Techniques such as EMG, or multi-electrode EMG, gather a
large amount of data that is the sum of hundreds to thousands of
individual motor units and the typical procedure is to perform
complex decomposition analysis on the total signal to attempt to
tease out individual motor units and their behavior. The DM 920
will perform big data analysis over the total signal and recognize
patterns that relate to specific actions or even individual nerves
or motor units. This analysis can be performed over data gathered
in time from an individual, or over a population of TNSS Users.
[0066] The Therapy Library 922 contains various control regimens
for the TNSS devices 934. Regimens specify the parameters and
patterns of pulses to be applied by the TNSS devices 934. The width
and amplitude of individual pulses may be specified to stimulate
nerve axons of a particular size selectively without stimulating
nerve axons of other sizes. The frequency of pulses applied may be
specified to modulate some reflexes selectively without modulating
other reflexes. There are preset regimens that may be loaded from
the Cloud 942 or 3rd party apps 936. The regimens may be static
read-only as well as adaptive with read-write capabilities so they
can be modified in real-time responding to control signals or
feedback signals or software updates. Referring to FIG. 3, one such
example of a regimen has parameters A=40 volts, t=500 microseconds,
T=1 Millisecond, n=100 pulses per group, and f=20 per second. Other
examples of regimens will vary the parameters within ranges
previously specified.
[0067] The Tissue Models 924 are specific to the electrical
properties of particular body locations where TNSS devices 934 may
be placed. As previously disclosed, electric fields for production
of action potentials will be affected by the different electrical
properties of the various tissues that they encounter. Tissue
Models 924 are combined with regimens from the Therapy Library 922
and Electrode Placement Models 928 to produce desired actions.
Tissue Models 924 may be developed by MRI, Ultrasound or other
imaging or measurement of tissue of a body or particular part of a
body. This may be accomplished for a particular User 932 and/or
based upon a body norm. One such example of a desired action is the
use of a Tissue Model 924 together with a particular Electrode
Placement Model 928 to determine how to focus the electric field
from electrodes on the surface of the body on a specific deep
location corresponding to the pudendal nerve in order to stimulate
that nerve selectively to reduce incontinence of urine. Other
examples of desired actions may occur when a Tissue Model 924 in
combination with regimens from the Therapy Library 22 and Electrode
Placement Models 928 produce an electric field that stimulates a
sacral nerve. Many other examples of desired actions follow for the
stimulation of other nerves.
[0068] Electrode Placement Models 928 specify electrode
configurations that the TNSS devices 934 may apply and activate in
particular locations of the body. For example, a TNSS device 934
may have multiple electrodes and the Electrode Placement Model 928
specifies where these electrodes should be placed on the body and
which of these electrodes should be active in order to stimulate a
specific structure selectively without stimulating other
structures, or to focus an electric field on a deep structure. An
example of an electrode configuration is a 4 by 4 set of electrodes
within a larger array of multiple electrodes, such as an 8 by 8
array. This 4 by 4 set of electrodes may be specified anywhere
within the larger array such as the upper right corner of the 8 by
8 array. Other examples of electrode configurations may be circular
electrodes that may even include concentric circular electrodes.
The TNSS device 934 may contain a wide range of multiple electrodes
of which the Electrode Placement Models 928 will specify which
subset will be activated. The Electrode Placement Models 928
complement the regimens in the Therapy Library 922 and the Tissue
Models 924 and are used together with these other data components
to control the electric fields and their interactions with nerves,
muscles, tissues and other organs. Other examples may include TNSS
devices 934 having merely one or two electrodes, such as but not
limited to those utilizing a closed circuit.
[0069] Sensor/Actuator Control
[0070] Independent sensors 938 and actuators 940 can be part of the
TNSS system. Its functions can complement the electrical
stimulation and electrical feedback that the TNSS devices 934
provide. An example of such a sensor 938 and actuator 940 include,
but are not limited to, an ultrasonic actuator and an ultrasonic
receiver that can provide real-time image data of nerves, muscles,
bones, and other tissues. Other examples include electrical sensors
that detect signals from stimulated tissues or muscles. The
Sensor/Actuator Control module 950 provides the ability to control
both the actuation and pickup of such signals, all under control of
the MCP 910.
[0071] Application Program Interfaces
[0072] The MCP 910 is also responsible for supervising the various
Application Program Interfaces (APIs) that will be made available
for 3rd party developers. There may exist more than one API 914
depending upon the specific developer audience to be enabled. For
example many statistical focused apps will desire access to the
Data Warehouse 926 and its cumulative store of data recorded from
TNSS 934 and User 932 inputs. Another group of healthcare
professionals may desire access to the Therapy Library 922 and
Tissue Models 924 to construct better regimens for addressing
specific diseases or disabilities. In each case a different
specific API 914 may be appropriate.
[0073] The MCP 910 is responsible for many software functions of
the TNSS system including system maintenance, debugging and
troubleshooting functions, resource and device management, data
preparation, analysis, and communications to external devices or
programs that exist on the smart phone or in the cloud, and other
functions. However, one of its primary functions is to serve as a
global request handler taking inputs from devices handled by the
Communications Controller 930, external requests from the Sensor
Control Actuator Module (950), and 3rd party requests 936. Examples
of High Level Master Control Program (MCP) functions are disclosed
below.
[0074] The manner in which the MCP handles these requests is shown
in FIG. 9B. The Request Handler (RH) 960 accepts inputs from the
User 932, TNSS devices 934, 3rd party apps 936, sensors 938 and
other sources. It determines the type of request and dispatches the
appropriate response as set forth in the following paragraphs.
[0075] User Request: The RH 960 will determine which of the
plurality of User Requests 961 is present such as: activation;
display status, deactivation, or data input, e.g. specific User
condition. Shown in FIG. 9B is the RH's 960 response to an
activation request. As shown in block 962, RH 960 will access the
Therapy Library 922 and cause the appropriate regimen to be sent to
the correct TNSS 934 for execution, as shown at block 964 labeled
"Action."
[0076] TNSS/Sensor Inputs: The RH 960 will perform data analysis
over TNSS 934 or Sensor inputs 965. As shown at block 966, it
employs data analysis, which may include techniques ranging from
DSP decision making processes, image processing algorithms,
statistical analysis and other algorithms to analyze the inputs. In
FIG. 9B two such analysis results are shown; conditions which cause
a User Alarm 970 to be generated and conditions which create an
Adaptive Action 980 such as causing a control feedback loop for
specific TNSS 934 functions, which can iteratively generate further
TNSS 934 or Sensor inputs 965 in a closed feedback loop.
[0077] 3rd Party Apps: Applications can communicate with the MCP
910, both sending and receiving communications. A typical
communication would be to send informational data or commands to a
TNSS 934. The RH 960 will analyze the incoming application data, as
shown at block 972. FIG. 9B shows two such actions that result. One
action, shown at block 974 would be the presentation of the
application data, possibly reformatted, to the User 932 through the
MCP User Interface 912. Another result would be to perform a User
932 permitted action, as shown at 976, such as requesting a regimen
from the Therapy Library 922.
[0078] Referring to FIG. 10, an example TNSS in accordance to one
example is shown. The TNSS has one or more electronic circuits or
chips 1000 that perform the functions of: communications with the
controller, nerve stimulation via one or more electrodes 1008 that
produce a wide range of electric field(s) according to treatment
regimen, one or more antennae 1010 that may also serve as
electrodes and communication pathways, and a wide range of sensors
1006 such as, but not limited to, mechanical motion and pressure,
temperature, humidity, chemical and positioning sensors. In another
example, TNSS interfaces to transducers 1014 to transmit signals to
the tissue or to receive signals from the tissue.
[0079] One arrangement is to integrate a wide variety of these
functions into an SOC, system on chip 1000. Within this is shown a
control unit 1002 for data processing, communications, transducer
interface and storage and one or more stimulators 1004 and sensors
1006 that are connected to electrodes 1008. An antenna 1010 is
incorporated for external communications by the control unit. Also
present is an internal power supply 1012, which may be, for
example, a battery. An external power supply is another variation
of the chip configuration. It may be necessary to include more than
one chip to accommodate a wide range of voltages for data
processing and stimulation. Electronic circuits and chips will
communicate with each other via conductive tracks within the device
capable of transferring data and/or power.
[0080] The TNSS interprets a data stream from the control device,
such as that shown in FIG. 9A, to separate out message headers and
delimiters from control instructions. In one example, control
instructions contain information such as voltage level and pulse
pattern. The TNSS activates the stimulator 1004 to generate a
stimulation signal to the electrodes 1008 placed on the tissue
according to the control instructions. In another example the TNSS
activates a transducer 1014 to send a signal to the tissue. In
another example, control instructions cause information such as
voltage level and pulse pattern to be retrieved from a library
stored in the TNSS.
[0081] The TNSS receives sensory signals from the tissue and
translates them to a data stream that is recognized by the control
device, such as the example in FIG. 9A. Sensory signals include
electrical, mechanical, acoustic, optical and chemical signals
among others. Sensory signals come to the TNSS through the
electrodes 1008 or from other inputs originating from mechanical,
acoustic, optical, or chemical transducers. For example, an
electrical signal from the tissue is introduced to the TNSS through
the electrodes 1008, is converted from an analog signal to a
digital signal and then inserted into a data stream that is sent
through the antenna 1010 to the control device. In another example
an acoustic signal is received by a transducer 1014 in the TNSS,
converted from an analog signal to a digital signal and then
inserted into a data stream that is sent through the antenna 1010
to the control device. In certain examples sensory signals from the
tissue are directly interfaced to the control device for
processing.
[0082] An open loop protocol to control current to electrodes in
known neural stimulation devices does not have feedback controls.
It commands a voltage to be set, but does not check the actual
Voltage. Voltage control is a safety feature. A stimulation pulse
is sent based on preset parameters and cannot be modified based on
feedback from the patient's anatomy. When the device is removed and
repositioned, the electrode placement varies. Also the humidity and
temperature of the anatomy changes throughout the day. All these
factors affect the actual charge delivery if the voltage is
preset.
[0083] In contrast, examples of the TNSS stimulation device have
features that address these shortcomings using the Nordic
Semiconductor nRF52832 microcontroller to regulate charge in a
TNSS. The High Voltage Supply is implemented using a LED driver
chip combined with a Computer controlled Digital Potentiometer to
produce a variable voltage. A 3-1 step up Transformer then provides
the desired High Voltage, "VBOOST", which is sampled to assure that
no failure causes an incorrect Voltage level as follows. The
nRF52832 Microcontroller samples the voltage of the stimulation
waveform providing feedback and impedance calculations for an
adaptive protocol to modify the waveform in real time. The Current
delivered to the anatomy by the stimulation waveform is integrated
using a differential integrator and sampled and then summed to
determine actual charge delivered to the user for a Treatment.
After every pulse in a Stimulation event, this measurement is
analyzed and used to modify, in real time, subsequent pulses.
[0084] This hardware adaptation allows a firmware protocol to
implement the adaptive protocol. This protocol regulates the charge
applied to the body by changing VBOOST. A treatment is performed by
a sequence of periodic pulses, which insert charge into the body
through the electrodes. Some of the parameters of the treatment are
fixed and some are user adjustable. The strength, duration and
frequency may be user adjustable. The user may adjust these
parameters as necessary for comfort and efficacy. The strength may
be lowered if there is discomfort and raised if nothing is felt.
The duration will be increased if the maximum acceptable strength
results in an ineffective treatment.
[0085] A flow diagram in accordance with one example of the
Adaptive Protocol disclosed above is shown in FIG. 11. The Adaptive
Protocol strives to repeatedly and reliably deliver a target charge
("Q.sub.target") during a treatment and to account for any
environmental changes. Therefore, the functionality of FIG. 11 is
to adjust the charge level applied to a user based on feedback,
rather than use a constant level.
[0086] The mathematical expression of this protocol is as follows:
Q.sub.target=Q.sub.target(A*dS+B*dT), where A is the Strength
Coefficient--determined empirically, dS is the user change in
Strength, B is the Duration Coefficient--determined empirically,
and dT is the user change in Duration.
[0087] The Adaptive Protocol includes two phases in one example:
Acquisition 1100 and Reproduction 1120. Any change in user
parameters places the Adaptive Protocol in the Acquisition phase.
When the first treatment is started, a new baseline charge is
computed based on the new parameters. At a new acquisition phase at
1102, all data from the previous charge application is discarded.
In one example, 1102 indicates the first time for the current usage
where the user places the TNSS device on a portion of the body and
manually adjusts the charge level, which is a series of charge
pulses, until it feels suitable, or any time the charge level is
changed, either manually or automatically. The treatment then
starts. The mathematical expression of this function of the
application of a charge is as follows: The charge delivered in a
treatment is
Q target = i = 1 T * f Q pulse ( i ) ##EQU00001##
Where T is the duration; f is the frequency of "Rep Rate";
Q.sub.pulse(i) is the measured charge delivered by Pulse (i) in the
treatment pulse train provided as a voltage MON_CURRENT that is the
result of a Differential Integrator circuit shown in FIG. 12 (i.e.,
the average amount of charge per pulse). The Nordic microcontroller
of FIG. 12 is an example of an Analog to Digital Conversion feature
used to quantify voltage into a number representing the delivered
charge and therefore determine the charge output. The number of
pulses in the treatment is T*f.
[0088] At 1104 and 1106, every pulse is sampled. In one example,
the functionality of 1104 and 1106 lasts for 10 seconds with a
pulse rate of 20 Hz, which can be considered a full treatment
cycle. The result of phase 1100 is the target pulse charge of
Q.sub.target.
[0089] FIG. 13 is a table in accordance with one example showing
the number of pulses per treatment measured against two parameters,
frequency and duration. Frequency is shown on the Y-axis and
duration on the X-axis. The Adaptive Current protocol in general
performs better when using more pulses. One example uses a minimum
of 100 pulses to provide for solid convergence of charge data
feedback. Referring to the FIG. 13, a frequency setting of 20 Hz
and duration of 10 seconds produces 200 pulses, which is desirable
to allow the Adaptive Current Protocol to reproduce a previous
charge.
[0090] The reproduction phase 1120 begins in one example when the
user initiates another subsequent treatment after acquisition phase
1100 and the resulting acquisition of the baseline charge,
Q.sub.target. For example, a full treatment cycle, as discussed
above, may take 10 seconds. After, for example, a two-hour pause as
shown at wait period 1122, the user may then initiate another
treatment. During this phase, the Adaptive Current Protocol
attempts to deliver Q.sub.target for each subsequent treatment. The
functionality of phase 1120 is needed because, during the wait
period 1122, conditions such as the impedance of the user's body
due to sweat or air humidity may have changed. The differential
integrator is sampled at the end of each Pulse in the Treatment. At
that point, the next treatment is started and the differential
integrator is sampled for each pulse at 1124 for purposes of
comparison to the acquisition phase Q.sub.target. Sampling the
pulse includes measuring the output of the pulse in coulombs.
[0091] The output of the integrator of FIG. 12 in voltage, referred
to as Mon_Current 1201, is a direct linear relationship to the
delivered charge in micro-coulombs and provides a reading of how
much charge is leaving the device and entering the user. At 1126,
each single pulse is compared to the charge value determined in
phase 1100 (i.e., the target charge) and the next pulse will be
adjusted in the direction of the difference.
NUM_PULSES=(T*f)
After each pulse, the observed charge, Q.sub.pulse(i), is compared
to the expected charge per pulse.
Q.sub.pulse(i)>Q.sub.target/NUM_PULSES?
The output charge or "VBOOST" is then modified at either 1128
(decreasing) or 1130 (increasing) for the subsequent pulse by:
dV(i)=G[Q.sub.target/NUM_PULSES-Q.sub.pulse(i)]
where G is the Voltage adjustment Coefficient--determined
empirically. The process continues until the last pulse at
1132.
[0092] A safety feature assures that the VBOOST will never be
adjusted higher by more than 10%. If more charge is necessary, then
the repetition rate or duration can be increased.
[0093] In one example, in general, the current is sampled for every
pulse during acquisition phase 1100 to establish target charge for
reproduction. The voltage is then adjusted via a digital
potentiometer, herein referred to as "Pot", during reproduction
phase 1120 to achieve the established target_charge.
[0094] The digital Pot is calibrated with the actual voltage at
startup. A table is generated with sampled voltage for each wiper
value. Tables are also precomputed storing the Pot wiper increment
needed for 1v and 5v output delta at each pot level. This enables
quick reference for voltage adjustments during the reproduction
phase. The tables may need periodic recalibration due to battery
level.
[0095] In one example, during acquisition phase 1100, the minimum
data set=100 pulses and every pulse is sampled and the average is
used as the target_charge for reproduction phase 1120. In general,
less than 100 pulses may provide an insufficient data sample to use
as a basis for reproduction phase 1120. In one example, the default
treatment is 200 pulses (i.e., 20 Hz for 10 seconds). In one
example, a user can adjust both duration and frequency
manually.
[0096] In one example, during acquisition phase 1100, the maximum
data set=1000 pulses. The maximum is used to avoid overflow of 32
bit integers in accumulating the sum of samples. Further, 1000
pulses in one example is a sufficiently large data set and
collecting more is likely unnecessary.
[0097] After 1000 pulses for the above example, the target_charge
is computed. Additional pulses beyond 1000 in the acquisition phase
do not contribute to the computation of the target charge.
[0098] In one example, the first 3-4 pulses are generally higher
than the rest so these are not used in acquisition phase 1100. This
is also accounted for in reproduction phase 1120. Using these too
high values can result in target charge being set too high and over
stimulating on the subsequent treatments in reproduction phase
1120. In other examples, more advanced averaging algorithms could
be applied to eliminating high and low values.
[0099] In an example, there may be a safety concern about
automatically increasing the voltage. For example, if there is poor
connection between the device and the user's skin, the voltage may
auto-adjust at 1130 up to the max. The impedance may then be
reduced, for example by the user pressing the device firmly, which
may result in a sudden high current. Therefore, in one example, if
the sample is 500 mv or more higher than the target, it immediately
adjusts to the minimum voltage. This example then remains in
reproduction phase 1120 and should adjust back to the target
current/charge level. In another example, the maximum voltage
increase is set for a single treatment (e.g., 10V). More than that
should not be needed in normal situations to achieve the
established target_charge. In another example, a max is set for
VBOOST (e.g., 80V).
[0100] In various examples, it is desired to have stability during
reproduction phase 1120. In one example, this is accomplished by
adjusting the voltage by steps. However, a relatively large step
adjustment can result in oscillation or over stimulation.
Therefore, voltage adjustments may be made in smaller steps. The
step size may be based on both the delta between the target and
sample current as well as on the actual VBOOST voltage level. This
facilitates a quick and stable/smooth convergence to the target
charge and uses a more gradual adjustments at lower voltages for
more sensitive users.
[0101] The following are the conditions that may be evaluated to
determine the adjustment step. [0102]
delta-mon_current=abs(sample_mon_current-target_charge) [0103] If
delta_mon_current>500 mv and VBOOST>20V then step=5V for
increase adjustments [0104] (For decrease adjustments a 500 mv
delta triggers emergency decrease to minimum Voltage) [0105] If
delta_mon_current>200 mv then step=1V [0106] If
delta_mon_current>100 mv and
delta_mon_current>5%*sample_mon_current then step=1V
[0107] In other examples, new treatments are started with voltage
lower than target voltage with a voltage buffer of approximately
10%. The impedance is unknown at the treatment start. These
examples save the target_voltage in use at the end of a treatment.
If the user has not adjusted the strength parameter manually, it
starts a new treatment with saved target_voltage with the 10%
buffer. This achieves target current quickly with the 10% buffer to
avoid possible over stimulation in case impedance has been reduced.
This also compensates for the first 3-4 pulses that are generally
higher.
[0108] As disclosed, examples apply an initial charge level, and
then automatically adjust based on feedback of the amount of
current being applied. The charge amount can be varied up or down
while being applied. Therefore, rather than setting and then
applying a fixed voltage level throughout a treatment cycle,
implementations of the invention measure the amount of charge that
is being input to the user, and adjust accordingly throughout the
treatment to maintain a target charge level that is suitable for
the current environment.
[0109] Several examples are specifically illustrated and/or
described herein. However, it will be appreciated that
modifications and variations of the disclosed examples are covered
by the above teachings and within the purview of the appended
claims without departing from the spirit and intended scope of the
invention.
* * * * *